Volume 71, Number 5, September-October 2019

Siriraj Medical Journal

SMJ

ISSN 2629-995X

ORIGINAL ARTICLE

324

The Effectiveness of Computer-Assisted Instruction for Caring

Nasogastric Tube: a Randomized Controlled Study

Chuthamat Yuchareon, et al.

331 DIR/Floortime® Parent Training Intervention for Children

with Developmental Disabilities: a Randomized Controlled

Trial

Kingkaew Pajareya, et al.

339

Predictive Factors for Survival Outcomes of High-Risk

Febrile Neutropenic Patients: a 3-Year Study at a Single

Center in Thailand

Wannaphorn Rotchanapanya, et al.

349 The Components of Strategic Leadership of Prosthetic

and Orthotic Practitioners in Thailand

Thatchanan Manopetkasem, et al.

356 Correlation between Heart Failure with N-Terminal-Pro-B

Type Natriuretic Peptide and Cardiothoracic Ratio from

Chest Radiograph in Pediatric Heart Disease

Worawan Jittham

364 Ocular Manifestations in Acute Herpes Zoster

Ophthalmicus

Ranida Thamphithak, et al.

370 Association Between Body Mass Index and Moderate-

to-Severe Vasomotor Symptoms in Thai Postmenopausal

Women

Kitirat Techatraisak, et al.

377 Factors Associated with Xerostomia in Non-Radiated

Patients

Vannipa Vathanophas, et al.

385 Correlation Between CT Findings of Invasive Pulmonary

Aspergillosis and Severity of Neutropenia

Suwimon Wonglaksanapimon, et al.

392 Success Rate of Cervical Cerclage at Siriraj Hospital

Phanitra Maneeratprasert, et al.

399 The Vertical Ground Reaction Force and Temporal-Spatial

Parameters of Transfemoral Amputees Wearing Three

Prosthetic Knee Joints Available in Thailand: a Pilot Study

Thanyaporn Rakbangboon, et al.

405 Utility of the Siriraj Psoriatic Arthritis Screening Tool, the

Thai Psoriasis Epidemiology Screening Tool, and the Early

Arthritis for Psoriatic Patients Questionnaire to Screen for

Psoriatic Arthritis in an Outpatient Dermatology Clinic

Setting, and Identification of Factors Significantly

Associated with Psoriatic Arthritis

Leena Chularojanamontri, et al.

REVIEW ARTICLE

414 The Development of Personalized Medicine:

Acute Myeloid Leukemia as a Model

Ployploen Phikulsod, et al.

CASE REPORT

426 Hybrid Central Odontogenic Fibroma with Central Giant

Cell Granuloma Like Lesion ; A Case Report and Review of

the Literature

Sakanus Vijintanawan, et al.

www.smj.si.mahidol.ac.th

E-mail: sijournal@mahidol.ac.th

International Association of Surgeons

Gastroenterologists & Oncologists

Thailand Chapter

Thai Association for Gastrointestinal

Endoscopy

Indexed by

By Ployploen Phikulsod, et al.

SIRIRAJ MEDICAL JOURNAL

First Editor: Ouay Ketusinh Emeritus Editors: Somchai Bovornkitti, Adulya Viriyavejakul, Sommai Toongsuwan,

Nanta Maranetra, Niphon Poungvarin, Prasit Watanapa, Vithya Vathanophas, Pipop Jirapinyo, Sanya Sukpanichnant,

Somboon Kunathikom

Executive Editor: Prasit Watanapa Editorial Director: Manee Rattanachaiyanont

Managing Editor: Gulapar Srisawasdi, Chenchit Chayachinda

Editor-in-Chief: awatchai Akaraviputh

Associate Editor: Varut Lohsiriwat, Prapat Wanitpongpan Online Editor: Puttinun Patpituck

SIRIRAJ MEDICAL JOURNAL is published bimonthly, 6 issues a year (Jan-Feb, Mar-Apr, May-Jun, Jul-Aug, Sep-Oct and Nov-Dec)

and distributed by the end of the last month of that issue.

SIRIRAJ MEDICAL JOURNAL is listed as a journal following the Uniform Requirements for Manuscripts Submitted to Biomedical Journals (URM)

by the International Committee of Medical Journal Editors (ICMJE) since 9 July 2010 [http://www.icmje.org/journals.html].

Philip Board (Australian National University, Australia)

Richard J. Deckelbaum (Columbia University, USA)

Yozo Miyake (Aichi Medical University, Japan)

Yik Ying Teo (National University of Singapore, Singapore)

Harland Winter (Massachusetts General Hospital, USA)

Philip A. Brunell (State University of New York At Bualo, USA)

Noritaka Isogai (Kinki University, Japan)

Yuji Murata (Aizenbashi Hospital, Japan)

Keiichi Akita (Tokyo Medical and Dental University Hospital, Japan)

Shuji Shimizu (Kyushu University Hospital, Japan)

David S. Sheps (University of Florida, USA)

Robin CN Williamson (Royal Postgraduate Medical School, UK)

Tai-Soon Yong (Yonsei University, Korea)

Anusak Yiengpruksawan (e Valley Robotic Institute, USA)

Stanlay James Rogers (University of California, San Francisco, USA)

Kyoichi Takaori (Kyoto University Hospital, Japan)

Tomohisa Uchida (Oita University, Japan)

Yoshiki Hirooka (Nagoya University Hospital, Japan)

Hidemi Goto (Nagoya University Graduate School of Medicine, Japan)

Kazuo Hara (Aichi Cancer Center Hospital, Japan)

Shomei Ryozawa (Saitama Medical University, Japan)

Christopher Khor (Singapore General Hospital, Singapore)

Yasushi Sano (Director of Gastrointestinal Center, Japan)

Mitsuhiro Kida (Kitasato University & Hospital, Japan)

Seigo Kitano (Oita University, Japan)

Ichizo Nishino (National Institute of Neuroscience NCNP, Japan)

Masakazu Yamamoto (Tokyo Women’s Medical University, Japan)

Dong-Wan Seo (University of Ulsan College of Medicine, Korea)

George S. Baillie (University of Glasgow, UK)

G. Allen Finley (Delhousie University, Canada)

Sara Schwanke Khilji (Oregon Health & Science University, USA)

Matthew S. Dunne (Institute of Food, Nutrition, and Health, Switzerland)    

Marianne Hokland (University of Aarhus, Denmark)

Marcela Hermoso Ramello (University of Chile, Chile)

Ciro Isidoro (University of Novara, Italy)

Moses Rodriguez (Mayo Clinic, USA)

Robert W. Mann (University of Hawaii, USA)

Wikrom Karnsakul (Johns Hopkins Children’s Center, USA)

Frans Laurens Moll (University Medical Center Ultrecht, Netherlands)

James P. Dolan (Oregon Health & Science University, USA)

John Hunter (Oregon Health & Science University, USA)

Nima Rezaei (Tehran University of Medical Sciences, Iran)

Dennis J. Janisse (Subsidiary of DJO Global, USA)

Folker Meyer (Argonne National Laboratory, USA)

David Wayne Ussery (University of Arkansas for Medical Sciences, USA)

Intawat Nookaew (University of Arkansas for Medical Sciences, USA)

Victor Manuel Charoenrook de la Fuente 

(Centro de Oalmologia Barraquer, Spain)

Karl omas Moritz

(Swedish University of Agricultural Sciences, Sweden)

Nam H. CHO (University School of Medicine and Hospital, Korea)

www.smj.si.mahidol.ac.th

SMJ

Statistician: Saowalak Hunnangkul (Mahidol University, ailand)

Medical Illustrator: Chananya Hokierti (Nopparat Rajathanee Hospital, ailand)

Online Assistant: Surang Promsorn, Wilailuck Amornmontien, Hatairat Ruangsuwan Editorial Oce Secretary: Amornrat Sangkaew

International Editorial Board

Editorial Board

Watchara Kasinrerk (Chiang Mai University, ailand)

Rungroj Krittayaphong (Siriraj Hospital, Mahidol University, ailand)

Wiroon Laupattrakasem (Khon Kaen University, ailand)

Anuwat Pongkunakorn (Lampang Hospital, ailand)

Nopporn Sittisombut (Chiang Mai University, ailand)

Vasant Sumethkul (Ramathibodi Hospital, Mahidol University, ailand)

Yuen Tanniradorm (Chulalongkorn University, ailand)

Saranatra Waikakul (Siriraj Hospital, Mahidol University, ailand)

Pa-thai Yenchitsomanus (Siriraj Hospital, Mahidol University, ailand)

Surapol Issaragrisil (Siriraj Hospital, Mahidol University,ailand)

Jaturat Kanpittaya (Khon Kaen University, ailand)

Suneerat Kongsayreepong (Siriraj Hospital, Mahidol University, ailand)

Pornchai O-Charoenrat (Siriraj Hospital, Mahidol University, ailand)

Nopphol Pausawasdi (Siriraj Hospital, Mahidol University, ailand)

Supakorn Rojananin (Siriraj Hospital, Mahidol University, ailand)

Jarupim Soongswang (Siriraj Hospital, Mahidol University, ailand)

Suttipong Wacharasindhu (Chulalongkorn University, ailand)

Prapon Wilairat (Mahidol University, ailand)

Pornprom Muangman (Siriraj Hospital, Mahidol University, ailand)

Ampaiwan Chuansumrit

(Ramathibodi Hospital, Mahidol University, ailand)

Sayomporn Sirinavin

(Ramathibodi Hospital, Mahidol University, ailand)

Vitoon Chinswangwatanakul

(Siriraj Hospital, Mahidol University, ailand)

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

324

Original Article

SMJ

Chuthamat Yuchareon, B.Tech.*, Pranom Phromdaeng, M.NS.**, Aree Booranakul, M.NS.**, Pratuang Naksit,

BS.*

*Department of Education, **e School of Nurse Assistant Siriraj Hospital, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700,

ailand.

The Effectiveness of Computer-Assisted

Instruction for Caring Nasogastric Tube:

a Randomized Controlled Study

ABSTRACT

Objective: To compare the eectiveness of Computer-Assisted Instruction (CAI) with traditional instruction for

the training of nurse assistant students in the care of patients with nasogastric tube (NG tube).

Methods: Nurse assistant students were randomized to either the CAI group or the control group. Student academic

performances, including both the theoretical and practical parts were observed.

Results: 170 participants were enrolled in this study and were equally randomized to CAI group (n=85) and the

control group (n=85). 24.7% of CAI group (24/85) and 14.1% of control group (12/85) passed both examinations

(Dierence 10.6% (95% CI: -1.37% to 22.57%), p = 0.081). Interestingly, 54.1% of the CAI group whereas only 35.3%

of the control group passed the theoretical part examination (Dierence 18.8% (95% CI: 3.8% to 33.0%, p = 0.041).

For practical part, 64.7% of the CAI group and 58.8% of the control group passed the examination (Dierence 5.9%

(95% CI: -8.73% to 20.26%, p = 0.430).

Conclusion: For nurse assistant students, CAI for caring the patients with NG tube did not demonstrate benet

for the whole examination. However, CAI seemed to oer some benet for teaching especially the theoretical part.

Keywords: A randomized controlled trial; computer-assisted instruction; nurse assistant students; patients with

nasogastric tube (Siriraj Med J 2019; 71: 324-330)

Corresponding author: Chuthamat Yuchareon

E-mail: chuthamat.yuj@mahidol.edu

Received 11 September 2017 Revised 7 March 2018 Accepted 14 March 2018

ORCID ID: http://orcid.org/0000-0002-9494-6715

http://dx.doi.org/10.33192/Smj.2019.50

INTRODUCTION

Students enrolled at the School of Nurse Assistant

Siriraj Hospital endeavor to develop their ability to become

productive, contributing, and essential members of the total

healthcare experience team at Siriraj Hospital. Creativity,

critical thinking, communication, and collaboration

are essential components of an eective educational

program.

eoretical and practical medical education

at all levels is necessary for providing eective care and

treatment and to assist in preventing healthcare-related

errors and omissions. Care of patients with nasogastric

tube (NG tube) is a key responsibility of nurse assistants.

Instruction given to nurse assistants in NG tube-related

patient care protocols is generally delivered via textbooks

and lectures in a traditional classroom setting. ough

NG tube-related care involves gastric intubation, enteral

tube feeding, NG tube cleaning, and NG tube removal,

nurse assistant students receive only one combined

theoretical and practical course in NG tube-related

care. is abbreviated training in NG tube-related care

may lead to complications in NG tube management,

unprofessional techniques, and negative physical and

psychological outcomes for patients. Accordingly, the

inclusion of computer-assisted instruction (CAI) as

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

325

Yuchareon et al.

a method for training nurse assistants in the care of

patients with NG tube may yield signicant benets for

both our patients and our center.

e ongoing development of technology continues

to provide additional and oen improved methods for

supporting instructors. Dedicated computer-assisted

instruction (CAI) modules have been created to help

teach nurse assistants the theories and methods used

in the care of patients with NG tube. Specically, CAI

involves learning via computer using specically designed

soware. Several educational computer programs are

available online, from computer stores, and from textbook

publishing companies. ese instructional soware packages

oer several instructional enhancements, including well-

produced animation, sound, and visualized and detailed

demonstration. Using these programs, students are able

to progress at their own pace and work individually.

Aer receiving instruction, students can form groups

and engage in the solving of hypothetical clinical NG

tube-related problems. Another benet of CAI is that

these programs oen provide immediate feedback, letting

students know whether their answer is correct or not.

CAI

also improves instruction for students with disabilities,

making access to instruction easier and more convenient.

CAI has the ability to capture student attention for the

following reasons: instructional soware programs are

interactive and interesting, programs are designed to

engage the student’s spirit of competitiveness to increase

scores, and computers are fundamental to a student’s

digital world experience. Another advantage of CAI is

that it moves at the student’s pace and normally does not

move ahead until the student has mastered the skill being

taught. Programs also provide dierentiated lessons to

challenge students who are at-risk, average, or advanced.

2-3

Demonstrated advantages of active learning, reective

learning, independent learning, and task-based teaching

were all observed in students who learned using CAI.

Computer-assisted instruction (CAI) was investigated

in this study to determine if it is a better instruction

method than traditional class-based instruction for

nurse assistant training in the care of patients with NG

tube at the School of Nurse Assistant Siriraj Hospital.

MATERIALS AND METHODS

is experimental study was conducted among

students enrolled at the School of Nurse Assistant Siriraj

Hospital, a subsidiary unit of the Faculty of Medicine

Siriraj Hospital, Mahidol University, Bangkok, ailand.

Inclusion criteria were nurse assistant students who were

enrolled in SIPN 043-NG Tube Healthcare course and who

were older than 18 years of age. Students were excluded

if they were for some reason unable to participate in or

follow the protocol of the study. e study design was a

randomized controlled trial. One hundred and seventy

nurse assistant students were enrolled in this study, of

which 85 were randomly assigned to the computer-

assisted instruction group (CAI group) and 85 were

allocated to the PowerPoint instruction group (control

group). Study group randomization was determined by

computerized randomization. Student age, GPA, and

O-net (Ordinary National Educational Test) scores were

not dierent between groups (Table 1).

TABLE 1. Demographic and academic performance data of 170 nurse assistant students.

Sample

Data CAI Group Control Group

(n=85) (n=85)

Gender

Male 7 (8.2%) 3 (3.5%)

Female 78 (91.8%) 82 (96.5%)

Age (yrs) 19.2±1.3 19.3±1.3

GPA 3.1±0.4 3.1±0.4

O-Net score 179.8±16.8 179.6±15.5

Anatomy and physiology score 41.5±7.5 41.8±7.2

Data are presented as n (%) or mean ± standard deviation

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

326

Original Article

SMJ

e hypothesis of our study comprised a primary and

a secondary outcome. Instructor’s teaching and outcome

experience found that 60% of students passed both the

theory and practical skills course components with a

score >70%. As such, the primary outcome hypothesis

was that >80% of the students in the CAI group would

pass both components of the exam with a score of 20 for

theory and 11 for practical skills or higher. According

to a calculation by nQuery Advisor soware (Statistical

Solutions Ltd., Cork, Ireland), a p-value = 0.05 was

calculated using a 2-sided test (Z

(0.05/2)

=1.96) and a

power = 80% (Z

1-β

0.8

=0.84). e secondary outcome

hypothesis was that >70% of students in the CAI group

would report satisfaction with computer-based learning

in order to compare students behavior during studying of

media learning between CAI and control groups such as

motivation, interest in lesson and cooperative learning.

Data were collected from 6 October 2014 to 9 October

2014. e student instruction and data collection process

for each day of the training was conducted, as follows:

Day 1: All students in both study groups attended

traditional lecture-based instruction in NG tube-related

patient care protocols in a traditional classroom setting.

Day 2: Students were randomly separated into the

2 study groups for lecture review. e control group was

provided with a textbook and PowerPoint instruction

and the CAI group was provided with computer-assisted

instruction (CAI). e duration of the review and study

period was 3 hours, with student study behavior observed

and recorded by School of Nurse Assistant Siriraj Hospital’s

sta.

Day 3: A 25 question multiple-choice examination was

given to students in both groups to test their knowledge.

e same exam was given to both groups.

Day 4: A practical examination consisting of 16

operations on a medical model was given to students in

both study groups. Enteral feeding exam was also given

to both groups on day 4.

Data were analyzed using SPSS Statistics version

18 (SPSS, Inc., Chicago, IL, USA). Descriptive statistics

are presented as mean (X) ± standard deviation (SD).

Student examination scores were analyzed by chi-square

test, with results presented as odds ratio (OR) and 95%

condence interval (CI). A p-value<0.05 was considered

statistically signicant (Tables 2-3).

e protocol for this study was approved by the

Siriraj Institutional Review Board (SIRB) of the Faculty of

Medicine Siriraj Hospital (Si 429/2015). CAI instructional

content was reviewed and approved by 3 experts by

report evaluation and instructional content validity was

veried by the course instructor.

TABLE 2. Knowledge and practice skill scores for lesson on patient with NG tube health care (n=170).

TABLE 3. Pass rates by type of assessment for CAI and control groups.

Sample Mean difference P-value

Score CAI Group (n=85) Control Group (n=85) (95% CI)

(mean±SD) (mean±SD)

Knowledge score 18.1±2.4 16±3.0 2.1 (1.3 - 2.9) <0.001

Practice skill score 20.9±4.6 20.6±5.0 0.3 (-1.2 - 1.7) 0.735

A p-value<0.05 indicates statistical signicance

Type of assessment CAI Group Control Group Difference (95% CI) P-value

(n=85) (n=85)

n (%) n (%)

Pass theory 46 (54.1) 30 (35.3) 18.8% (3.8% - 33.0%) .041

Pass practice skill 55 (64.7) 50 (58.8) 5.9% (-8.73% - 20.26%) .430

Pass theory and practice skill 21 (24.7) 12 (14.1) 10.6% (-1.37% - 22.57%) .081

A p-value<0.05 indicates statistical signicance

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

327

RESULTS

A total of 170 students were enrolled, with students

randomly allocated into one of two study groups [CAI

group (n=85) or control group (n=85)]. ere were no

exclusions and all enrolled participants were analyzed

(Fig 1). Objective testing results for subject theory were

statistically signicantly higher in the CAI group than

in the control group (18.1 vs. 16) (p<0.001, T-test) with

mean dierence equal to 2.1 (95% CI; 1.3-2.9). Scores

for the practical examination were only slightly higher

in the CAI group than in the control group, with the

dierence failing to achieve statistical signicance (20.9

vs. 20.6) (p=0.735, T-test) (Table 2).

e procedure for enteral feeding of patients with

NG tube was important for students to carefully follow

- particularly 2 key points. e students were required

to position the patient with the head at a 30-45 degree

angle and drain gastric content for purposes of testing

the position of the feeding tube. A failure to perform

these two key steps resulted in a failing grade for the

practical component of the test. Regarding the 25 question

multiple-choice examination, the CAI group performed

signicantly better than the control group (46 vs. 30)

(p=0.041, Chi-square). e level of passing scores for

the practical examination were not signicantly dierent

between groups (55 vs. 50) (p=0.430, Chi-square). Pass

rates by type of assessment for CAI and control groups are

presented in Table 3. Regarding the number of students

that passed both the theoretical and practical portions

of the exam, the CAI group exceeded the control group

by 10% (21 vs. 12) (p=0.081, Chi-square).

Student levels of satisfaction regarding dierent

aspects of computer-assisted instruction (CAI) are shown

in Table 4. Mean level of student satisfaction with CAI

was 4.58 out of 5. Nine students (10.6%) commented that

CAI was “modern”, “easy to understand”, “colorful”, and

“exciting to learn”. Regarding other feedback results, 7.65%

of students requested that CAI be used for instruction

in other lessons, 2.55% said they would like to have

more available exercises, and 1.7% indicated a desire

to have the ability to study at home with CAI. Eight

students (9.4%) expressed displeasure that the playback

button was necessary in the exercise section. Regarding

student behavior during the study period, students in the

control group demonstrated a low level of attention and

minimal active learning behavior. In contrast, the CAI

group demonstrated much higher levels of attentiveness

during the study period (Table 5).

DISCUSSION

Our null hypothesis in this study (primary outcome:

eoretical instruction pass rate of >80% and practical

instruction pass rate of >70%) was not able to be rejected

based on the combined pass rate (theory and practical) in

Fig 1. Flow diagram of participant eligibility and randomization process.

Yuchareon et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

328

Original Article

SMJ

TABLE 4. Student level of satisfaction with computer-assisted instruction (CAI) (n=58).

Topic Level of satisfaction

Mean±SD

Total for content presentation 4.57±0.12

Presentation easy to understand 4.67±0.52

Language easy to understand 4.64±0.61

Appropriateness of lesson sequence 4.64±0.55

 Appropriatenessoflessoncontentdifculty 4.48±0.65

Appropriateness of number of exercises in the lesson 4.40±0.76

Total for design of learning media 4.39±0.20

Convenient use of button 4.05±0.98

Composition of media learning appropriate and interesting 4.48±0.68

Appropriateness and clarity of characters 4.58±0.62

Appropriateness of sound effects 4.40±0.90

Ability of media learning to motivate students 4.45±0.76

Total for lesson satisfaction 4.65±0.06

Students are attentive and enjoy the lesson 4.58±0.59

 Studentsndtheanswerbythemselvesandcanrepeatthelesson 4.62±0.62

 Studentsaresatisedwiththeoutcomeofthelesson 4.62±0.53

Students can easily study by themselves 4.66±0.52

Students can study independently 4.75±0.46

Overall total 4.53±0.17

TABLE 5. Student learning behavior in the CAI and control groups.

Behavior CAI Group (n=85) Control Group (n=85)

Display interest in and Students display interest in and pay Students do not pay attention

attention to media learning attention to media learning and to or display interest in media

ask staff when they have a question learning; students look tired and

unmotivated

Half of students watch another

video that is not related to the lesson

Behavior during media Some students talk to each other,

Students play on their mobile telephones,

learning but most pay attention to media learning talk to each other, sleep, and/or wait

Some students reset their lesson to for the time they can leave the room

repeat again

Behavior after media Students cooperate with each other Students continue to display a lack of

learning to answer questionnaire interest in learning

Some students ask their friends about Some students remain attentive to

the lesson and how to use media their lesson as their friends leave

learning the room

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

329

the study group of only 24.7%. Students became nervous

during the examination, causing them to demonstrate

incorrect process of nursing care. Although most students

knew the related processes, a loss of concentration and

test anxiety adversely aected their scores. Previous study

of the eects of instructional videos on knowledge and

practical skills of labor and delivery nursing students

demonstrated that these learning materials were useful

in improving student knowledge in clinical practice

during the intra partum period.

Similar to the results of

a comparative study on the eectiveness of two teaching

medias: use of video and use of ip chart instructions on

the knowledge of postpartum woman, the video group in

this study had higher practicum and satisfaction scores

than those of the control group.

However, one-time use

of CAI for students to practice their NG tube healthcare

skills could not improve student skills in this study. As

suggested by several opinions, students require several

learning and practice opportunities, including repeated

demonstrations, before they should be asked to perform a

skill.

3,7

A study that used CAI to enhance student critical

thinking ability found a dierence in knowledge scores

between pre-test and post-test scores, but there was no

dierence in critical thinking skills.

8,9

Exposing students

to new types of CAI has been proposed.

Park EJ found computer-assisted instruction (CAI)

was likely to be useful and practical in the training of

abstract skills to nursing students, though certain challenges

remain, such as the precise understanding of cognitive

or aective responses to ethical issues.

eir nding

suggests eective use of the constructivist learning theory,

which actively emphasizes students as the key factor in

improving learning outcomes. e variability in teaching

levels and instructional resources that are available via

CAI enables students to develop knowledge and skills

and construct meaning according to his/her preferred

style of learning. Attractive and eye-catching colors,

endearing animated characters, music and sound eects,

and compelling visual eects and animation eectively

attract student attention and create a fun and relaxed

learning atmosphere. CAI methods can achieve excellent

results, with students’ reading and speaking abilities being

continuously enhanced during the learning process.

ere is limited empirical evidence regarding the

effectiveness of CAI in clinical skills education.

previous study in Information and Communication

Technologies (ICTs) found many obstacles to CAI nursing

education.

ey reported age, previous knowledge, and

preferred learning style as factors that aected learning

of nursing clinical skills. ey also recommended the

following research guidelines for studies investigating

CAI instruction: randomized controlled trial study design,

avoid voluntary recruitment, and test for dierences in

the demographic characteristics of participants in the

study groups.

e results of this study revealed computer-assisted

instruction (CAI) to be useful in improving student

knowledge, but not in improving student practical skills.

In addition to instruction by CAI, demonstration of

practical skills by the instructor is required to improve

student practical skills. Based on the promising ndings

of this study, the authors recommend the integration

of CAI as an instruction method for students at the

School of Nurse Assistant Siriraj Hospital. Integration

should include other nursing care training subjects and

the use of pre-test and post-test to evaluate training

eectiveness. Moreover, the results of this study should

inform teachers of nurses and nurse assistants not only

to condently incorporate CAI for teaching theoretical

parts but also to build more “real skill” training with

patients in practical part.

ACKNOWLEDGMENTS

e authors gratefully acknowledge the students

who participated in this study and the School of Nurse

Assistant Siriraj Hospital for their kind cooperation

and support. The authors would also like to thank

Dr.Kullathorn ephamongkhol for his facilitation

and Ms.KesareePunlee for her support.

Conicts of interest and source of funding: is study

was funded by a grant from the Routine to Research Unit,

Faculty of Medicine Siriraj Hospital, Mahidol University.

REFERENCES

1. Caroline N, Catherine MF, Anne M, Tony F. Creative Arts:

An Essential Element in the Teacher’s Toolkit When Developing

Critical inking in Children. AJTE 2013;38:1-17.

2. DuBois MR, Volpe RJ, Hemphill EA.A randomized trial of a

computer-assisted tutoring program targeting letter sound

expression. SPR2014.p.210-21.

3. Hill NTM, Mowszowski L, Naismith SL, Chadwick VL, Valenzuela

M, Lampit A. Computerized Cognitive Training in Older

Adults with Mild Cognitive Impairment or Dementia: A

Systematic Review and Meta-Analysis. Am J Psychiatry 2017;

174:329-40.

4. Panich W. Instructor for student support to enjoy with learning

[Internet]. 2012 [cited 2014 November 15];1. Available from:

https://www.gotoknow.org/posts/504150

5. Kala S, Youngwanichsate S, Chunuan S. Eects of instructional

videos on the knowledge and labour practical skills of nursing

students. Songkla Med J 2008;26:111-21.

6. Sangwan M, Rojananirankit N. Comparative Study on the

Eectiveness of Two Teaching Medias: Use of Video and Use

of Flip Chart Instructions on the Knowledge of Postpartum

Yuchareon et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

330

Original Article

SMJ

Woman for Postpartum Practices and Breastfeeding Skills.

JPHN 2015;29:56-66.

7. Chaweethamawat A, Kongsaktrakul C, Kaveevivitchai C.

Development of Computer-Assisted Instruction on Knowledge

of Congenital Heart Disease for School-Aged Children with

Congenital Heart Disease. KJN 2012;19:118-34.

8. Longmuir KJ. Interactive computer-assisted instruction in

acid-base physiology for mobile computer platforms. Adv

Physiol Educ 2014;38:34-41.

9. Srisorachatr S,Huadong Y,Hudthagosol C,Danthanavanich

S. Computer assisted instructionon “learning nutrition ags

for deaf 5

grade and 6

grad students”: eectiveness ofinstruction.

J Med Assoc ai 2013;96 Suppl 5:49-54.

10. Park EJ. e development and implications of a case-based

computerprogram to train ethical decision-making. Nursing

Ethics 2013;20:943-56.

11. Carrillo de Gea JM, FernándezAlemán JL, Ana García BS.

Computer based nursing education: An integrative review of

empirical studies. JNEP 2012;2:162-72.

12. Kish G, Cook SA, Kis G. Computer-assisted learning in anatomy

at the International Medical School in Debrecen, Hungary: A

preliminary report. Anat Sci Ed 2013;6:42-47.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

331

Kingkaew Pajareya, M.D.*, Sureelak Sutchritpongsa, M.D.**, Ratcharin Kongkasuwan, M.D.*

*Department of Rehabilitation Medicine, **Department of Pediatric, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.

DIR/Floortime

Parent Training Intervention for

Children with Developmental Disabilities:

a Randomized Controlled Trial

ABSTRACT

Objective: To determine whether adding a parent training program utilizing the Developmental Individual-dierence

Relationship-based (DIR/Floortime®) approach for children with developmental disabilities can improve their

capacities for attention and initiation.

Methods: Forty-eight pairs of parents and their preschool children with developmental disabilities were randomly

assigned to three 1-hour one on one DIR/Floortime® parent training sessions during a 4-month period or to a

control group.

Results: Between-group comparison demonstrated a signicant dierence in favor of the intervention group for

attention (F(1, 46) = 34.52, p = .031, ph

= .099) and initiation composite scores (F(1, 46) = 6.55, p = .014, ph

.127). Eect sizes were medium to large for the attention composite score (Cohen’s d = .526) and initiation composite

score (Cohen’s d = .653) respectively.

Conclusion: Adding a modest home-based DIR/Floortime® parent training approach may yield clinically meaningful

improvements in attention and initiation for children with a range of developmental challenges.

Keywords: Child development; developmental disability; parent training; DIR/Floortime® (Siriraj Med J 2019; 71:

331-338)

Corresponding author: Kingkaew Pajareya

E-mail: kingkaew.paj@mahidol.ac.th

Received 5 March 2019 Revised 1 July 2019 Accepted 5 August 2019

ORCID ID: http://orcid.org/0000-0002-3098-1896

http://dx.doi.org/10.33192/Smj.2019.51

INTRODUCTION

Children with non-autistic developmental disabilities

typically exhibit challenging limitations in self-care and

are at high risk of developing emotional and behavioral

problems compared to typically developing children.

1-3

Studies of parent training for children with developmental

disabilities are promising.

4-7

ere is good evidence that

parent training can yield increased positive parenting and

reduction in problematic behaviors in children. ese

approaches utilize contingency management, planned

activities, compliance training, or a combination of

these. Some emphasize the importance of fostering more

positive relationships between parents and children.

8-13

ese strategies, however, are designed to be applied

in a generalized manner without explicitly taking into

account the neurological and developmental capacities

of individual children.

e Developmental Individual-dierences Relationship-

based (DIR)/Floortime® model

focuses on the child’s

developmental capacity for relating and communicating,

taking in, regulating, responding, understanding sensations

and information, and planning actions. It is a developmental

intervention involving meeting a child at his or her current

developmental level, in which the parent or therapist

Pajareya et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

332

Original Article

SMJ

follows the child’s lead, with playful positive attention

while tuning into the child’s interests. Once the child

connects with the adult, specic techniques are used

to challenge and entice the child to move toward ever

more complex developmental milestones outlined in

the DIR model. In DIR/Floortime® relationships are the

pivotal force that nurtures and optimizes development.

DIR/Floortime® parent training can improve the core

challenges of autism, including relating, interacting, and

communicating.

15-17

Prior to this study, most research on DIR has focused

on investigating the eectiveness of DIR/Floortime® in

the children with ASD. It has been less clear whether

DIR might help children with developmental disabilities

who generally have better social abilities than children

with ASD.

18-20

It would, therefore, be helpful to learn

whether DIR might help other vulnerable populations.

is study seeks to determine whether DIR/Floortime®

parent training program for children with development

disabilities confers benet beyond treatment as usual

(TAU).

MATERIALS AND METHODS

Study design and participants

Aer the Institutional Review Board of Faculty of

Medicine Siriraj Hospital, Mahidol University approved

this study (Si 089/2013), caregivers (parents, or an adult in

a caregiving role) were recruited through posters and

paper

advertising from February 2013 to December 2016.

One of the

researchers who was a developmental pediatrician

conrmed

the diagnosis. Inclusion criteria included: patient:

diagnosed

with a developmental disability, global developmental

delay, or a cerebral palsy; age 2 to 6 years; and living

with the caregiver for at least 6 months. Children with

co-occurring ASD, signicant impairments in hearing

or vision serious medical problems, intractable seizures,

and those whose caregivers were illiterate were excluded.

Aer the participants agreed and signedinformed

consent and had undergone a baseline assessment, the

children were sorted into 2 strata according to age (24-

47 or 48-72 months).

e study is a single-blind, parallel design clinical

trial with an allocation ratio of 1: 1. A research assistant

who was not directly involved with the study used a

random number table to randomly allocate participants

to either the DIR/Floortime® parent training group or a

waitlist control group. Opaque envelopes were used to

conceal the allocation sequence.

Intervention

Treatment was carried out using the DIR/Floortime®

approach. e rst author was an experienced clinician

who had her expert training leader certication from the

Interdisciplinary Council on Developmental and Learning

Disorders. Caregivers in the intervention group viewed

a 2-hour DVD about DIR/Floortime® and received an

accompanying book, an approach previously utilized to

study the method with parents of children with ASD.

16,21

Caregivers were trained for 1 hour on how to observe

their children’s cues and responses. Caregivers were

coached to join the children in pleasurable activities,

maintain engagement, and to sustain reciprocal interactions.

For children with symbolic capacity, pretend play was

encouraged. Controlling and intrusive responses on the part

of caregivers were identied, discouraged, and substituted

with two-way emotional gestural communication.

Caregivers in the intervention group were asked to

spend a minimum of 15 hours per week utilizing DIR/

Floortime® techniques throughout the study period. Parents

of the children in both groups were asked to continue

their other usual treatment interventions such as physical

therapy or speech therapy throughout the study period.

At the end of the rst and third months, the researcher

observed and gave feedback to parents in the intervention

group about how well they were related to their children.

e assessment was based on the child and caregiver’s

functional emotional capacities.

Controlling and intrusive

responses were again identied and discouraged and

substituted with responses aimed at facilitating two-way

emotional signaling and communication between parents

and their children. Modeling and coaching were used

to improve caregiver performance. e goals, method,

and techniques of home program were adjusted to meet

the child’s current needs. Both groups were reassessed

at the end of the fourth month.

Measurements

Baseline measurement included assessment of

the children’s development and family demographic

characteristics.

Child Behavior Rating Scale (CRBS)

e CBRS

was the primary outcome measurement,

rating the children’s interactions with their caregivers. e

caregiver and child were video-recorded while playing

together in any manner they chose. e assessors were

two developmental psychologists who were blinded

to the group status. ey rated Attention (attention

to activity, persistence, involvement, and compliance

or cooperation subscales), and Initiation (initiation of

activity, joint attention, and aect). e assessors rated

each of the items on a 5-point Likert scale. e two

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

333

assessors’ scores were averaged and subscale scores were

averaged for composite scores. Cronbach’s alpha for the

two assessors was 0.94 for the attention composite score

and 0.77 for initiation.

Functional Emotional Developmental Questionnaire

(FEDQ)

e FEDQ

is a questionnaire for the parent to assess

Greenspan’s six Functional Emotional Developmental

Levels (FEDLs): (1) shared attention and regulation,

(2) engagement and relating, (3) purposeful emotional

interaction, (4) social problem-solving, (5) creating

ideas, and (6) logical use of symbolic ideas in social

problem solving. All are scored on a 7-point scale. e

ai version of the FEDQ has a satisfactory internal

consistency (Cronbach’s alpha = 0.83) with an ICC of

0.89.

Mullen Scales of Early Learning (MSEL)

e MSEL

is a developmental test for children from

birth to 68 months of age. Four cognitive domains of

the MSEL were used: visual reception, ne motor skills,

receptive language, and expressive language. is instrument

was administered by a developmental pediatrician who

was blinded to intervention status. Subscale scores are

compared against age equivalents and summarized as

the Early Learning Composite (ELC). Higher scores

reect more intact development.

Parenting Stress Index-Short Form (PSI-SF)

e PSI –SF

is a 36-item questionnaire designed

to measure stress in the parent-child system on three

subscales: parental distress, parent-child dysfunctional

interaction, and dicult child. Higher scores indicate

greater stress. e ICC of the total Parenting Stress Index

score in the current study was 0.82.

Compliance and co-interventions

Parents kept logs of hours per week they used DIR/

Floortime® technique as well as other interventions.

Sample size calculation

A prior study found a mean CBRS score (initiation)

of 2.8 [standard deviation (SD), 0.9] for a population of

children with developmental disorders other than autism.

In the present study, we sought to detect a minimum

clinically important between-group dierence of 0.8.

A sample size of 40 was needed to provide 80% power to

detect this dierence at a two-tailed signicance level of

0.05. If the estimated drop-out rate was 20%, the number

of subjects to be recruited was 48.

Data analysis

Data were analyzed using SPSS, version 18.0. An

intent-to-treat analysis was used. For drop-outs, we used

the last-observation-carried-forward method. Paired

t-tests compared pre-test and post-test scores. Analysis

of covariance was conducted to investigate the impact of

the treatment intervention between groups, with follow-

up scores as the dependent measure and baseline scores

as covariates.

RESULTS

Sixty families were screened: 4 declined to participate,

2 did not respond to the baseline evaluation, and 6 were not

eligible because of the children’s health problems. Forty-

eight families were randomized to either the treatment

group (n = 23) or control (n = 25). Six families dropped

out, 3 from each group. Reasons included an inability to

reach the caregivers, moving away, and ‘too busy’. (Fig 1)

e average age of the children was 39.98 months.

Sixty-eight percent of the diagnosis was global developmental

delay. Most were living in a two-parent household.

Seventy-ve percent had an ELC of ≤50, indicating

moderate to severe developmental delay. Eighty-ve

percent received usual services averaging 1.14 hours per

week. Groups did not signicantly dier concerning for

average service hours. e parent and child demographic

characteristics were not signicantly dierent between

groups (Table 1). We used intention to treat analysis

and the last-observation-carried-forward method for

missing data.

Primary outcome

Control group post-test attention and initiation

composite scores of the CRBS were not signicantly

dierent from pre-test scores. Intervention group post-test

attention and initiation composite scores were signicantly

higher than pre-test scores (p < .001). Between-group

comparison demonstrated a signicant dierence in

favor of the intervention group for attention (F(1, 46)

= 34.52, p = .031, pη

= .099) and initiation composite

scores (F(1, 46) = 6.55, p = .014, pη

= .127). Eect size

for the dierence between groups was medium to large

for both the attention composite score (Cohen’s d =

.526) and initiation composite score (Cohen’s d = .653)

(Table 2).

Subscale results demonstrated signicantly better

outcomes for the persistence, involvement, initiation

of activity, and joint attention subscales. ere were no

signicant dierences in attention to activity, compliance,

or aect (Table 2).

Pajareya et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

334

Original Article

SMJ

Control group Intervention group

P value

N=25 N=23

The children

Mean (SD) age in months 39.36 (11.4) 40.65 (13.6) .301

Male 11 (44%) 14 (56%) .243

Diagnosis

Down syndrome 2 (8%) 4 (17%)

Cerebral Palsy 7 (28%) 7 (30%) .561

Global developmental delay 16 (64%) 12 (52%)

Mean (SD) CBRS

Attention composite scores 3.04 (0.90) 2.85 (0.93) .473

Initiation composite scores 2.91 (0.77) 2.53 (0.77) .096

Mean (SD) FDQ 32.24 (11.21) 29.61 (11.03) .564

Early Learning Composite

More than 85 3 (12%) 2 (9%)

71-85 4 (16%) 1 (4%) .538

50-70 7 (28%) 9 (39%)

Less than 50 11 (44%) 11 (48%)

History of convulsion 1 (4%) 2 (9%) .532

Participation in daycare or kindergarten 5 (20%) 7 (30%) .617

Their families

Primary caregiver:

Parent / Grandparent 19 (76%)/ 6 (24%) 15 (65%)/ 8 (35%) .412

Bachelor degree or higher 11 (44%) 6 (26%) .195

Two-parent household 22 (88%) 20 (87%) .195

Worked inside of the home 14 (56%) 11 (48%) .466

Sibling in family (yes/ no) 19 (76%) 20 (87%) .571

Mean (SD) Total PSI score 87.88 (21.09) 93.52 (27.5) .390

TABLE 1. Demographic information and statistical tests of group dierences for participants.

Secondary outcomes

Intervention group post-test FEDQ scores were

signicantly higher than the pre-test score (p = .001) but

not signicant in the control group (p = .099). Between-

group analysis revealed signicant dierences in favor

of the invention (F(1, 46) = 7.37, p = .031, pη

= .141).

e eect size for the dierence between groups was

medium to large (Cohen’s d = .729).

For both groups, post-test MSEL subscale scores were

signicantly higher than the pre-test scores. Between-

group analysis showed that the invention group had

signicantly better scores for the receptive language

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

335

TABLE 2. Presents mean (SD) of the outcome measures and the results of the analysis of covariance (ANCOVA)

for group analyses.

Control Group Intervention Group

(n=25) (n=23) F

P value Cohen’s d

(1, 46)

Pre-test Post-test Pre-test Post-test

Child Behavior Rating Scale (CBRS): Attention Domain

Attention to Activity 3.40 (1.00) 3.44 (1.12) 3.21 (1.08) 3.54 (0.91) 2.12 .153 .421

Persistence 2.80 (0.92) 3.04 (0.97) 2.61 (1.08) 3.19 (1.03) 4.21 .046* .574

Involvement 3.02 (1.04) 3.24 (1.01) 2.83 (0.94) 3.48 (0.74) 4.95 .031* .570

Compliance 2.92 (0.98) 3.02 (0.94) 2.74 (0.94) 2.96 (0.94) 0.18 .671 .116

Attention composite scores 3.04 (0.90) 3.18 (3.29) 2.85 (0.93) 3.29 (0.79) 4.94 .031* .526

Child Behavior Rating Scale (CBRS): Initiation Domain

Initiation to activity 2.80 (1.00) 2.84 (1.04) 2.26 (0.93) 2.87 (0.98) 11.99 .009 .944

Affect 2.62 (0.83) 2.82 (1.06) 2.43 (0.83) 3.00 (0.87) 3.35 .074 .527

Joint attention 3.30 (0.93) 3.34 (0.85) 2.89 (0.80) 3.30 (0.78) 4.91 .032* .443

Initiation composite scores 2.91 (0.77) 3.30 (0.86) 2.53 (0.77) 3.06 (0.77) 6.55 .014* .653

Total FEDQ Score 32.24 (11.21) 33.76 (11.81) 29.61 (11.03) 37.35 (11.37) 7.37 .009* .729

Mullen Scales of Early Learning (MSEL)

Visual Reception 23.68 (9.26) 26.04 (9.57) 20.91 (11.9) 25.35 (11.31) 1.97 .167 .648

Fine Motor 21.64 (7.41) 23.80 (8.03) 20.48 (8.12) 22.91 (8.91) 0.079 .780 .170

Receptive Language 21.56 (8.89) 22.92 (9.45) 19.17 (10.33) 23.55 (11.15) 4.33 .043* .663

Expressive Language 17.64 (10.69) 18.68 (9.45) 15.39 (9.85) 18.90 (9.85) 5.53 .023* 1.11

Parental Stress Index

Parental Distress 30.08 (8.92) 29.04 (7.20) 31.43 (10.6) 29.09 (9.83) 0.148 .702 .140

Parent-Child 29.72 (6.10) 28.64 (6.96) 28.09 (8.28) 32.00 (7.98) 5.27 .026* .414

Dysfunctional Interaction

 DifcultChild 29.16(8.89) 28.6(8.16) 30.74(9.82) 30.09(9.12) 0.237 .629 .112

Total Parental Stress Index 87.88 (21.09) 85.30 (24.28) 93.52 (23.98) 85.03 (24.29) 1.240 .271 .275

* p value < 0.05

Pajareya et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

336

Original Article

SMJ

subscale (F(1, 46) = 1.97, p = .043, pη

= .088) and the

expressive language subscale (F(1, 46) = 1.97, p = .043,

pη

= .109). ere was no signicant dierence in the

visual reception or ne motor subscales. Mean changes

in the ELC score in this study could not be calculated

because scores were lower than 49.

Intervention group PSI-SF post-test parent–child

dysfunctional interaction score was signicantly lower

than pre-test (p = .011). Parental stress and dicult child

scores did not signicantly change from baseline. Between-

group analysis showed that the parent–child dysfunctional

interaction score in the intervention group was the only

individual subscale score that was signicantly higher

than that in the control group when the pre-test score

was a covariate (F(1, 46) = 5.27, p = .026, pη

= .033).

No other adverse effects of the intervention were

detected.

Moderator analysis

Regression analysis was carried out to determine the

moderator eect of baseline variables on the improvements

in the attention and initiation composite scores. Age,

groups of children classied by ELC, and baseline score

demonstrated some association with the improvement

in the attention (R

= .284, F(3, 44) = 5.83, p = .002) and

initiation composite scores (R

= .381, F(3, 44) = 5.83,

p = .001). Pre-test attention and initiation composite

scores had a signicant negative weight, indicating the

children with higher baseline scores might have a lower

chance of improving. Age of the children was signicantly

associated with the improvement in initiation composite

scores, i.e., older children were more likely to show

improved initiation. Pre-test ELC did not demonstrate

moderator eect on both composite scores.

Compliance and co-interventions

35% of the intervention group reported using DIR/

Floortime® techniques an average of 14 hours/week;

52% spent 7 to 14 hours/week; 13% spent <7 hours/

week. Children in both the control and intervention

groups received additional services (e.g., speech therapy,

occupational therapy, or physical therapy) for an average

of 1.20 (SD, 0.58) hours/week and 1.08 (SD, 0.73) hours/

week, respectively. ere were no signicant dierences

in the time spent receiving such services between the

two groups (p = .554).

DISCUSSION

is study supports the feasibility of studying DIR/

Floortime® parent training intervention across groups of

children beyond those with autism.

15-17,28

Children with

non-autistic developmental disorders in the intervention

group showed signicantly greater involvement and had

better communication with their parents or caregivers.

Eect sizes were medium to large. e lack of signicance

in CBRS changes in compliance or cooperation and

attention to activity scores between groups is similar to

Casenhiser.

However, DIR/Floortime® does not focus

on compliance nor activity but on encouraging initiation

and interaction.

In this research protocol, we asked parents to spend

at least 15 hours per week using the intervention with

their children. Parent logs showed that only one-third

of parents met the requested time. e studies of parent

training programs found that when the programs were

implemented with high delity, the parenting practices

improved signicantly, but the eect was much less when

implementation delity was low.

29-30

Program adherence

relates to the content and dose of the intervention. It

may be moderated by the quality of training, supporting

system, and participant responsiveness. More research is

needed to clarify the moderating impact of the components

included here in order to encourage the family to practice

DIR/Floortime® at home.

Further analysis was conducted to compare gains

in attention and initiation composite scores of CBRS

between the parents in the invention groups who spent

dierent amounts of time using the DIR intervention.

ere is no statistical dierence in improvements for

those children whose parents documented spending more

or less time using the intervention with their children.

A similar analysis was attempted to look at the eect of

the dierent kinds of developmental disabilities across

children, however, no statistical dierence was detected.

is nding was likely due to the small number of subjects.

Prior studies have noted a concern that parent-

mediated interventions might cause increased parental

stress.

Of additional note, we found a statistically elevated

Parent-Child Dysfunctional Interaction Score of PSI.

We interpret this nding in the context of both the

process of training and the improvement in initiation

and attention. On the one hand learning techniques for

improving interactions might increase parental awareness

of the problematic aspects of the relationship. Further,

when a child is more able to attend to what she wants

and is more able to initiate, this too may create a feeling

in the parent that their relationship is harder. ese

concerns should be studied in more detail in future

studies, particularly in light of no dierences between

groups in the Parental Distress Score of PSI and total PSI

scores. Qualitative assessments might shed further light

on the clinical impact of DIR for individual families.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

337

Limitations

e main limitation of the current study were,

rst, the families in the intervention group used DIR/

Floortime® in addition to their children’s routine care,

whereas the control group received only their routine care.

Improvements with intervention may be attributable to

any amount of increase in time spent between caregivers

and children.

While the eects of DIR Floortime® parent training

are promising at four months, long-term eects of the

intervention are unknown. Further study with more

diverse caregivers, better stratication of the children

regarding their challenges, and longer follow-up are

needed.

Summary

Adding parent training intervention and the home-

based DIR/Floortime® for 1-2 hours per day for four

months produced statistically signicant and clinically

meaningful improvements. Further research should

investigate long-term outcomes and the effects on

parents’ perception about their child, themselves, and

their relationships with their children.

ACKNOWLEDGMENTS

is research project was supported by Faculty of

Medicine Siriraj Hospital, Mahidol University (Grant

Number R15632012). We would like to sincerely thank

to Dr. Devin M. Casenhiser and Dr. Joshua Feder who

provided us with intellectual assistance and editing this

manuscript.

REFERENCES

1. Baker B, Blacher J, Crnic K, Edelbrock C. Behavior problems

and parenting stress in families of three-year-old children with

and without developmental delays. Am J Ment Retard 2002;107:

433-44.

2. Green VA, O’Reilly MO, Itchon J, Sigafoos J. Persistence of

early aberrant behavior in children with developmental disabilities.

Res Dev Disabil 2005;26:47-55.

3. Fenning RM, Baker BL, Juvonen J. Emotion discourse, social

cognition, and social skills in children with and without

developmental delays. Child Dev 2011;82:717-31.

4. Harrold M, Lutzker JR, Campbell RV, Touchette PE. Improving

parent–child interactions for families of children with developmental

disabilities. J Behav er Exp Psychiatry 1992;23:89-100.

5. Huynen KB, Lutzker JR, Bigelow KM, Touchette PE, Campbell RV.

Planned activities training for mothers of children with

developmental disabilities. Behav Modif 1996;20: 406-27.

6. Hudson AM, Matthews JM, Gavidia-Payne ST, Cameron

CA, Mildon RA, Radler GA, et al. Evaluation of an intervention system

for parents of children with intellectual disability and challenging

behavior. J Intellect Disabil Res 2003;47:238-49.

7. Roberts C, Mazzucchelli T, Studman L, Sanders MR. Behavioral

family intervention for children with developmental disabilities

and behavioral problems. J Clin Child Adolesc Psychol 2006;

35:180-93.

8. Hampel OA, Hasmann SE, Schaadt A, Holl R, Petermann F,

Hasmann R. Stepping stones parenting training as an eective

support for families with a handicapped child. Kindh Entwickl

2010;19:36-46.

9. Webster-Stratton C, Hammond M. Treating children with

early-onset conduct problems: A comparison of child and

parent training interventions. J Consult Clin Psychol 1997;65:

93-109.

10. Roberts C, Mazzucchelli T, Studman L, Sanders MR. Behavioral

family intervention for children with developmental disabilities

and behavioral problems. J Clin Child Adolesc Psychol

2006;35:180-93.

11. McIntyre LL. Parent training for young children with

developmental disabilities. Randomized controlled trial. Am

J Ment Retard 2008;113:356-68.

12. Whittingham K, Wee D, Boyd R. Systematic review of the

ecacy of parenting interventions for children with cerebral

palsy. Child Care Health Dev 2011;37:475-83.

13. Roux G,Sofrono K,Sanders MR. A randomized controlled

trial of Group Stepping Stones Triple P for families of children

with disabilities. Fam Process 2013;52:411-24.

14. Greenspan SI, Weider S. Developmental patterns and outcomes

in infants and children with disorders in relating and

communicating: A chart review of 200 cases of children with

autistic spectrum diagnoses. J Dev Learn Disord 1997;1:87-141.

15. Solomon R, Necheles J, Ferch C, Bruckman D. Pilot study of a

parent training program for young children with autism: the

PLAY Project Home Consulting Program. Autism 2007;11:

205-24.

16. Pajareya K, Nopmaneejumruslers K. A pilot randomized

controlled trial of DIR/ Floortime™ parent training intervention

for pre-school children with autistic spectrum disorders.

Autism 2011;15:563-77.

17. Casenhiser DM, Shanker SG, Stieben J. Learning through

interaction in children with autism: preliminary data from

asocial-communication-based intervention.Autism 2013:17:

220-41.

18. Rodrigue JR, Morgan SB, Geen GR. A comparative evaluation

of adaptive behavior in children and adolescents with autism,

Down Syndrome, and normal development.J Autism Dev

Disord 1991;21:187-96.

19. Gillham JE, Carter AS, Volkmar FR, Sparrow SS. Toward a

developmental operational denition of autism.J Autism Dev

Disord 2000;30:269-78.

20. Jackson CT, Fein D, Wolf J, Jones G, Houck M, Waterhouse

L,et al. Responses and sustained interactions in children with

mental retardation and autism. J Autism Dev Disord 2003;33:

115-21.

21. Pajareya K, Nopmaneejumruslers K. A one-year prospective

follow-up study of a DIR/ Floortime parent training intervention

for pre-school children with autistic spectrum disorders. J

Med Assoc ai 2012;95:1184-93.

22. Greenspan SI, DeGangi G. Research on the FEAS: Test development,

reliability, and validity studies. In: Greenspan SI, DeGangi G,

Wieder S, eds. e functional emotional assessment scale (FEAS)

for infancy and early childhood: Clinical and research applications.

Pajareya et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

338

Original Article

SMJ

Bethesda: Interdisciplinary council on developmental and

learning disorders; 2001.p.167-247.

23. Mahoney G, Perales F. Relationship-focused early intervention

with children with pervasive developmental disorders and

other disabilities: a comparative study. J Dev Behav Pediatr

2005;26:77-85.

24. Greenspan J, Greenspan SI. Functional emotional developmental

questionnaire (FEDQ) for childhood: A preliminary report

on the questions and their clinical meaning. J Dev Learn Disord

2002;71:71-116.

25. Pajareya K, Sutchritpongsa S, Sanprasath N. Sensitivity and

specicity of the ai version of the Functional Emotional

Developmental Questionnaire (FEDQ) for Childhood. Siriraj

Med J 2014;66:168-78.

26. Mullen EM. Mullen Scales of Early Learning (AGS ed). Circle

Pines: American Guidance Service, 1995.

27. Abidin R. Parenting Stress Index, 3

Edition. Odessa: Psychological

Assessment Resources, 1995.

28. Casenhiser DM, Binns A, McGill F, Morderer O, Shanker SG.

Measuring and supporting language function for children

with autism: evidence from a randomized control trial of a

social-interaction-based therapy. J Autism Dev Disord 2015;45:

846-57.

29. Forgatch M, Patterson G, DeGarmo D. Evaluating delity:

predictive validity for a measure of competent adherence to

the Oregon Model of Parent Management Training. Behav

er 2005;36:3-13.

30. omas RE, Baker P, Lorenzetti D. Family-based programmes

for preventing smoking by children and adolescents. Cochrane

Database Syst Rev 2007; CD004493-1.

31. Herring S, Gray K, Tae J, Tonge B, Sweeney D, Einfeld S.

Behaviour and emotional problems in toddler with pervasive

developmental disorders and developmental delay: Associations

with parental mental health and family functioning. J Intellect

Disabil Res 2006;50:874-82.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

339

Wannaphorn Rotchanapanya, M.D.*, Weerapat Owattanapanich, M.D.**, Nonlawan Chueamuangphan, M.D.*

*Department of Medicine, Chiangrai Prachanukroh Hospital, Chiang Rai 57000, **Department of Medicine, Faculty of Medicine Siriraj Hospital,

Mahidol University, Bangkok 10700, ailand.

Predictive Factors for Survival Outcomes of High-

Risk Febrile Neutropenic Patients: a 3-Year Study

at a Single Center in Thailand

ABSTRACT

Objective: is study aimed to identify the risk factors associated with mortality in febrile neutropenic patients.

Materials and Methods: is 3-year, single center, retrospective, observational study was conducted at Chiangrai

Prachanukroh Hospital, Chiangrai Province, ailand. e inclusion criteria consisted of a patient age of over 15

years and a diagnosis of febrile neutropenia.

Results: Most of the 303 febrile neutropenic inpatients had a Multinational Association for Supportive Care in

Cancer (MASCC) risk score < 21. e median length of stay was 6 days (interquartile range: 4-11 days). During 30

days of admission, 24.8% of the patients succumbed. In a univariate analysis, patients receiving G-CSF, the post-

chemotherapy-related group, patients with MASCC score > 16, and patients admitted in private had signicantly

higher survival rate. In a multivariate analysis, a MASCC score ≤ 16 and non-chemotherapy-related groups were

associated with an increased mortality risk.

Conclusion: e 30-day survival rate of febrile neutropenic patients in ailand is seventy-ve percent. Low MASCC

score and non-chemotherapy-related neutropenia are associated with a higher risk of unfavorable outcomes.

Keywords: Febrile neutropenia; risk factors; mortality; ailand; developing country (Siriraj Med J 2019; 71: 339-348)

Corresponding author: Wannaphorn Rotchanapanya

E-mail: rot.wannaphorn@gmail.com

Received 18 April 2019 Revised 24 May 2019 Accepted 2 July 2019

ORCID ID: http://orcid.org/0000-0003-4679-8647

http://dx.doi.org/10.33192/Smj.2019.52

INTRODUCTION

Febrile neutropenia is usually accompanied by other

signs and symptoms of infection. e treatment outcomes

of these patients, especially those in the high-risk group,

are associated with high morbidity and mortality.

In cases

with severe infections, previous guidelines recommend

that patients should be admitted to hospital.

2,3

e hospital

stays of febrile neutropenic patients with solid tumors

and hematological malignancies have been reported to be

approximately 1 to 2 weeks.

Most febrile neutropenia occurs

during chemotherapy, especially in cases of hematologic

malignancies and solid cancers.

Moreover, it could result

from other neutropenia-causing diseases, for example,

aplastic anemia, drug-induced agranulocytosis, HIV

infections, and autoimmune diseases.

6-9

e treatment

is costly due to both the length of the hospital stay and

the associated complications. A retrospective study

has reported the clinical and economic burden arising

from the hospitalization of this patient group. e mean

duration of hospitalization with febrile neutropenia in

all cancer patients was 6 days and the median cost was

$8,376 per episode of febrile neutropenia.

ese problems are present in all countries. From a

previous study of patients with febrile neutropenia in a

developed country, the mortality rate during admission

was 10%, and the mean hospitalization duration was

Rotchanapanya et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

340

Original Article

SMJ

8.6 days.

e outcomes for developing countries still

need to be documented. e present cohort study aimed

to demonstrate data on the predictive factors and survival

outcomes of patients with febrile neutropenia at a single

center in a developing country.

MATERIALS AND METHODS

is 3-year, single center, retrospective, observational

study was conducted at Chiangrai Prachanukroh Hospital,

Chiangrai Province, ailand. It focused on patients

with febrile neutropenia who had been admitted to

medical wards between January 1, 2014 and December

31, 2016. e inclusion criteria were (1) a patient age of

over 15 years and (2) a diagnosis of febrile neutropenia.

e patients were identied through the ICD-10-CM

diagnosis code database using the codes for fever (R50)

and agranulocytosis (D70). e primary outcome of the

study was to identify the risk factors associated with

mortality in febrile neutropenia. e clinical characteristics,

clinical outcomes, laboratory ndings, treatments and

complications were recorded as secondary outcomes.

is study was approved by the Ethical Committee for

Research in Human Subject at Chiangrai Prachanukroh

Hospital (0032.102/15693).

Terminology

Febrile neutropenia is a condition of fever > 38.3°C

or > 38°C for 1 hour, with an absolute neutrophil count

(ANC) of ≤ 500 cells/cu mm, or an ANC of ≤ 1,000 cells/

cu mm with an expected nadir of ≤ 500 cells/cu mm.

e Multinational Association for Supportive Care

in Cancer (MASCC) risk index is a scoring method for

identifying low-risk of complications in cancer patients

with febrile neutropenia. is score comprises the following

factors: burden of febrile neutropenia, hypotension,

chronic obstructive pulmonary disease, previous fungal

infection, dehydration, and age. Scores ≥ 21 signify

a low-risk group of complications, while scores < 21

indicate a high-risk group of complications in patients

with febrile neutropenia.

Multi-drug resistant (MDR) organism is dened

as acquired non-susceptibility to at least one agent in

three or more antimicrobial categories.

Severe infection

is determined as severe sepsis by clinical of systemic

inflammatory response syndrome with evidence of

infection and end-organ dysfunction such as altered

mental status, hypotension, oliguria (urine output < 0.5

ml/kg/hr) or hypoxia.

Statistical analysis

PASW Statistics for Windows, version 18.0 (SPSS

Inc., Chicago, IL, USA) was used to analyze the data. e

sample size (330 cases) was calculated by the proportion

of mortality rate between two groups from the previous

study.

e critical value was 1.96 with 90% power.

Continuous variables were represented as mean ± standard

deviation, or median and interquartile range (IQR). For

the continuous variables comparison consisting of age,

complete blood counts and blood chemistry investigations,

the results were calculated by the Mann–Whitney U test

or student’s t-test. Categorical variables including gender,

diseases, MASCC risk index, infections and clinical

outcomes were presented as frequency and percentage,

and they were compared using Fisher’s exact or chi-

square test. e survival analysis and predictive factors

associated with survival were compared using the log-rank

test and presented as a Kaplan–Meier survival curve. A

Cox proportional hazards analysis (enter method) was

used for the univariate and multivariate factors associated

with survival outcome, and they were exhibited as hazard

ratio (HR) and 95% CI. A p-value < 0.05 was dened as

statistically signicant.

RESULTS

Baseline characteristics of febrile neutropenic patients

is study enrolled 303 cases of febrile neutropenic

patients. In all, 328 cases were admitted to medical wards

during the 3-year study period. Of the included cases,

twenty-ve were excluded: twenty-two were incompatible

with the diagnostic criteria of febrile neutropenia, and

the remaining three had inadequate documentation. e

mean age was 53±2 years, and females predominated

(51.4%). A large number of patients had hematologic

diseases and solid cancers, which included acute leukemia

(27.7%), solid tumors (21.1%), aplastic anemia (18.8%),

and lymphoma (18.2%). e clinical severity of the patients

was determined by their MASCC risk score. e median

was 16 (IQR: 13-18), with 91.4% (277 cases) having a

score below 21, indicating that they had a high-risk of

febrile neutropenia.

According to the initial laboratory ndings, the

complete blood counts showed a median hemoglobin of

8.0 g/dl (IQR: 6.0–9.0 g/dl); a median ANC of 150 cells/

cu mm (IQR: 13–385 cells/cu mm); a median platelet

count of 28,000 cells/cu mm (IQR: 10,000-80,000 cells/

cu mm). All initial laboratory results are summarized

in Table 1.

Infectious etiologies, treatments, and survival outcomes

Organ-specic symptoms of infection were seen in

57.8% of patients. Respiratory tract infections were the most

common (22.8%); they were followed by gastrointestinal

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

341

TABLE 1. e demographic and clinical characteristics and laboratory ndings of febrile neutropenic patients.

Febrile neutropenic patients

(n = 303)

Positive results (%)

Age (years), mean ± SD 53 ± 17

Gender

Female 166 (54.8%)

Diseases

Solid cancers 64 (21.1%)

Acute leukemia 84 (27.7%)

Aplastic anemia 57 (18.8%)

Lymphoma 55 (18.2%)

Myelodysplastic syndrome 10 (3.3%)

Multiple myeloma 3 (1.0%)

Infections 23 (9.6%)

Autoimmune disease 1 (0.3%)

Median MASCC score 16 (13–18)

0-10 63 (20.8%)

11-20 214 (70.6%)

21-23 26 (8.6%)

Laboratoryndings

Complete blood count

Median Hb level (g/dl) 8.0 (6.0–9.0)

Median WBC count (cells/cu mm) 1,100 (600–1,800)

Median ANC count (cells/cu mm) 150 (13–385)

Median platelet count (cells/cu mm) 28,000 (10,000–80,000)

Blood chemistry

Median creatinine (mg/dl) 1 (0.7–1.0)

Median albumin (g/dL) 3.0 (2.6–3.6)

Median total bilirubin (mg/dl) 1 (0.6–1.6)

Median direct bilirubin (mg/dl) 0.2 (0–0.7)

Median aspartate aminotransferase (U/L) 30 (20–52)

Median alanine aminotransferase (U/L) 22 (14–42)

Abbreviations: MASCC = the Multinational Association for Supportive Care in Cancer; SD = standard deviation; WBC = white blood cell;

ANC = absolute neutrophil count

Rotchanapanya et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

342

Original Article

SMJ

infections (16.8%). Nearly 20% of the patients had positive

blood culture. e most common infectious organisms

were gram negative bacteria: Acinetobacter spp. (9.2%),

Escherichia coli (8.3%), Pseudomonas spp. (5.3%), and

Klebsiella spp. (5%). Furthermore, multidrug-resistant

bacterial infections were recorded at 14.5%, while fungal

infections were present in a small number of patients.

ese results are illustrated in Table 2.

The median length of hospital stay was 6 days

(IQR: 4-11 days). The median duration of an ANC

below 500 cells/cu mm was 3 days (IQR: 2-7 days).

Approximately 85% of cases were admitted to a common

ward. Eighty-three percent of the patients received an

antibiotic within 4 hours. Sixty percent of the patients

had post-chemotherapy associated neutropenia, and

nearly half received granulocyte colony stimulating factor

(G-CSF) aer developing febrile neutropenia. As to the

complications, acute respiratory failure developed in 64

cases (21.1%), while another 60 (19.8%) had clinical of

shock.

Of the 303 febrile neutropenic patients, 228 (75.2%)

survived and were discharged from the hospital. By the

day of febrile neutropenia, Fiy-three patients (70%

of dead cases) died during neutropenic time, before their

neutrophil recovering. e common causes of death were

severe infections (83.5%) and bleeding associated with

thrombocytopenia (12.7%). In the mortality group, the

median survival during admission was 6 days (IQR: 2–9

days), as shown in Table 2.

Risk factors associated with mortality outcome

We classied the patients into 2 groups to determine

predictive factors that might impact on the survival

outcomes during 30 days of admission. In a univariate

analysis, a patient age ≤ 60 years had a 30-day survival

rate of 77.3%, while the corresponding gure for those

aged > 60 years was 69.6% (p = .175; Fig 1A). e 30-

day survival rate in patients with ANC ≤ 100 cells/cu

mm was not significantly different from those with

ANC > 100 cells/cu mm (71.9% versus 76.2%, p = .421).

e 30-day survival rate for patients receiving G-CSF

was 84.9%, compared with 65.6% for those who did

not receive G-CSF (p = < .001; Fig 1B). Furthermore,

the post-chemotherapy-associated neutropenia group

had 30-day survival rate of 83.2%, which was superior

to the non-chemotherapy-related group (60.5%; p =

< .001; Fig 2A). For MASCC score, the lower score is

associated with higher risk of complications of febrile

neutropenia. A MASCC score ≤ 16 yielded a 30-day

survival rate of 58.2% whereas a MASCC score > 16

had a 30-day survival rate of 96.7% (p = < .001; Fig 2B).

Patients who were admitted to a private room must

be clinically stable and needed to copay for their stay.

ese group was found to have a 30-day survival rate

of 88.4%, which was moderately higher than the 72.0%

for those admitted to a general ward (p = .034; Fig 3A).

In addition, patients who received antibiotics < 4 hours

and ≥ 4 hours aer arrived at ward had 30-day survival

rate of 73.7% and 78%, respectively (p = .481; Fig 3B).

Focusing on the duration of neutropenia in each group,

the groups with aged > 60 years, ANC > 100 cells/cu mm,

receiving G-CSF and receiving antibiotics < 4 hours had

signicantly shorter neutropenic duration (Table 3).

In a multivariate analysis, the results showed that an

MASCC score ≤ 16 and the non-chemotherapy-related

group were associated with an increased risk of mortality

(HR = 16.88, 95% CI = 6.15–46.32, p = < .001; and HR =

2.37, 95% CI = 1.19-4.72, p = .014, respectively). Table 3

exhibits the univariate and multivariate analyses of the

risk factors impacting on the survival outcomes of febrile

neutropenic patients during 30 days of admission.

Subgroup analysis

We selected 277 patients who were categorized as a

high-risk group by MASCC score to analyze the survival

outcome. e patients who received G-CSF, or developed

febrile neutropenia aer chemotherapy, or had MASCC

score > 16 were more likely to have signicantly better

survival outcome in a univariate analysis. In a multivariate

analysis, the patients with MASCC score > 16 and those

who developed febrile neutropenia aer chemotherapy

were correlated with better outcome (Supplementary

Table S1).

Furthermore, subgroup analysis of the patients

who developed febrile neutropenia aer chemotherapy

revealed that MASCC score ≤ 16 leaded to an increased

risk of mortality when compared to those with MASCC

score > 16 (HR = 58.14, 95% CI = 3.05–1,108.10,

p = .007).

DISCUSSION

is study aimed to identify the risk factors associated

with a fatality outcome for febrile neutropenic patients

during a 30-day admission period. Approximately 70%

of the patients were cancer cases. Almost all cases had a

clinical severity by MASCC risk score below 21, which

is considered to signify a high risk.

In a univariate

analysis, favorable factors of survival outcome consist of

chemotherapy-associated neutropenia, G-CSF exposure,

admitted to private room, and MASCC score > 16. Post-

chemotherapy-associated neutropenia patients had more

favorable outcomes. is would be because these patients

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

343

TABLE 2. e infections, complications and outcomes of febrile neutropenic patients.

Febrile neutropenic patients

(n = 303)

Positive results (%)

Infections

Site of infection 57.8%

Respiratory 22.8%

Gastrointestinal 16.8%

Genitourinary 11.2%

Bacteremia 19.1%

Skin and soft tissue 8.9%

Head and neck 3%

Central nervous system 0.7%

Infectious organisms 37.3%

Bacteria

Acinetobacter spp. 9.2%

Escherichia coli 8.3%

Pseudomonas spp. 5.3%

Klebsiella spp. 5.0%

Streptococcus spp. 4.6%

Staphylococcus spp. 3.3%

Enterococcus spp. 2.3%

Multidrug-resistant bacteria 14.5%

Fungus

Penicillium spp. 2.3%

Aspergillus spp. 1.7%

Candida spp. 1.0%

Complications

Respiratory failure 21.1%

Shock 19.8%

Median duration of ANC < 500 cells/cu mm (days) 3 (2-7)

Median length of hospital stays (days) 6 (4-11)

Outcomes on 30

day

Death 75 (25.6%)

Median time to death (days) 6 (2-9)

Cause of death

Infection 83.5%

Bleeding 12.7%

Abbreviation: ANC = absolute neutrophil count

Rotchanapanya et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

344

Original Article

SMJ

Fig 1. Overall survival curves according to (A) age groups (p-value = .175); (B) receiving G-CSF groups (p-value < .001).

Fig 2. Overall survival curves according to (A) chemotherapy related groups (p-value <.001); (B) MASCC score groups (p-value < .001).

Fig 3. Overall survival curves according to (A) admitted ward-type groups (p-value = .034); (B) time to receive antibiotics (p-value =.481).

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

345

TABLE 3. Univariate and multivariate analyses for overall mortality of febrile neutropenic patients (days).

Univariate N Median neutropenic Overall P-value HR (95% CI)

duration survival (days)

(days, IQR)

Age .175

 ≤60years 181 4(2–10) 24.8  1

> 60 years 112 2 (1–4) 23.4 1.37 (0.87–2.16)

ANC

 ≤100cells/cumm 121 4(2–8) 23.9 .421 1

> 100 cells/cu mm 172 2 (1–6) 25.4 0.83 (0.52–1.31)

G-CSF < .001

Yes 133 2 (2–4) 26.7 1

No 160 5 (2–11) 22.2 2.56 (1.54–4.28)

Chemotherapy < .001

Yes 179 2 (2–7) 26.4 1

No 114 4 (2–7) 20.9 2.74 (1.72–4.35)

MASCC score < .001

> 16 123 3 (2–6) 29.6 1

 ≤16 170 3(2–9) 20.4  16.42(5.99–44.99)

Private room .034

Yes 43 4 (2–10) 27.6 1

No 250 3 (2–6) 23.7 2.67 (1.08–6.61)

Time to antibiotic (hours) .481

< 4 243 3 (2–6) 25.6 1

 ≥4 50 6(2–15) 24.0  1.26(0.66–2.39)

A p-value < .05 indicates statistical signicance

Abbreviations: ANC = absolute neutrophil count; CI = condence interval; G-CSF = granulocyte colony stimulating factor; HR = Hazard

ratio; IQR = interquartile range, MASCC = the Multinational Association for Supportive Care in Cancer

Multivariate P-value HR (95% CI)

No G-CSF .501 1.30 (0 .61–2.78)

No chemotherapy .014 2.37 (1.19–4.72)

MASCC≤16 < .001 16.88 (6.15–46.32)

No private room .210 1.80 (0.72–4.49)

Rotchanapanya et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

346

Original Article

SMJ

SUPPLEMENTARY TABLE S1. Univariate and multivariate analyses for overall mortality of high-risk febrile

neutropenic patients (days).

Univariate Overall survival (days) P-value HR (95% CI)

Age .396

 ≤60years 24.2  1

> 60 years 23.3 1.22 (0.77–1.94)

ANC

 ≤100cells/cumm 23.1 .531 1

> 100 cells/cu mm 24.3 0.86 (0.54–1.37)

G-CSF < .001

Yes 26.4 1

No 21.8 2.57 (1.53–4.31)

Chemotherapy < .001

Yes 26.1 1

No 20.1 2.86 (1.79–4.57)

MASCC score < .001

> 16 29.8 1

 ≤16 20.4  43.87(6.09–315.96)

Private room .086

Yes 27.7 1

No 23.2 2.42 (0.88–6.65)

Time to antibiotic (hours) .859

< 4 23.6 1

 ≥4 24.9  1.06(0.55–2.05)

A p-value < .05 indicates statistical signicance

Abbreviations: ANC = absolute neutrophil count; CI = condence interval; G-CSF = granulocyte colony stimulating factor; HR = Hazard

ratio; MASCC = the Multinational Association for Supportive Care in Cancer

Multivariate P-value HR (95% CI)

No G-CSF .438 1.36 (0 .63–2.96)

No chemotherapy .022 2.28 (1.13–4.62)

MASCC≤16 < .001 44.07 (6.09–318.79)

No private room .716 1.21 (0.43–3.41)

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

347

could spontaneously achieve neutrophil count recovery

aer their nadir period. Furthermore, a previous study

has conrmed that the use of G-CSF is benecial to

short neutropenic duration in solid cancer patients with

high-risk febrile neutropenia receiving chemotherapy.

Aplastic anemia and infection were common causes in

non-chemotherapy related neutropenic group. A previous

large cohort study of aplastic anemia in developing

country represented the mortality rate of 9 percents.

In addition, increase severity of neutropenia had been

reported in advanced HIV infection and leaded to fatal

secondary infections.

G-CSF may be advantageous to

temporary increase of neutrophil counts in patients with

severe aplastic anemia with fever.

In addition, patients

who admitted to private room had better outcomes

because this group was possibly correlated with clinically

stable, higher socioeconomic status and better nursing

care. As to the factors associated with poor outcomes

in the multivariate analysis, an MASCC score ≤ 16 and

non-chemotherapy-related neutropenia were susceptible

to a higher mortality risk. We decided to use MASCC

score ≤ 16 and > 16 to divide patients into two groups

because 16 is the median score of this cohort and this

level was the beginning of the score that can discriminate

the outcome between alive and dead. is result was

correlated with a large observational study carried out by

MASCC score at the beginning of febrile neutropenia in

relation to bacteremia, the score <15 represented much

higher mortality of 36% compared with 3% in patients

with MASCC score ≥21.

Taking all into account, we

encourage physicians to determine the MASCC score

of febrile neutropenia patients early. For patients with

lower MASCC scores, especially ≤ 16, or patients who

have non-chemotherapy-related neutropenia, extensive

investigations, intensive vital sign monitoring, and promptly

antibiotics and specic treatments are needed to improve

their survival outcomes.

More than 25% of the cases in this cohort died during

admission, which is a higher rate than the previously

reported results in a developed country.

We particularly

focused on the survival outcomes on the 30

day of

admission because the previous study had shown that

the mortality rate ratios were highest within 1 month

for patients treated with chemotherapy for malignant

diseases.

rombocytopenic bleeding is listed as the

second most common cause of death. Because there are

some problems of platelet’s availability and utilization of

platelets depending on each physician’s decision making.

e median duration of an ANC below 500 cells/

cu mm was 3 days. is short duration for neutropenic

recovery was probably related to the fact that 60.1% of the

cases were post-chemotherapy-associated neutropenia,

who typically would be able to restore neutrophils to

normal levels within a few days.

However, many patients

in this study died during neutropenic period so we

cannot demonstrate the real neutrophil recovery time

in this group.

e three most common sources of infection in

our study were respiratory tract infections, bacteremia,

and gastrointestinal tract infections. Similar to previous

results, pneumonia and gastrointestinal infections were

the important causes of morbidity and mortality for

febrile neutropenic patients.

is study found that the

most common infectious organisms were gram negative

bacteria, with 14.5% being multidrug-resistant organisms.

However, fungal infections were only present in a small

number of the patients in our study. is is because of

the limited availability of computerized tomography of

the lungs and the long period for serum galactomannan

results. ese result in an under-detection of fungal

infections, especially invasive pulmonary aspergillosis.

e main limitation of our study is that it was a retrospective

study; some data, therefore, are lacking or limited. For

instance, the exact duration of febrile neutropenia could

not be estimated because some patients had history of fever

before admitted to the hospital. e complications of this

study were mainly focused on acute respiratory failure

and shock; nevertheless, other complications might be

co-occurred but did not be recorded. Moreover, several

primary physicians took care of the febrile neutropenic

patients followed an old specic hospital protocol to

prescribe antibiotics within 4 hours aer making diagnosis.

However, current guidelines recommend prescribing

antibiotics within an hour. e indications of G-CSF

for febrile neutropenia in our hospital during the study

period were vary in each physicians. Fortunately, a policy

for G-CSF usage in febrile neutropenia in our hospital is

recently applied. Further investigation about the outcome

aer releasing the new policy is ongoing collected.

CONCLUSION

e 30-day survival rate of febrile neutropenic

patients in ailand is seventy-ve percents. A low MASCC

score and non-chemotherapy-related neutropenia are

associated with a higher risk of unfavorable outcomes.

Declaration of conicting interest: e authors have

declared that no conicts of interest exist.

REFERENCES

1. Talcott JA, Siegel RD, Finberg R, Goldman L. Risk assessment

in cancer patients with fever and neutropenia: a prospective,

Rotchanapanya et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

348

Original Article

SMJ

two-center validation of a prediction rule. J Clin Oncol

1992;10:316-22.

2. Taplitz RA, Kennedy EB, Bow EJ, Crews J, Gleason C, Hawley

DK, et al. Outpatient Management of Fever and Neutropenia

in Adults Treated for Malignancy: American Society of Clinical

Oncology and Infectious Diseases Society of America Clinical

Practice Guideline Update. J Clin Oncol 2018;36:1443-53.

3. National Comprehensive Cancer Network (NCCN), Prevention

and Treatment of Cancer-Related Infections (version 1.2019),

Available at: https://www.nccn.org/professionals/physician_gls/

pdf/infections.pdf, Accessed 1 November 2018.

4. Osmani AH, Jabbar AA, Gangwani MK, Hassan B. Outcomes

of High Risk Patients with Febrile Neutropenia at a Tertiary

Care Center. Asian Pac J Cancer Prev 2017;18: 2741-5.

5. Rapoport BL, Aapro M, Paesmans M, van Eeden R, Smit T,

Krendyukov A, et al. Febrile neutropenia (FN) occurrence

outside of clinical trials: occurrence and predictive factors in

adult patients treated with chemotherapy and an expected

moderate FN risk. Rationale and design of a real-world prospective,

observational, multinational study. BMC Cancer 2018;18:917.

6. Shi X, Sims MD, Hanna MM, Xie M, Gulick PG, Zheng YH,

et al. Neutropenia during HIV infection: adverse consequences

and remedies. Int Rev Immunol 2014;33:511-36.

7. Quarello P, Saracco P, Giacchino M, Caselli D, Caviglia I,

Longoni D, et al. Epidemiology of infections in children with

acquired aplastic anaemia: a retrospective multicenter study

in Italy. Eur J Haematol 2012;88:526-34.

8. Newburger PE, Dale DC. Evaluation and management of

patients with isolated neutropenia. Semin Hematol 2013;50:

198-206.

9. Capsoni F, Sarzi-Puttini P, Zanella A. Primary and secondary

autoimmune neutropenia. Arthritis Res er 2005;7:208-14.

10. Kuderer NM, Dale DC, Crawford J, Cosler LE, Lyman GH.

Mortality, morbidity, and cost associated with febrile neutropenia

in adult cancer patients. Cancer 2006;106: 2258-66.

11. Dulisse B, Li X, Gayle JA, Barron RL, Ernst FR, Rothman KJ,

et al. A retrospective study of the clinical and economic burden

during hospitalizations among cancer patients with febrile

neutropenia. J Med Econ 2013;16:720-35.

12. Freifeld AG, Bow EJ, Sepkowitz KA, Boeckh MJ, Ito JI, Mullen

CA, et al. Clinical practice guideline for the use of antimicrobial

agents in neutropenic patients with cancer: 2010 update by

the infectious diseases society of america. Clin Infect Dis 2011;

52:e56-e93.

13. Klastersky J, Paesmans M. e Multinational Association

for Supportive Care in Cancer (MASCC) risk index score:

10 years of use for identifying low-risk febrile neutropenic

cancer patients. Support Care Cancer 2013;21:1487-95.

14. Magiorakos AP,Srinivasan A,Carey RB, Carmeli Y, Falagas

ME, Giske CG, et al. Multidrug-resistant, extensively drug-

resistant and pandrug-resistant bacteria: an international expert

proposal for interim standarddenitionsfor acquired resistance.

Clin Microbiol Infect2012;18:268-81.

15. Osmani AH, Jabbar AA, Gangwani MK, Hassan B. Outcomes

of high risk patients with febrile neutropenia at a tertiary care

center. Asian Pac J Cancer Prev 2017;18:2741-5.

16. Klastersky J, Paesmans M, Rubenstein EB, Boyer M, Elting L, Feld

R, et al. e Multinational Association for Supportive Care in

Cancer risk index: A multinational scoring system for identifying

low-risk febrile neutropenic cancer patients. J Clin Oncol

2000;18:3038-51.

17. García-Carbonero R,Mayordomo JI,Tornamira MV, López-

Brea M, Rueda A, Guillem V, et al. Granulocyte colony-

stimulating factor in the treatment of high-riskfebrile neutropenia:

a multicenter randomized trial. J Natl Cancer Inst2001;93:

31-38.

18. Höchsmann B,Moicean A,Risitano A,Ljungman P. Supportive

care in severe and very severe aplastic anemia. Bone Marrow

Transplant2013;48:168-73.

19. Klastersky J, Ameye L, Maertens, Georgala A, Muanza F, Aoun

M, et al. Bacteraemia in febrile neutropenic cancer patients.

Int J Antimicrob Agents 2007;30 Suppl 1:S51-S59.

20. Nordvig J, Aagaard T, Daugaard G, Brown P, Sengeløv H,

Lundgren J, et al. Febrile Neutropenia and Long-term Risk of

Infection Among Patients Treated With Chemotherapy for

Malignant Diseases. Open Forum Infect Dis 2018;5:ofy255.

21. Chan A, Wong QX, Ali MK, Wong M, Hsu LY. Clinical ecacy

of adjunctive G-CSF on solid tumor and lymphoma patients with

established febrile neutropenia. Support Care Cancer 2014;22:

1105-12.

22. Calik S, Ari A, Bilgir O, Cetintepe T, Yis R, Sonmez U, et al.

e relationship between mortality and microbiological parameters

in febrile neutropenic patients with hematological malignancies.

Saudi Med J 2018;39:878-85.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

349

atchanan Manopetkasem, Ph.D.*, Suree Kanjanawong, Ph.D.**, Phaisarn Jantarungsri, Ph.D.**

*Sirindhorn School of Prosthetics and Orthotics, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, **Christian University, Bangkok,

ailand.

The Components of Strategic Leadership of

Prosthetic and Orthotic Practitioners in Thailand

ABSTRACT

Objectives: To determine the components and to test the validity of factor analysis of strategic leadership among

prosthetic and orthotic practitioners in ailand with the empirical data.

Methods: e sample consisted of 754 prosthetic and orthotic practitioners from 200 government hospitals in

ailand. e two instruments used were structured interview and questionnaires consisting of 66 items with Item-

Objective Congruence (IOC) at 0.96 and the reliability of questionnaire in the form of Cronbach alpha coecient

was 0.98. e content analysis, exploratory factor analysis and conrmatory factor analysis were used to analyze data.

Results: ere were 7 components of strategic leadership among prosthetic and orthotic practitioners in ailand

which consisted of 42 factors with factor loadings as follows: relating the part to the whole (0.997), corporate spirit

(0.904), strategic thinking and planning (0.899), building partnerships (0.896), developing today’ and tomorrow’s

leader (0.878), making it happen (0.865), and giving direction (0.823). e validity test results of factor analysis of

strategic leadership among prosthetic and orthotic practitioners in ailand were conrmed with the empirical data.

Conclusion: From the research ndings, 7 components of strategic leadership should be developed among prosthetic

and orthotic practitioners, particulary the rst three components i.e., relating the part to the whole, corporate spirit,

and strategic thinking and planning.

Keywords: Strategic leadership; prosthetic and orthotic practitioners (Siriraj Med J 2019; 71: 349-355)

Corresponding author: atchanan Manopetkasem

E-mail: thatchanan.mao@mahidol.edu

Received 29 November 2017 Revised 6 January 2019 Accepted 14 February 2019

ORCID ID: http://orcid.org/0000-0003-2021-3129

http://dx.doi.org/10.33192/Smj.2019.53

INTRODUCTION

e WHO states that an estimated 0.5 percent of

people with disabilities require a prosthesis or orthosis.

Furthermore, in 2010, among people in Africa, Asia,

and Latin America, 30 million people with disabilities

required a prosthesis or orthosis and were in need of

rehabilitation practitioners; Physiatrist, Certied Prosthetist

and Orthotist, and Prosthetist and Orthotist Technician.

A Certied Prosthetist and Orthotist could treat 250

patients which means a requirement of approximately

120,000 practitioners. However there are only 34 certied

institutes that can produce 600-800 practitioners annually.

e production of Prosthetists and Orthotists is still a

global issue which requires huge capital for machinery and

adequate teaching personnel. is results in a substantial lack

of Prosthetists and Orthotists practitioners in developing

countries.

In ailand, there are 836,224 people with

disabilities in need of prosthetic and orthotic devices

as of December 2016.

According to the WHO, a single

device has a 3-year lifespan, meaning that ailand

Manopetkasem et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

350

Original Article

SMJ

requires 1,150 Prosthetists and Orthotists. Physiatrists

have specic training in rehabilitation through prosthetics

and orthotics. Currently there are only 381 physiatrists

working in the ai healthcare scheme,

and there are

only 173 licensed Prosthetists and Orthotists.

Nevertheless, the Sirindhorn School of Prosthetics

and Orthotics, established in 2002, is an ISPO Category

1 program which has produced 100 qualified Thai

Prosthetists and Orthotists until now. However, despite

100 graduates, only 71 are working in ailand.

Strategic

management planning is essential in order to progress

human resource development. John Adair is an expert in

‘Strategic Leadership’ and states that leadership comprises

strategic thinking and planning, giving direction, relating

the part to the whole, corporate spirit, developing today’s

and tomorrow’s leaders, building partnerships, and making

it happen.

Strategic Leadership is a characteristic of a

leader who brings about advancement to an organization

and begins with a good vision and successful practices.

Senior management or manager is required to directly

involve with every step of strategic management which

consists of 3 main parts; 1) Strategic Formulation, 2)

Strategic Implementation, and 3) Strategic Control &

Evaluation.

A review of literature did not provide any examples

of research on strategic leadership among prosthetic and

orthotic practitioners in ailand and International level.

ere is literature on leadership in nursing

and some

literature focuses on a model for strategic leadership

development for secondary school administrators according

to Durbin’s theory.

7-9

is information is useful because

it lays the groundwork for preliminary research on the

prosthetic and orthotic profession.

is study aimed to identify the components of

strategic leadership among prosthetic and orthotic

practitioners in ailand. e outcome of this research

will be useful for the development of strategic leadership

among prosthetic and orthotic practitioners as key succes

factors of eective health service management. According

to the new way of management, not only the top level

is important to run the business but also middle level

and frontline level. To develop strategic leadership for

prosthetic and orthotic in all levels will help increase an

eective health service management as a whole.

Research question

1) What are the components of the strategic leadership

among prosthetic and orthotic practitioners in ailand?

and 2) Does the component structure of the strategic

leadership among prosthetic and orthotic practitioners

in ailand conform with empirical data?

Objective

1) Analyze the components of strategic leadership among

prosthetic and orthotic practitioners in ailand.

2) Analyze variables of each component of strategic

leadership among prosthetic and orthotic practitioners

in ailand.

3) Analyze the conformity between the components

of strategic leadership among prosthetic and orthotic

practitioners in ailand with the empirical data.

Conceptual Framework

A conceptual framework, which was created based on

the strategic leadership framework of John Adair (2010)

was used.

e framework demonstrates responsibilities

of the Chief Executive Ocer within 7 components;

Giving direction, Making it happen, Relating the part

to the whole, Building partnerships, Strategic thinking

and planning, Releasing corporate spirit, and Developing

today’s and tomorrow’s leader (Fig 1).

Fig 1. Conceptual framework of the components of the strategic

leadership among prosthetic and orthotic practitioners in ailand.

MATERIALS AND METHODS

From May to October 2016, questionnaires were

provided to 754 prosthetics and orthotics practitioners

from 200 government institutes in ailand via post.

e research was a correlational descriptive research and

utilized a questionnaire of 76 questions divided into 3

parts;

Part 1: Personal Information (10 questions)

Part 2: Questionnaire of the 7 components of strategic

leadership among prosthetic and orthotic practitioners

in ailand: Giving direction (7 questions), Make it

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

351

happen (7 questions), Relating the part to the whole (15

questions), Building partnerships (6 questions), Releasing

corporate spirit (6 questions), Developing today’s and

tomorrow’s leader (10 questions), and Strategic thinking

and planning (14 questions)

Part 3: Comments and suggestions (1 question)

The questionnaire was adopted from “The Competencies

of Strategic Leadership of Chief Nurse Executives, Private

Hospital in Bangkok Metropolitan

which was based on

Adair’s concept (2010)

. It comprised of 7 components

76 items of strategic leadership. e researcher contacted

the copyright holder and asked the permission to utilize

the questionnaire which has already been translated into

ai language. At the rst stage, small group discussion

among 20 experts in prosthetic and orthotic, as well

as human resources management was set up in order

to gather comments and ideas related to the strategic

leadership among prosthetic and orthotic practitioners

in ailand. e second stage, in-depth interview with

8 prosthetic and orthotic experts was arranged in order

to verify the variables. Finally, the questionnaire was

modied based on small group discussion and in-depth

interview ndings. Eleven items were deleted because

they did not t for the samples in this research study.

e original questionnaire was used for the top level,

but this research concerned not only the top level but

the middle and the frontline level managers. Evaluated

with the Index of Item-Objective Congruence (IOC),

the content validity of the questionnaire was 0.96 and

the reliability demonstrated in the form of Cronbach’s

Alpha Coecient was 0.98.

Participants

e participants were prosthetic and orthotic executives

and practitioners in government medical institutes,

secondary and tertiary hospitals all over the country

consisting of 200 management executives in hospitals,

381 physiatrists, 73 prosthetic and orthotic practitioners,

and 100 prosthetic and orthotic technicians, so in total

754 persons participated in the study e prosthetic and

orthotic executives and practitioners in private sector were

not included because the administrative system between

public and private medical institues was dierent.

Samples

e samples were physiatrists, prosthetic and orthotic

practitioners, prosthetic and orthotic technicians, and

hospital management executives. A questionnaire was

used to rst discuss 76 variables through a small group (20

persons) discussion and in-depth interview of 8 prosthetic

and orthotic management executives and experts. A

resulting 65 variables were used in this research because

contents of variables use were for senior chief exceutive,

while the prosthetic and orthotic practitioners were junior

chief exceutives. As there were 754 participants, all of them

were considered samples

and 427 questionnaires were

answered on rst request (56.63%). Hair et al (2010) and

Comrey and Lee (1992) suggested use of factor analysis

and determined that a sample should be 10-20 times of

variables (650-1,300 sample of 65 variables) and 300 was

sucient for this research using AMOS program.

12-14

Ethic approval

Ethical approval was approved by Christian University

of ailand Institutional Review Board (Protocol number

4/2015), Faculty of Medicine Siriraj Hospital, Mahidol

University (Si 695/2016), Chulalongkorn University

(accceptance letter no. 1955/2016) and other institutes

as required.

Outcome measurement

e questionnaire was adapted from a strategic

leadership capacity questionnaire provided to administrators

of nursing departments at private hospitals in Bangkok and

surrounding areas,

following concepts of John Adair.

e questionnaire was already translated into ai and

used in a small group discussion among 20 experts in the

Prosthetics and Orthotics eld in this research. Secondly,

in-depth interviews with 8 prosthetic and orthotic experts

were performed to verify nal variables by proposive

method (experts, lecturers and executive of human

resource department in prosthetics and orthotics eld).

e questionnaire was then evaluated with the Index of

Item-Objective Congruence (IOC) which scored 0.96.

irty questionnaires were provided to prosthetic and

orthotic practitioners and human resource management

executives which were similar to original research samples

for reliability test by Cronbach’s alpha coecient. Also,

since the questionnaire was a 5-level measurement, a

Cronbach’s alpha coecient was applied and resulted

in a value of 0.98.

Statistical analysis

Final data analysis was performed using descriptive

statistics; frequency, percentage, mean, mode, standard

deviation, and inferential statistics; Exploratory Factor

Analysis and Conrm Factor Analysis with AMOS version

no.18.

RESULTS

Personal information of the participants veried that

the proportion of male and female participants were

Manopetkasem et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

352

Original Article

SMJ

almost equal. Most were 31-40 years old and married

(33.5%). ey possessed more than 15 years of experience

in prosthetic and orthotic practice (37%), had Bachelor’s

Degrees in Health Science (48.9%), and worked in hospitals

under the Ministry of Public Health (57.6%).

Variable analysis of each component of strategic leadership

among prosthetic and orthotic practitioners in ailand

provided 7 components in ranking order as Table 1.

ere were 7 Factor Loadings which ranged from 0.823

to 0.997; the highest was Relating the part to the whole

at 0.997 while the lowest was Giving Direction at 0.823.

Result from an analysis of each component of the

strategic leadership among prosthetic and orthotic

practitioners in ailand

1) Relating the part to the whole (RL) consisted

of 8 variables which were (1) the leader has to clearly

understand the objective of the department, (2) the

leader has to clearly understand the objective of aliated

organization or unit, (3) all management executives in the

department are capable of working as a team in order to

create a High Performance Team, (4) the leader is able to

get all members to work together to reach objectives of

the organization, (5) the leader is able to get all members

in the department to work together and reach objectives,

(6) the leader should possess a capability to manage

personal and work life for better leadership eciency,

(7) the leader should create a work life balance, and (8)

creating too many changes may result in confusion and

disorganization among practitioners.

2) Realizing corporate spirit (SP) consisted of 5

variables which were; (1) the leader has to be able to

draw out eciencies or strengths in practitioners upon

their work, (2) practitioners expect their leader to have

strategic leadership and teamwork ability, (3) the leader

is required to trust in practitioners until their actions

are not trustworthy, (4) the human mind is a powerful

tool which can inspire practitioners to do great things,

and (5) the nature of practitioners are normally good

as a human beings.

3) Strategic thinking and planning (ST) consisted of

7 variables which were; (1) the leader ought to regularly

listen to the prosthetic and orthotic management team

regarding their strategic and innovative thinking, (2) the

leader’s strategic planning of the department should be

exible and up-to-date, (3) the leader ought to regularly

listen to the hospital executive or management team

upon their strategic and innovative thinking, (4) the

leader should improve the strategic management of the

department in order to provide services at ease, (5) the

leader should encourage providing rewards for creative

innovation, (6) the leader should create opportunity

for practitioners to take part in developing innovation

within the department, and (7) creative innovation can

encourage achievement in terms of personnel, team, and

device productions.

4) Building partnerships (PN) consisted of 5 variables

which were; (1) other than income from service provisions,

practitioners are required to be aware of benets for the

community, (2) the leader should take part in assisting

the department and should get along well with domestic

TABLE 1. Variables analysis of each component of strategic leadership among prosthetic and orthotic practitioners

in ailand.

No Components Factor loadings

1. Relating the part to the whole (RL) .997

2. Releasing Corporate Spirit (SP) .904

3. Strategic Thinking and Planning (ST) .899

4. Building Partnerships (PN) .896

5. Developing Today’ and Tomorrow’s Leader (DV) .878

6. Making it happen (HP) .865

7. Giving Direction (DR) .823

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

353

alliances outside the hospital, (3) the device service

provision, employment policy, and environmental impacts

are corporate social responsibilities of prosthetic and

orthotic department, (4) the leader should take part in

assisting the department to get along well with international

alliances of dierent cultures, and (5) the department

will be more successful if the leader expands to have

more alliances in terms of device service provision.

5) Developing today’ and tomorrow’s leader (DV)

consisted of 7 variables which were; when a practitioner is

appointed to be a management executive, the leader will

also have to (1) evaluate the team management of each

executive, (2) evaluate the management competency of the

assigned works, (3) evaluate the leadership regularly, (4)

evaluate the characteristic of a leader, (5) the leader should

initiate a career development system for practitioners to

be at a management executive level in the future, (6) the

leader should make the management executives believe

that they are able to train leadership to practitioners, and

(7) the leader should launch a policy for the management

executives to be responsible for self-development.

6) Making it happen (HP) consisted of 5 variables

which were; the leader has to administrate all segments in

the department to focus on the customers (1) by initiating

quality control system of prosthetic and orthotic service

provision, (2) by initiating a quality control system of

service provision, (3) by initiating a security system

of service provision, (4) the leader should take part in

rectifying the problem in case the management executives

under the department failed to achieve their goals, and

(5) the leader should always follow up progression and

review the strategic planning of the department.

7) Giving direction (DR) consisted of 5 variables

which were; (1) the leader has to create a mutual vision

in order to make the department a future organization,

(2) the leader can clearly explain the core value of the

responsible department, (3) practitioners have to clearly

understand their roles which conform to the direction of

organization, and their own responsibility, (4) the leader

can clearly explain the core values which indicate the

direction of the organization, and (5) the leader has to

clearly understand the main objectives of the department

or the organization, and know what to do in the moment

and in the future.

Result from the analysis of the congruence of the

component of the strategic leadership among prosthetic

and orthotic practitioners in ailand with the empirical

data

Aer adjusting the relationship model through a

Conrmatory Factor Analysis, the CMIN/Df = 1.590,

Root Mean Squared Residual (RMR) = 0.18, Root Mean

Squared Error of Approximate Residual (RAMSEA) =

.037, Tucker Lewis Index (TLI) = .968, and Comparative

Fit Index (CFI)=.975, were found to conform to the

empirical data by consistent tolerance index. erefore,

the model was consistent with the empirical data on the

components as illustrated in Table 2 and Fig 2.

TABLE 2. Model of a Conrmatory Factor Analysis of the component of the strategic leadership among prosthetic

and orthotic practitioners in ailand with the empirical data.

Consistency Index Tolerance Index Adjusted Model

Statistical Index Consistency

χ(FD/NIMC) fd/

Less than 3 1.590 Consistent

Root Mean Squared Residual (RMR) Not more than 0.8 .018 Consistent

Goodness of Fit Index (GFI) More than 0.9 .900 Consistent

Adjust Goodness of Fit Index (AGFI) More than 0.9 .863 Consistent

Root Mean Square Effort of Appr.(RMSEA) Less than 0.08 .037 Consistent

Tucker Lewis Index (TLI) More than 0.9 .968 Consistent

Comparative Fit Index (CFI) More than 0.9 .975 Consistent

Manopetkasem et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

354

Original Article

SMJ

DISCUSSION

e dierence between the survey of prosthetic and

orthotic practitioners in ailand and the Conrmatory

Factor Analysis (CFA) of the component of the strategic

leadership among prosthetic and orthotic practitioners

in ailand

e survey outcome on the components of strategic

leadership among practitioners in ailand demonstrated

giving precedence to the component consecutively as

follows; giving direction (Х= 4.56, SD = 0.58), make it

happen (Х = 4.53, SD = 0.59), strategic thinking and

planning (Х = 4.46, SD = 0.62), relating the part to the

whole (Х = 4.43, SD = 0.63), develop today’s and tomorrow’s

leader (Х = 4.32, SD = 0.67), building partnerships (Х

= 4.31, SD = 0.68), and releasing corporate spirit (Х =

4.14, SD = 0.74). On the other hand, the outcome of CFA

turned out to fall in an alternative order of precedence;

relating the part to the whole (factor loading = 0.997),

releasing corporate spirit (factor loading = 0.904), strategic

thinking and planning (factor loading = 0.899), building

partnerships (factor loading = 0.896), develop today’s

and tomorrow’s leader (factor loading = 0.878), make

it happen (factor loading = 0.865), and giving direction

(factor loading = 0.823).

Such dierent outcomes of precedence levels from the

research were due to the fact that practitioners considered

the components of strategic leadership separately and

gave precedence to external environment which evidences

an increase of people with disabilities that augments

the demand of practitioners and treatment.

Still, the

CFA helped to determine whether the variable was a

reasonable and systematic construct in the theoretical

model or not.

According to the CFA, each variable has

an eect on each other and concerns a possible deviation

on the overall model.

e Conrmatory Factor Analysis of the component of

the strategic leadership among prosthetic and orthotic

practitioners in Thailand demonstrated that all 7

components were consistent with the empirical data.

e research outcome rearmed the component of

strategic leadership of John Adair (2010) responsibility for

Chief Executive Ocer. All 7 components conformed with

the research outcome of strategic leadership capability of

the chief executive of the nursing department at hospitals

in Bangkok and surrounding areas.

is research outcome rearmed that a strategic

leader is required to predict future directions of the

organization and initiate strategic planning to pursue

the objectives.

is research outcome demonstrated

that the most signicant components relate the part

to the whole, releasing corporate spirit, and strategic

thinking and planning. Whereas the research outcome

observed in nursing management demonstrated that the

most signicant components were strategic thinking and

planning, relating the part to the whole, and developing

Fig 2. Structural pattern of the component of the strategic leadership among prosthetic and orthotic practitioners in ailand

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

355

today’s and tomorrow’s leader.

is similarity is as a result

of practitioners being medical personnel who provide

assessment, design, invent, and produce prostheses and

orthoses for people with disabilities,

while nurses provide

hospitalized care to people, families, and communities

by instructing, suggesting, consulting, giving health

counsel, physical and mental contact, environmental care,

rehabilitating, provision of primary care, and assisting

the specialist in treating diseases.

By doing so, both

careers are connected by a medical profession which

shares a similar responsibility to take care of people with

disabilities.

e component of strategic leadership among prosthetic

and orthotic practitioners in terms of relating the part

to the whole is a rst priority.

Regarding to a process of providing medical

care, prosthetics and orthotic devices to the patient

requires communication, coorperation, and assistance

from various practitioners; prosthetic and orthotic

practitioners, technicians, orthopaedists, physiatrists,

surgeon, physiologists, nurses, mechanical engineers, and

rehabilitation sta. Relating the part to the whole, thus,

is the most important component for the strategic leader

to understand about the objectives and working system

of their department and organization. Also, individuals

will have to work as a team to in order to create a chief

executive team that is highly ecient.

Recommendations to apply the research outcome in

the organization administration

1.1 is research discovered that relating the part

to the whole is the most signicant component, and

therefore the organization must pay close attention in

order to enhance the strategic leadership of practitioners,

especially in their understanding of objectives of the

organization. Being consistent with the objectives of the

aliated organization and team work is compulsory, and

will lead to an eective chief executive and team leader.

1.2 e providing direction is the least signicant

component, but the most agreed component by practitioners.

erefore, the organization should support and improve

by creating shared vision in order to move forward to

a future organization, and increase a clear core value

picture in responsibility of departments. Moreover, a

clear understanding of the role, direction of organization,

and personal responsibilities is also essential.

Recommendation for future research

1. Creating guidelines for strategic leadership

of practitioners in ailand and facilitating progress

by initiating training centered on strategic leadership

development programs.

2. Future research should consider related components

of strategic leadership with operational performance of

practitioners.

3. Creating research of strategic leadership in

other medical elds such as Physiotherapists, and ai

Traditional Medicne, etcetera.

REFERENCES

1. World Health Organization: World Health Report on Disabilities

[Internet]. 2016 [cited 2017 September 30]. Available from:

http://apps.who.int/iris/bitstream/10665/70670/1/WHO_

NMH_VIP_11.01_eng.pdf

2. Department of Empowerment of Person with Disabilities: People

with Disabilities Situation [Internet]. 2016 [cited 2017 September

30]. Available from: http://dep.go.th/sites/default/les/les/

news/ailan_PWD_situation_dec2016.pdf

3. Opartkiattikul N, editor. Minute of meeting of Sirindhorn

School of Prosthetics and Orthotics; Bangkok, Thailand:

2015.p.1-30.

4. e Royal College of Physiatrist of ailand: list of physiatrists

in ailand [Internet]. 2016 [cited 2017 September 30]. Available

from: http://rehabmed.or.th/main/nd-a-doctor/

5. Wongkhomthong J, Srithumma N, Jintanadilok N. The

Competencies of Strategic Leadership of Chief Nurse Executives

Private Hospital in Bangkok Metropolitan. CUTJ 2014;20:353-9.

6. Yawiraj N. Leadership and Strategic Leader. 6

Edition.

Bangkok: Central express, 2007.

7. Pongsriwat S. Contexts for Transformational and

Transactional Leadership. 2005;7:127-30.

8. Jitsanguan K. A model of strategic leadership development

for secondary school administrators. EAU Heritage Journal

Social Science and Humanity 2014;4: 201-11.

9. Janthapoon P, Sumettikul P. Development of Collaborative

Network Management Strategies for Small Sized School

Development. EDUCU 2014;42:81-94.

10. Wajanawongpanya P, Wisetsiri P, Sinlarat P. Strategies for the

Development of ai-Style Leaders’ Characteristics of Administrators

in Basic Education Institution. EAU Heritage Journal Social

Science and Humanity 2014;3:208-21.

11. Adair J. Strategic Leadership: How to think and plan strategically

and provide direction. London: Kogan Page; 2010.

12. Comrey AL, Howard BL. The Factor Analytic Model. A

First Course in Factor Analysis. 2

edition. Hillsdale, New

Jersey: Lawrence Erlbaum Associate Inc.; 1992.p.21-52.

13. Hair JF Jr, Black WC, Babin BJ, Anderson RE, Tatham RL.

Multivariate data analysis. 6

ed. New Jersey: Prentice Hall;

2006.

14. Silpjaru T. Research and Data Analysis (SPSS and AMOS).

ed. Bangkok: Business R&D; 2012.

15. Opartkiattikul N. A decade of Sirindhorn School of Prosthetics

and Orthotics (2002-1012). 1

ed. Bangkok: P.A. Living Co.,

Ltd; 2012.p.1-66.

16. Nursing Council of ailand. Scope of Professional Practice in

Nursing and Midwifery. [Internet]. 2016 [cited 2017 September

30]. Available from: http://nursing-unit.pn.psu.ac.th/data/sar/

organization/le_up/5_nurse.pdf

Manopetkasem et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

356

Original Article

SMJ

Worawan Jittham, M.D.

Department of Pediatric, Faculty of Medicine, Naresuan University, Phitsanulok 65000, ailand.

Correlation between Heart Failure with N-Terminal-

Pro-B Type Natriuretic Peptide and Cardiothoracic

Ratio from Chest Radiograph in Pediatric Heart

Disease

ABSTRACT

Objective: The aim of this study is to determine the correlation between the symptom of heart failure and basic investigations; N-Terminal-

Pro-B Type Natriuretic Peptide (NT-proBNP) and cardiothoracic (CT) ratio from chest radiograph in pediatric heart disease.

Methods: One-hundred-eighty children (aged 1-15 years) with underlying heart disease were enrolled

in this prospective cross-sectional study. e heart failures were categorized based on the Ross classication

into 2 groups, non-heart failure (Ross classication I) and heart failure (Ross classication II-IV). e NT‐

proBNP level was determined and chest radiograph was done in posteroanterior upright position for CT ratio.

Results: e mean NT-proBNP level was 223.1 pg/ml (±180.3) and a mean CT ratio was 53.6% (±5.6) in the non-heart

failure group. e mean NT-proBNP level was 1,054 pg/ml (±1,840.3) and a mean CT ratio was 58.6% (±6.1) in heart failure

group. ere was a signicantly positive correlation between heart failure symptoms and the level of NT-proBNP and

CT ratio. In this study, the cut-o value of NT-proBNP for heart failure was more than 400 pg/ml (OR 6.97, ROC 0.694,

sensitivity 52.6%, specicity 86.3%) and CT ratio more than 55% (OR 3.57, ROC 0.654, sensitivity 68%, specicity 62.8%).

Conclusion: In pediatric heart diseases, there are strong positive correlations between heart failure with both

NT-proBNP and CT ratio. ese correlations help in the diagnosis of heart failure. NT-proBNP level more than

400 pg/ml and CT ratio more than 55% are indicative of heart failure in our population. Both investigations are

inexpensive, readily available and do not require specialist experts.

Keywords: NT-proBNP; cardiothoracic ratio; heart failure (Siriraj Med J 2019; 71: 356-363)

Corresponding author: Worawan Jittham

E-mail: ann_pitlok@yahoo.com

Received 23 April 2019 Revised 13 August 2019 Accepted 16 August 2019

ORCID ID: http://orcid.org/0000-0002-2227-3136

http://dx.doi.org/10.33192/Smj.2019.54

INTRODUCTION

Pediatric heart diseases are more oen than not

congenital and less frequently acquired disease. Heart

failure is one of the most important problems in pediatric

cardiology clinic. e etiologies of heart failure from

congenital heart diseases are mainly from volume over

load and/or pressure over load. A normally-structured

heart in which the disease le ventricle function in

opposite to its normal function is most common cause

of heart failure.

e clinical signs of heart failure in

children include poor feeding, respiratory distress, poor

weight gain, edema and exercise intolerance. ere are

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

357

standard guidelines for diagnosis and treatment of heart

failure for adult in worldwide practice.

2-4

In children, the

practice guidelines reviewed by many studies showed no

uniformed recommendation for diagnosis and treatment

of heart failure due to variant pathophysiology in pediatric

heart failure and small studies subjects.

5-7

ese studies

suggested that we should combine symptoms, physical

examinations, investigations for diagnosis and treatment

and reliant on the pathophysiology of heart failures in

individual children.

In out-patient department, the investigations which

most readily available in every hospital and useful for

diagnosis of heart failures in children are chest radiography,

electrocardiogram, blood test for neuro-hormonal activation;

brain natriuretic peptide (BNP) and N-Terminal-Pro-B

Type Natriuretic Peptide (NT-proBNP). NT-proBNP and

cardiothoracic ratio from chest radiograph are widely

used with a standard cut-o value for heart failure in

adult, but there is no consensus for children. Contributing

factors are the various types of ventricular impairment,

underlying cardiac morphology, age, gender, and assay

method for NT-proBNP aecting the reference values for

both markers. For chest radiograph one usually calculates

the CT ratio for determination of cardiomegaly in the

diagnosis of heart failure. e aim of this study was to

determine the correlation between symptoms and signs

of heart failure with basic investigations; NT-proBNP

and CT ratio from chest radiograph in pediatric heart

diseases.

MATERIALS AND METHODS

We performed a prospective cross-sectional study

on 180 children (aged 1-15 years) with underlying heart

diseases with diagnosis and follow-ups at the cardiology

clinic of Department of Pediatric, Faculty of Medicine,

Naresuan University, ailand from September 2015

to May 2018. e study was approved by the Ethics

Committee of Naresuan University (COA No. 208/2015).

Heart failure was categorized by a pediatric cardiologist

based on Ross classification

during follow up in

cardiology clinic. e patients were categorized into

2 groups: non-heart failure (Ross classication I) and

heart failure (Ross classication II-IV). e baseline

characteristics: age, gender, underlying heart diseases,

current medication and body mass index (BMI) were

recorded.

e NT‐ProBNP level was assayed with 3 ml of

heparinized blood and immunoassay by Roche cobas

h 232 instruments (Roche Diagnostics Ltd. CH-6343,

Rotkreuz, Switzerland). e normal adult range of this

assay is 0-300 pg/ml, and a value of more than 450 pg/

ml is suggestive of heart failure in adult with age below

50 years.

3,4

Presently, there are no clear standards and

there is no consensus of normal range for children.

e chest radiograph was done in posteroanterior

upright position for CT ratio. Patients who could not have

chest radiographs done in upright position were excluded

from the study. e CT ratio is the ratio of maximal

horizontal cardiac diameter to maximal horizontal thoracic

diameter (inner edge of ribs) in chest radiography.

Prior to performing the procedures, all patients and

their parents (or families) were briefed on the risks and

benets of the procedure. Only upon consent from the

patients and their families were the procedure carried

out.

Sample size

e proportion of NT–pro-BNP in patients with

heart failure in Evelyn Lechner et al showed the NT–

pro-BNP in patients with heart failure to be 0.39 and

patients without heart failure was 0.6; 2, Alpha of 0.05; 3

and Beta of 0.2 for testing two independent proportions

(two-tailed test).

e total sample size of 176 patients.

Statistical analysis

Continuous data with normal distribution were

shown as mean and standard deviation (SD) and those

with non-normal distribution as median and minimum:

maximum. Categorical data was shown as frequency and

percentage. e odds ratio and 95% condence interval

(95%CI) were used to compare the association between

NT–pro-BNP and CT ratio presence of heart failure. A

receiver operating characteristic (ROC) analysis was

used to determine the diagnostic utility of NT-proBNP

and CT ratio for heart failure. A p-value of less than 0.05

was regarded statistically signicant.

RESULTS

e study consisted of 180 patients, 91 males (50.6%)

and 89 females (49.4%) patients. In all the subjects, the

mean age was 6.3 years (±4.3 years), the mean weight

was 21.7 kg (±12 kg), the mean height was 112.9 cm

(±26.7 cm) and the mean BMI was 15.6 (±2.9). e

diagnosis of heart disease was categorized into 3 major

hemodynamics groups; ventricular volume load (62.2%),

ventricular pressure load (22.2%) and combine ventricular

load (15.6%). e diagnosis in ventricular volume load

group was (1) le to right shunt lesion: ventricular septal

defect (VSD), atrial septal defect (ASD), patent ductus

arteriosus (PDA) (2) valvular regurgitation: mitral valve

regurgitation (MR), aortic valve regurgitation (AR),

Ebstein’s anomaly with tricuspid valve regurgitation

Jittham et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

358

Original Article

SMJ

(TR), rheumatic heart disease with valvular regurgitation

(3) postoperative total correction of tetralogy of Fallot

(TOF) with pulmonary valve regurgitation (PR). e

diagnosis in ventricular pressure load group was (1)

valvular stenosis: pulmonary valve stenosis (PS), aortic

valve stenosis (AS), corrected translocation of great artery

(TGA) with severe PS, VSD with severe PS (2) TOF with

moderate to severe PS (3) pulmonary hypertension (PHT)

or Eisenmenger’s complex: complete atrioventricular

canal defect (CAVD) with PHT, Double Outlet Right

Ventricle (DORV) with PHT (4) coarctation of aorta

(COA). e diagnosis in combine ventricular load group

was single ventricle (SV) with postoperative Glenn’s

operation, SV with postoperative aortopulmonary shunt

or pulmonary artery banding, pulmonary atresia (PA)

with VSD postoperative aortopulmonary shunt. For

all patients, the mean NT-proBNP was 583.1 pg/ml

(±1,282.9) with median of 225 pg/ml (19:9,000) and a

mean of CT ratio was 55.7% (±6.3) with median of 56%

(40.3:84) (Table 1). e Fig 1 and 2 are the distribution

of NT-proBNP level and CT ratio of all patients in each

Ross classication group.

TABLE 1. Clinical characteristics, NT-proBNP and CT ratio of patients in heart failure and non-heart failure group.

Characteristic All patients heart failure Non-heart failure

(n=180) (n=78) (n=102) P-value

n % n % n %

Sex (male) 91 50.6 35 44.9 56 54.9 0.182

Age (year) 0.199

1-5 94 52.2 45 57.7 49 48.0

6-15 86 47.8 33 42.3 53 52.0

Mean (±S.D.) 6.3 (±4.3) 5.9 (±4.5) 6.6 (±4.1)

Weight (kg) 0.074

Mean (±S.D.) 21.7 (±12) 19.9 (±12) 23.1 (±11.8)

Height (cm) 0.006

Mean (±S.D.) 112.9 (±26.7) 106.7 (±27.8) 117.7 (±24.9)

BMI (kg/m

) 0.963

Mean (±S.D.) 15.6 (±2.9) 15.6 (±3.3) 15.6 (±2.6)

Diagnosis <0.001

Ventricular volume load 112 62.2 33 42.3 79 77.5

Ventricular pressure load 40 22.2 17 21.8 23 22.5

Combine ventricular load 28 15.6 28 35.9 0 0.0

NT-proBNP (pg/ml)

Mean (±S.D.) 583.1 (±1282.9) 1054 (±1840.3) 223.1 (±180.3)

Median (min : max) 225 (19 : 9000) 406.5 (19 : 9000) 171 (44 : 1080)

CT ratio

Mean (±S.D.) 55.7 (±6.3) 58.6 (±6.1) 53.6 (±5.6)

Median (min : max) 56 (40.3 : 84) 57.9 (40.3 : 84) 53.9 (42.3 : 68.8)

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

359

Non-heart failure group

e non-heart failure group (n=102, 56.7%), dened

as asymptomatic patients with Ross classication class I,

included 56 males (54.9%) and 46 females (45.1%). e

mean age was 6.6 years (±4.1 years), the mean weight

was 23.1 kg (±11.8 kg), the mean height was 117.7 cm

(±24.9 cm) and the mean BMI was15.6 (±2.6). The

diagnosis in this group revealed that the patients had

mainly acyanotic heart disease (99%) which consisted

of 77.5% of ventricular volume load 22.5% of ventricular

pressure load and no diagnosis of combine ventricular

load. e mean NT-proBNP was 223.1 pg/ml (±180.3)

with median of 171 pg/ml (44:1080) and a mean of CT

ratio was 53.6% (±5.6) with median of 53.9% (43.3:68.8)

(Table 1).

Heart failure group

e heart failure group (n=78, 43.3%), dened as

symptomatic patients with Ross classication class II-IV,

included 35 males (44.9%) and 43 females (55.1%). e

majority of Ross classication in this group was class II

(88%) e mean age was 5.9 years (±4.5 years), the mean

weight was 19.9 kg (±12 kg), the mean height was 106.7

cm (±27.8 cm) and the mean BMI was 15.6 (±3.3). e

diagnosis in this group was acyanotic heart disease in

49%, which consisted of 42.3% of ventricular volume load,

21.8% of ventricular pressure load and 35.9% of combine

ventricular load. e mean NT-proBNP was 1,054 pg/ml

(±1,840.3) with median of 406.5 pg/ml (19:9,000) and a

mean CT ratio was 58.6% (±6.1) with median of 57.9%

(40.3:84) (Table 1). ere were signicant dierences

between heart failure and non-heart failure group in

both NT-proBNP level and CT ratio (p<0.05) (Figs 3-4).

e NT-proBNP level base on hemodynamic groups

e mean NT-proBNP in ventricular volume load

group was 920.9 pg/ml (±1,772.7) in the heart failure

group and 213.6 pg/ml (±182) in the non-heart failure

Fig 1. Scatter plot of NT-proBNP level grouped by Ross classication

in all patients.

Fig 2. Scatter plot of CT ratio grouped by Ross classication in all

patients.

Fig 3. e box-plot of NT-proBNP level in patients with clinical

heart failure and non-heart failure.

Fig 4. e box-plot of CT ratio in patients with clinical heart failure

and non-heart failure.

Jittham et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

360

Original Article

SMJ

group. e mean of NT-proBNP in ventricular pressure

load group was 1381 pg/ml (±2656.3) in heart failure

group and 255.6 pg/ml (±174.3) in non-heart failure

group. e mean NT-proBNP in combine ventricular

load group was 1012.3 pg/ml (±1,289.3) in heart failure

group (Table 2).

Correlation between Heart failure and NT-proBNP

The correlation of NT-proBNP between the 2

subgroups of symptom (non-heart failure and heart

failure group) showed a signicant positive correlation

with area under the curve (AUC) of receiver operative

characteristic (ROC) curve of 0.697 (Fig 5). e level

of NT-proBNP showed an increase correlation with

a higher level indicating greater possibilities for heart

failure. e cut-o value of NT-proBNP from our study

for diagnosis of heart failure in pediatric heart disease

population was more than 400 pg/ml with sensitivity at

52.6%, specicity at 86.3% (Table 3).

Correlation between Heart failure and CT ratio

e correlation of CT ratio with 2 subgroups of

symptom (non-heart failure and heart failure group)

showed a signicant positive correlation with AUC of

ROC curve 0.728 (Fig 5). e CT ratio showed increase

correlation when higher level of CT ratio indicated higher

odd for heart failure. e appropriate cut-o value from

our study for diagnosis heart failure in pediatric heart

disease was a CT ratio more than 55% with sensitivity

at 68%, specicity at 62.8% (Table 4).

DISCUSSION

e brain natriuretic peptide (BNP) and NT-proBNP

are synthesized by ventricular myocytes in response to

the pressure load, the volume load or an increase in

myocardial wall stress. Both hormones are the most

well studied biomarker for diagnosis of heart failure in

adult and children. In this study, we found a signicant

and strong positive correlation between NT-proBNP

TABLE 2. e NT-proBNP level in hemodynamic subgroup.

All patients Heart failure Non-heart failure

Diagnosis (n=180) (n=78) (n=102)

NT-proBNP (pg/ml) NT-proBNP (pg/ml) NT-proBNP (pg/ml)

Mean (±S.D.) Mean (±S.D.) Mean (±S.D.)

Ventricular volume load 422 1016.9 920.9 1772.7 213.6 182

Ventricular pressure load 733.9 1797 1381 2656.3 255.6 174.3

Combine ventricular load 1012.3 1289.3 1012.3 1289.3 NA* NA*

*no patient in this subgroup

TABLE 3. e correlation of clinical of patients with NT-proBNP.

Outcome NT-proBNP (pg/ml)

>300 >350 >400

Odds ratio (95% CI) 4.43 (2.34 to 8.38) 6.14 (3.09 to 12.20) 6.97 (3.40 to 14.28)

ROC 0.672 0.692 0.694

Sensitivity (%) 59.0 55.1 52.6

Specicity(%) 75.5 83.3 86.3

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

361

and CT ratio with clinical of heart failure in children

with underlying heart disease. ese results are similar

to the previous studies showing a correlation between

NT-proBNP and symptoms and signs of pediatric heart

failure dened with New York Heart Association (NYHA),

Ross and modied Ross classication scores.

9-11

In non-heart failure group, the mean NT-proBNP

level (mean 223.1±180.3 pg/ml, median 171 pg/ml) in

this study was almost within the normal range in healthy

children, when matched with age and sex in A CALIPER

pilot study. e reference value of immunoassay for NT-

proBNP level using the same kit as ours from A CALIPER

pilot study showed the level in boys and girls age 1 to

5 years to be 12.5-308.4 pg/ml and those aged 5 to 15

years to be 7.9-177.9 pg/ml.

Another study by A Nir,

et al which studied the reference value of NT-proBNP

using a similar method in 58 normal children (4 months

to 15 years) showed normal 5-391 pg/ml, median 90

pg/ml.

In this study, in hemodynamic asymptomatic

children, there was no dierent in the mean of NT-

proBNP as compare to the previous normal reference

value. is result suggested that even with underlying

heart disease, the NT-proBNP would exhibit similar

level. ese children had minimal ventricular load and

wall stress. Our result was dierent from A Uner and

colleagues study on NT-proBNP level in symptomatic,

asymptomatic heart disease and control normal children.

It showed signicant dierent level of NT-proBNP among

the 3 groups.

is observation may be due to the larger

sample size of pediatric heart disease in ours study.

Similar to the previous studies, in heart failure

group, the mean NT-proBNP was signicantly higher

than asymptomatic patient. e previous studies showed

higher NT-proBNP level in children with heart failure

compared to normal children.

10,13-14

F-J Zhou, et al found

that the children with congenital heart disease admitted

to hospital with pneumonia and heart failure had higher

NT-proBNP level than those with pneumonia alone.

We also observed that ventricular wall stress might be

higher in the pressure load group than volume load or

combined ventricular load based on NT-proBNP level

(Table 2). e combined ventricular load group consisted

mainly of patients with cyanotic complex congenital

heart disease with all subjects having symptomatic heart

failure with elevation of NT-proBNP values and higher

than symptomatic ventricular volume load group. e

systematic review in usefulness of brain natriuretic

peptide in complex congenital heart disease by Jannet

A Eindhoven, et al showed result to be the same as our

study.

Evelyn Lechner, et al were studied in single

ventricle heart disease and the results showed that the RV

morphology yielded a higher NT-proBNP than the LV

morphology. It is thus concluded that the morphologic

right ventricle might be more stressed as a systemic

ventricle than the morphologic le ventricle.

TABLE 4. e correlation of clinical of patients with CT ratio.

Outcome CT ratio (%)

>50 >55 >60

Odds ratio (95% CI) 8.46 (2.84 to 25.14) 3.57 (1.92 to 6.65) 4.87 (2.18 to 10.86)

ROC 0.631 0.654 0.624

Sensitivity (%) 94.9 68.0 34.6

Specicity(%) 31.4 62.8 90.2

Fig 5. e ROC curves of NT-proBNP and CT ratio for the diagnosis

of heart failure in pediatric heart disease population.

Jittham et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

362

Original Article

SMJ

e cut-o value of NT-proBNP for diagnosis of

heart failure in children was not well established. e

NT-proBNP cut o value in acute heart failure for adult

aged below 50 years yielded a gure of more than 450

pg/ml with sensitivity at 98% and specicity at 76% form

PRIDE study.

In children, a previous study by Masaya

Sugimoto, et al showed cut-o value in symptomatic

heart failure for patients with heart disease aged below

3 years of more than 438 pg/ml with sensitivity at 88.7%,

specicity at 91.8%. In patient aged more than 3 years,

it was more than 295.2 pg/ml with sensitivity at 94.7%,

specicity at 95.9%.

e other study by Murat Sahin

showed that the cut-o value for diagnosis ventricular

dysfunction in children between 4 months to 17 years

old was more than 514 pg/ml with sensitivity at 95%,

specicity at 80%.

In our study done with the same

analysis of NT-proBNP showed a nearly cut o value as

Masaya Sugimoto study in which NT-proBNP of more

than 400 pg/ml had sensitivity at 52.6% and specicity

at 86.3% for diagnosis of heart failure in children aged

more than 1 year old.

e other commonly used investigation for diagnosis

heart failure is chest radiograph. Although echocardiogram,

cardiac catheterization, cardiac MRI and multi-detector

CT are now commonly used imaging techniques to

investigate cardiac disease, the chest radiograph is still

necessary for rst step of investigation. To evaluate of

cardiac enlargement, the CT ratio from chest radiograph

has long been accepted as the method of choice.

e

accepted upper limit of normal CT ratio is 50% in children

and adult and 60% in infant.

e diagnostic accuracyfrom

CT ratio for evaluation of function both LV and RV from

many studies showed low accuracy.

21-22

In LV systolic

function showed no signicant association with CT ratio

in patients with preserved le ventricular ejection fraction

who had undergone computed tomography coronary

angiography. But the CT ratio correlated with LV size and

LV ejection fraction (LVEF) in patients with depressed

LVEF.

In another study, e CT ratio reected atrial

dilatation rather than ventricular dilatation.

Our study

showed a signicant positive correlation of CT ratio with

symptomatic heart failure in pediatric heart disease.

e appropriate cut-o value of CT ratio was more

than 55% (OR 3.57; 1.92-6.65, ROC 0.654, sensitivity

at 68%, specicity at 62.8%) for the diagnosis of heart

failure in pediatric heart disease. is result was similar

to the study of Konstantinos Dimopoulos, et al, in which

cardiomegaly was present in both simple and complex

adult with congenital heart disease and CT ratio more

than 55% corrected with the patients’ symptom and

predicted an 8-fold increased risk of death.

CONCLUSION

e NT-proBNP level and CT ratio are helped in

the diagnosis of heart failure in pediatric heart disease.

NT-proBNP level more than 400 pg/ml and CT ratio more

than 55% were predictive of heart failure in our population.

Both investigations are inexpensive, readily available

and do not require specialist experts for interpretation.

REFERENCES

1. Hsu DT, Pearson GD. Heart Failure in Children Part I: History,

Etiology, and Pathophysiology. Circ Heart Fail 2009;2:63-70.

2. Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE, Drazner

MH, et al. 2013 ACCF/AHA Guideline for the Management

of Heart failure. Circulation 2013;128:e240-e327.

3. Yancy CW,Jessup M,Bozkurt B,Butler J,CaseyDE,Colvin

MM,et al. 2017 ACC/AHA/HFSA Focused Update of the

2013 ACCF/AHA Guideline for the Management of Heart

Failure A Report of the American College of Cardiology/American

Heart Association Task Force on Clinical Practice Guidelines

and the Heart Failure Society of America. J Card Fail 2017;23:

628-51.

4. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JG,

Coats AJ, et al. 2016 ESC Guidelines for the diagnosis and

treatment of acute and chronic heart failure: e Task Force

for the diagnosis and treatment of acute and chronic heart

failure of the European Society of Cardiology (ESC). Developed

with the special contribution of the Heart Failure Association

(HFA) of the ESC. Eur J Heart Fail 2016;18:891-975.

5. Rosenthal D, Chrisant MRK, Edens E, Mahony L, Canter

C, Colan S, et al. International Society for Heart and Lung

Transplantation: Practice Guidelines for Management of

Heart Failure in Children. J Heart Lung Transplant 2004;23:

1313-33.

6. Hsu DT, Pearson GD. Heart Failure in Children Part II: Diagnosis,

Treatment, and Future Directions. Circ Heart Fail 2009;2:

490-8.

7. Masarone D, Valente F, Rubino M, Vastarella R, Gravino R,

Rea A, et al. Pediatric Heart Failure: A Practical Guide to

Diagnosis and Management. Pediatr Neonatol 2017;58:303-12.

8. Ross RD, Bollinger RO, Pinsky WW. Grading of the Severity of

Congestive Heart Failure in Infant. Peddiatr Cardiol 1992;13:

72-5.

9. Lechner E, Schreier-Lechner EM, Hofer A, Gitter R, Mair R, Biebl A,

et al. Aminoterminal brain-type natriuretic peptide levels correlate

with heart failure in patients with bidirectional Glenn anastomosis

and with morbidity aer the Fontan operation. J orac Cardiovasc

Surg 2009;138:560-4.

10. Mir TS, Marohn S, Läer S, Eiselt M, Grollmus O, Weil J. Plasma

Concentrations of N-Terminal Pro-Brain Natriuretic Peptide

in Control Children From the Neonatal to Adolescent Period and

in Children With Congestive Heart Failure. Pediatrics 2002;110:

e76.

11. Şahin M,Portakal O,Karagöz T,Hasçelik G,Özkutlu S.

Diagnostic performance of BNP and NT-ProBNP measurements

in children with heart failure based on congenital heart defects

and cardiomyopathies. Clin Biochem2010;43:1278-81.

12. Kulasingam V, Jung BP, Blasutig IM, Baradaran S, Chan MK,

Aytekin M, et al. Pediatric reference intervals for 28 chemistries

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

363

and immunoassays on the Roche cobas 6000 analyzer--a

CALIPER pilot study. Clin Biochem2010;43:1045-50.

13. Nir A, Bar-Oz B, Perles Z, Brooks R, Korach A, Rein AJJT.

N-terminal pro-B-type natriuretic peptide: reference plasma

levels from birth to adolescence. Elevated levels at birth and in

infants and children with heart diseases. Acta Paediatr2004;93:

603-7.

14. Uner A, Dogan M, Ay M, Acar C. e evaluation of serum

N-terminal prohormone brain-type natriuretic peptide,

troponin-I, and high-sensitivity C-reactive protein levels

in children with congenital heart disease. Hum and Exp Toxicol

2014;33:1158-66.

15. Zhou FJ, Zhou CY, Tian YJ, Xiao AJ, Li PI, Wang YH, et al.

Diagnostic value of analysis of H-FABP, NT-proBNP, and

cTnI in heart function in children with congenital heart disease

and pneumonia. Eur Rev Med Pharmacol Sci 2014;18:1513-16.

16. Eindhoven JA, Van den Bosch AE, Jansen PR, Boersma E, Roos-

Hesselink JW. e usefulness of brain natriuretic peptide in

complex congenital heart disease: a systematic review. J AM Coll

Cardiol 2012;60:2140-9.

17. Januzzi JL Jr, Camargo CA, Anwaruddin S, Baggish AL, Chen

AA, Krauser DG, et al. e N-terminal Pro-BNP Investigation

of Dyspnea in the Emergency department (PRIDE) study. Am

J Cardiol 2005;95:948-54.

18. Sugimoto M, Manabe H, Nakau K, Furuya A, Okushima K,

Fujiyasu H, et al. eroleof N-terminal pro-B-typenatriuretic

peptidein the diagnosis ofcongestive heart failureinchildren.

- Correlation with theheart failurescore and comparison with

B-typenatriuretic peptide. Circ J 2010;74:998-1005.

19. Danzer CS. e cardiothoracic ratio. Am J Med Sci 1919;157:513-

54.

20. Gleeson FV, Rayner AC. e chest radiograph in heart disease.

Medicine 2010;38:357-61.

21. Zhu Y, Xu H, Zhu X, Wei Y, Yang G, Xu Y, et al. Association between

cardiothoracic ratio, le ventricular size and systolic function

in patients undergoing computed tomography coronary

angiography. Exp er Med 2014;8:1757-63.

22. Spiewak M, Malek LA, Biernacka EK, Kowalski M, Michalowska

I, Homan P, et al. Cardiothoracic ratio may be misleading

in the assessment of right- and le-ventricular size in patients

with repaired tetralogy of Fallot. Clin Radiol 2014;69:e1-8.

23. Dimopoulos K, Giannakoulas G, Bendayan I, Liodakis E, Petraco

R, Diller GP, et al. Cardiothoracic ratio from postero-anterior

chest radiographs: A simple, reproducible and independent

marker of disease severity and outcome in adults with congenital

heart disease. Int J Cardiol 2013;166:453-57.

Jittham et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

364

Original Article

SMJ

Ranida amphithak, M.D., Pitcha Panawattanawong, M.D., Sutasinee Boonsopon, M.D., Nattaporn Tesavibul,

M.D., Pitipol Choopong, M.D.

Department of Ophthalmology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.

Ocular Manifestations in Acute Herpes Zoster

Ophthalmicus

ABSTRACT

Objective: To determine the prevalence and clinical predictors of ocular involvement in acute herpes zoster ophthalmicus.

Methods: A retrospective chart review of 167 patients presenting with acute herpes zoster ophthalmicus at

the Outpatient Department of Ophthalmology, Siriraj Hospital, was performed. All skin lesions along the

ophthalmic branch of the trigeminal nerve (CN V

) and ocular inammatory signs were observed and documented.

Results: A total of 160 cases were analyzed. e prevalence of ocular inammation in acute herpes zoster ophthalmicus

was 73.1% (117/160). e types of ocular inammations included conjunctival injection (99.1%), keratitis (51.3%), and

iridocyclitis (21.4%). Nasociliary skin lesions (Hutchinson’s sign) were the best predictor of ocular inammation in acute

herpes zoster ophthalmicus (p = 0.006, OR = 3.6, 95% CI: 1.5-9.1). Other factors associated with ocular inammation

were a period of longer than 4 days from the onset of rash to an eye examination (p = 0.007, OR = 3.2, 95% CI: 1.4-7.5),

and the initiation of systemic acyclovir treatment aer 3 days from rash onset (p = 0.037, OR = 2.3, 95% CI: 1.05-4.96).

Conclusion: ere is a high prevalence of ocular inammation in acute herpes zoster ophthalmicus, especially

among individuals with Hutchinson’s sign and delayed systemic acyclovir treatment. General practitioners should

be aware of ocular involvement and refer high-risk patients for a complete ophthalmologic assessment.

Keywords: Herpes zoster ophthalmicus; Hutchinson’s sign; acyclovir; ocular inammation; uveitis (Siriraj Med J

2019; 71: 364-369)

Corresponding author: Pitipol Choopong

E-mail: pitipol.cho@mahidol.edu

Received 13 September 2018 Revised 22 February 2019 Accepted 15 May 2019

ORCID ID: http://orcid.org/0000-0001-8857-8488

http://dx.doi.org/10.33192/Smj.2019.55

INTRODUCTION

Herpes zoster ophthalmicus (HZO) is a reactivation

of the varicella-zoster virus (VZV), a DNA-virus in the

Herpesviridae family, along the ophthalmic division of

the trigeminal nerve (CN V

) aer a previous infection

such as chickenpox.

1-2

e clinical course usually starts

with prodromal symptoms like fever, malaise, headache,

and dysesthesia for approximately 1-4 days prior to the

development of skin lesions. e skin manifestations

are characterized by maculopapular rashes and vesicles

along the CN V

dermatome, including the forehead,

eyelids, and nose. e vesicles can progress to pustules

and occasionally hemorrhagic vesicles, with the formation

of crusts in 7-10 days.

e nasociliary nerve, a branch of the ophthalmic

nerve, supplies both the nasociliary dermatome and

intraocular tissue. e nasociliary dermatome extends

from the tip and sides of the nose to the nasal bridge

and medial canthus. e involvement of the nasociliary

nerve presents as a variety of vesicular eruptions at the

periorbital area, which can lead to a wide range of ocular

inammations (for instance, conjunctivitis, episcleritis,

scleritis, keratitis, uveitis, optic neuritis, and oculomotor

palsy). e subsequent ocular complications consist of

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

365

chronic uveitis, neuralgia, and neurotrophic keratopathy.

1,3,7-8

e intraocular involvement of HZO, especially uveitis

and optic neuritis, can lead to multimorbidity, including

blindness, if HZO is not recognized early or treatment

is delayed.

e involvement of this dermatome (Hutchinson’s

sign) is considered a highly sensitive sign for ocular

involvement in patients with HZO.

3-4

e immune status

of the patients is another factor that can dictate the

likelihood and types of ocular presentation of HZO.

5-7

To the best of our knowledge, there has been no study of

the prevalence and risk factors of ocular involvement in

acute HZO in ailand. erefore, this study collected

the clinical data of ai patients with acute HZO to

determine the prevalence of ocular involvement in HZO

and the predictors associated with ocular inammation.

MATERIALS AND METHODS

e study was approved by the Siriraj Institutional

Review Board (Si 179/2009). A retrospective chart review

of patients who were clinically diagnosed with HZO and

attended the Outpatient Department of Ophthalmology,

Siriraj Hospital, between 2005 and 2008 was performed.

All patients underwent an eye examination under a

slit-lamp biomicroscope within 1 month of the onset

of skin lesions. e degree of conjunctival injection

was referenced from the Cornea and Contact Lens

Research Unit grading scale (none, mild, moderate, and

severe).

However, the clinical grading of the intraocular

inammations was modied from the cell grading system

of the Standardization of Uveitis Nomenclature (SUN)

Working Group by using a 1 mm x 1 mm slit beam

under a slit-lamp examination (grade 0 = none; grades

0.5–1+ = mild; grade 2+ = moderate; and grades 3–4+

= severe).

Excluded were patients with previous ocular

diseases, ocular trauma, or ocular surgery, or patients

who had an eye examination later than 1 month aer

the onset of their skin lesions.

e diagnoses of HZO were based on the presence

of a unilateral group of vesicles on an erythematous

base and a painful skin rash within the ophthalmic

dermatome. The following data were collected and

recorded: demographic data (age, sex, and race), pre-

existing diseases (acquired immune deciency syndrome,

diabetes mellitus, tuberculosis, autoimmune diseases,

lymphoma, pregnancy, and others), and the date of onset

of the skin rash. e extent of herpes zoster skin lesions

was documented by marking anatomical drawings at the

eyelids, forehead, alar, and tip of the nose.

e patients were treated with systemic oral acyclovir

as soon as the diagnosis of HZO was established, and

the time interval between the onset of HZO and the

administration of the oral antiviral drug was recorded.

Oral acyclovir (800 mg, 5 times per day) was prescribed

for a period of 7–10 days, as per the guidelines of the

American Academy of Family Physicians.

e quantitative data were analyzed using the unpaired

t-test method. e qualitative data were analyzed with

the Chi-square and Fisher’s exact tests. Multivariate

analysis was performed with a backward stepwise

logistic regression method. e odds ratio (OR) with

95% condence interval (95% CI) and p-values were

calculated. Two-sided p-values ≤ 0.05 were considered

statistically signicant.

RESULTS

Having been diagnosed with HZO, 167 patients were

referred to the Ophthalmology Department to undergo

an eye examination. Of those, seven were excluded (two

due to an incomplete examination, one with a history

of ocular surgery, and four with incomplete records).

Data on the remaining 160 patients were summarized

and analyzed.

Table 1 shows the demographic data of the 160

patients, who comprised 103 females (64.4%) and 57

males (35.6%) with a mean age of 51.5 ± 18.8 years (range

5-88). e median age of our subjects was also 51 years

(interquartile range 39.0-65.7). ere were no dierences

between including or excluding patients with extreme

ages. e HIV patients tended to be younger than the

non-HIV patients (mean age 46.37 ± 17.63 years versus

40.32 ± 7.25 years, respectively). e study revealed 22

patients (33.8%) who were anti-HIV positive from the

65 patients who underwent an anti-HIV test; 23 patients

(14.4%) with hypertension; 11 patients (6.9%) with

diabetes mellitus; 8 patients (5.0%) with tuberculosis; 8

patients (5.0%) with immunodeciency; 4 patients (2.5%)

with lymphoma; and 3 patients (1.9%) with systemic

lupus erythematosus. e skin lesions were conned to

the forehead in 134 patients (83.8%), the eyelids in 138

patients (86.3%), and the tip of the nose (or Hutchinson’s

sign) in 54 patients (33.8%). e median time from the

appearance of the rash to acyclovir treatment was 3 days

(range 1-30 days), and from the rash appearance to eye

examination was 4 days (range 1-30 days).

Ocular involvement was found in 117 patients

(73.1%) with a female predilection (74 females and 43

males), but there was no dierence in laterality. Almost

all patients (116/117) had a conjunctival injection with

mild severity (70.1%). About half of the HZO patients

with ocular involvement demonstrated corneal lesions

that were dendritic, stromal keratitis, or both. Iridocyclitis

Thamphithak et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

366

Original Article

SMJ

Characteristics N (Total = 160)

Age (years) 51.5 ± 18.8

Gender: Female 103 (64.4%)

Male 57 (35.6%)

Underlying diseases

 Acquiredimmunodeciencysyndrome 22/65(33.8%)

Hypertension 23 (14.4%)

Diabetes mellitus 11 (6.9%)

Tuberculosis 8 (5.0%)

 Otherimmunodeciency 8(5.0%)

Lymphoma 4 (2.5%)

Systemic lupus erythematosus 3 (1.9%)

Skin involvements

Forehead 134 (83.8%)

Eyelids 138 (86.3%)

Tip or alar of nose (Hutchinson’s sign) 54 (33.8%)

Median duration (days)

Rash to acyclovir treatment (min, max) 3 (1, 30)

Rash to eye examination (min, max) 4 (1, 30)

TABLE 1. Demographic data of all 160 patients with acute herpes zoster ophthalmicus.

TABLE 2. Ocular findings in acute herpes zoster

ophthalmicus patients.

Characteristics N = 117

Laterality

Right eye 67 (57.3%)

Left eye 50 (42.7%)

Conjunctival injection

No 1 (0.9%)

Yes 116(99.1%)

Mild 82 (70.1%)

Moderate 32 (27.3%)

Severe 2 (1.7%)

Keratitis

No 57 (48.7%)

Yes 60 (51.3%)

Dendrite 47 (40.2%)

Stromal 7 (6.0%)

Combined 6 (5.1%)

Iridocyclitis

No 92 (78.6%)

Yes 25 (21.4%)

Mild 14 (12.0%)

Moderate 7 (6.0%)

Severe 4 (3.4%)

was the only intraocular involvement seen in this cohort,

accounting for 21.4% (25/117; Table 2).

A univariate analysis demonstrated that the prevalence

of ocular involvement increased signicantly if a patient

presented with Hutchinson’s sign (p = 0.008, OR = 3.5,

95% CI: 1.4-8.4) or an HIV infection (p = 0.024, OR =

9.1, 95% CI: 1.1-7.5); if the period from the onset of rash

to an eye examination was > 4 days (p = 0.009, OR =

3.1, 95% CI: 1.4-7.1); or if the period from the rash onset

to acyclovir treatment was > 3 days (p = 0.044, OR =

2.3, 95% CI: 1.08-4.9; Table 3). ere was no statistical

signicance for the age of onset, sex, or other pre-existing

diseases.

As to the multivariate analysis, HIV infection showed no

statistically signicant association with ocular involvement.

In contrast, the 3 remaining factors (Hutchinson’s sign,

a period from rash onset to eye examination > 4 days, and

a period from rash onset to acyclovir treatment > 3 days)

signicantly increased the risk of ocular complications

(Table 4).

A subgroup analysis for dierent types of ocular

involvement revealed associations between Hutchinson’s

sign and conjunctivitis (OR = 3.6, 95% CI: 1.48-8.76),

keratitis (OR = 1.97, 95% CI: 1.0-3.85), and anterior

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

367

TABLE 3. Univariate analysis of factors associated with ocular involvement in acute herpes zoster ophthalmicus.

Ocular No ocular Odds ratio

Factors involvement involvement (95%condence P-value

(N = 117) (N = 43) interval)

Age (years) 50.4 ± 18.6 54.5 ± 19.4 – 0.227

Gender: Female 74 (63.2%) 29 (67.4%) 1.3 (0.5–3.1) 0.76

Duration of rash to eye examination >4 days 53 (45.3%) 9 (20.9%) 3.1 (1.4–7.1) 0.009*

Duration of rash to acyclovir treatment >3 days 60/114 (52.6%) 13/40 (32.5%) 2.3 (1.08–4.9) 0.044*

Underlying diseases

HIV infection** 21/51 (41.2%) 1/14 (7.1%) 9.1 (1.1–75) 0.024*

Hypertension 15 (12.8%) 8 (18.6%) 0.6 (0.2–1.6) 0.503

Diabetes 9 (7.7%) 2 (4.7%) 1.7 (0.4–8.2) 0.729

Tuberculosis 7 (6.0%) 1 (2.3%) 2.7 (0.3–22.4) 0.683

 Immunodeciency 5(4.3%) 3(7.0%) 0.6(0.1–2.6) 0.444

Lymphoma 4 (3.4%) – – 0.575

Systemic lupus erythematosus 3 (2.6%) – – 0.564

Skin involvement

Forehead 99 (84.6%) 35 (81.4%) 1.3 (0.5–3.1) 0.804

Eyelid 103 (88.0%) 35 (81.4%) 1.7 (0.7–4.3) 0.411

Tip or alar of nose (Hutchinson’s sign) 47 (40.2%) 7 (16.3%) 3.5 (1.4–8.4) 0.008*

Unpaired t-test for quantitative data and Chi-square test, Fisher’s exact test for qualitative data

*Statistical signicance at p-value < 0.05, **Human immunodeciency virus infection

TABLE 4. Multivariate analysis of factors associated with ocular involvement in acute herpes zoster ophthalmicus.

Crude odds ratio Adjusted odds

Factors (95%condence P-value

ratio (95% P-value

  interval)  condenceinterval)

Hutchinson’s sign 3.5 (1.4-8.4) 0.008* 3.6 (1.5-9.1) 0.006*

Duration of rash to acyclovir treatment >3 days 2.3 (1.08-4.9) 0.044* 2.3 (1.05-4.96) 0.037*

Duration of rash to eye examination >4 days 3.1 (1.4-7.1) 0.009* 3.2 (1.4-7.5) 0.007*

HIV infection** 9.1 (1.1-75) 0.024* 6.3 (0.8-49.9) 0.084

Univariate analysis by Chi-square test

Multivariate analysis by backward stepwise logistic regression

*Statistical signicance at p-value < 0.05

**Human immunodeciency virus infection

uveitis (OR = 2.74, 95% CI: 1.13-6.62). Aer adjusting

for three factors (HIV status; a period from rash onset to

eye examination > 4 days; and a period from rash onset

to acyclovir treatment > 3 days), there were changes

in the associations between Hutchinson’s sign and

conjunctivitis (OR = 4.17, 95% CI: 1.58-10.97), keratitis

(OR = 2.18, 95% CI: 1.07-4.42), and uveitis (OR = 3.23,

95% CI: 1.28-8.18).

Thamphithak et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

368

Original Article

SMJ

DISCUSSION

HZO is a common disease, and it may present not

only as a skin manifestation but also involve ocular tissue.

e ocular manifestations of HZO are variable; the acute

eruptive phase can present as eyelid swelling, conjunctivitis,

punctate keratitis, and/or corneal pseudodendrites.

Aer the acute eruptive phase, the infection can cause

stromal keratitis, nummular keratitis, disciform keratitis,

neurotrophic keratopathy, and corneal mucous plaque.

Severe ocular involvement has been reported as serpiginous

keratitis, acute retinal necrosis, optic neuritis, orbital

myositis, and cranial nerve palsy, but especially CN III

and orbital apex syndrome, resulting in retinal detachment

or a loss of vision. e clinical ndings in the chronic

phase include persistent corneal epithelial defects and

postherpetic neuralgia.

8,11-13

In the current report, we studied the ocular involvement

of acute HZO. e age of onset of HZO for our cohort

was about 50 years. is was younger than the ages found

by Canadian and European studies, in which HZO was

reported to occur at an age of approximately 60,

3,4,14

but

it was much older than the age reported from Africa

(around 35 years).

15-16

is may suggest that the age of

HZO onset may have racial and/or geographic variations.

Furthermore, we also found females were more predominant

than males. In terms of the underlying diseases, previous

studies by Nassaji-Zavareh et al. and Kaiserman et al.

demonstrated a signicant correlation between herpes

zoster infection and diabetes mellitus, but not specic

to HZO.

17-18

In our study, the proportion of diabetes

mellitus patients was only 6.9% (11 patients), which is

low. erefore, we could not analyze the association

between diabetes mellitus and herpes zoster infection.

We found that the most common skin lesions in

HZO occurred on the eyelids, followed by the forehead

and, least commonly, the tip of the nose. It is known that

the virus replicates and migrates peripherally along the

sensory nerve; when it reaches the skin, it can penetrate

the epidermis and subsequently develop into skin lesions.

From previous studies, the most frequent nerve which the

VZV aects is the frontal branch of the trigeminal nerve,

which innervates the upper eyelid and forehead.

7,19-20

is explains why our study found those areas were the

most common sites of HZO involvement.

e rate of ocular involvement in acute HZO in

our study was 73.1%, which is comparable to the 74%

involvement reported by Adam et al.

On the other

hand, the rate is slightly higher than that found by Zaal

et al. (63%).

However, the rate of ocular involvement

in the HZO patients in the study by Zaal et al. was

lower than in the present study because Zaal et al. only

included immunocompetent patients, and they were also

referred to ophthalmologists earlier than in our study.

e vast majority of ocular involvement in the current

study was in the form of conjunctivitis (99.1%), which

tended to subside gradually only aer systemic oral

antiviral treatment; some of the patients needed further

attention. In addition, half of the patients (51.3%) with

eye complications demonstrated keratitis (dendritic,

stromal, or a combination), which needed topical antiviral

therapy. About one-h of the patients (21.4%) had

an intraocular involvement (iridocyclitis). As to other

studies, Szwto et al. also found that conjunctivitis was

the most common ocular manifestation.

Our univariate analysis established a signicant

correlation between HIV infection and ocular involvement

in HZO, but not for diabetes mellitus. Nevertheless,

HIV infection did not have a signicant association

with ocular involvement in HZO aer adjustment by

the multivariate analysis. is mainly resulted from the

low number of patients in the study who had undergone

HIV testing. In contrast, a larger number of HIV patients

was demonstrated by van Dyk et al. to have a strong

association between HIV and ocular involvement.

It is

recognized that a positive HIV status can lead to serious

complications due to an individual’s immunodeciency.

erefore, even though HIV status was not signicantly

associated with ocular complications in our study, we

suggest that all HIV patients with HZO should be examined

by ophthalmologists to detect ocular complications and

prevent the development of the most devastating forms.

e higher risk of ocular manifestations in HZO had a

statistically signicant association with Hutchinson’s sign,

a period > 4 days from rash onset to eye examination, and

a period > 3 days from rash onset to acyclovir treatment.

Hutchinson’s sign (or skin involvement of the alar and

tip of nose) demonstrated a statistically significant

association with ocular involvement, whereas other skin

involvement did not. In the study by Adam et al., ocular

involvement in HZO and rash along the supratrochlear

nerve distribution had a statistically signicant correlation.

VZV is reactivated from the ophthalmic division of the

CN V; the ophthalmic division gives rise to the terminal

branches, namely, the supraorbital, supratrochlear, and

nasociliary branches. e nasociliary branch innervates

the skin of the tip of the nose and divides into the long

ciliary nerve, which provides sensory innervation to

the cornea and uvea.

is therefore explains why the

common ocular complications in HZO patients are

keratitis and uveitis. A previous study by Harding et al.

also found that uveitis was the common manifestation

in HZO (46.48%).

From this evidence, a rash appearing

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

369

along the nasociliary branch (Hutchinson’s sign) or the

supratrochlear branch could suggest that the HZO patients

concerned have a high risk of ocular complications.

Treatment with an oral antiviral drug can arrest

viral replication in the early phase of HZO, thereby

stopping the progression and dissemination of the virus.

We found there was a signicant association between

ocular involvement and both a duration of > 4 days from

the onset of skin rash to the time of eye examination,

and a duration of > 3 days from the onset of skin rash to

systemic acyclovir administration. ese data therefore

indicate that a delayed examination or treatment allows

the virus to migrate to the distant ophthalmic nerve,

thereby increasing the risk of ocular involvement.

Given the low number of subjects, the current study

did not nd any severe ocular ndings. Other studies have

reported devastating forms of ocular manifestations, such

as orbital myositis, ophthalmoplegia, optic neuritis, cranial

nerve palsy, and orbital apex syndrome.

11-12

Nonetheless,

we should take special care with HZO patients who have

decreased visual acuities. is state could indicate that

the patient has a chance of developing serious ocular

complications.

e limitations of this study were related to its use

of a retrospective chart review, which provided limited

data, especially for underlying diseases such as HIV. An

HIV test was performed in only 65 patients (40.6%),

not all patients, and there were no records of viral loads

or CD4 counts available to stage the severity of the

HIV disease. Another limitation of our study was the

subjective evaluation of the ocular inammation grading

(both for conjunctival and uveal involvements), which

varied between the graders. However, this should not

aect the main study results of the binary evaluation of

ocular involvements.

CONCLUSION

HZO has been reported to cause many ocular

complications. It is reactivated by an impaired host

immune defense against VZV. ere is a high prevalence

of ocular inammation in acute HZO, especially among

individuals with positive Hutchinson’s sign, delayed eye

examination, and/or delayed acyclovir treatment. e data

suggests that general practitioners should circumspectly

assess HZO patients because any delay in diagnosis and

treatment can result in severe ocular complications

and increased morbidity. Patients who are suspected

of having an ocular involvement should be referred to

an ophthalmologist with haste.

Funding: None

Conicts of Interest: None

REFERENCES

1. Chern KC, Margolis TP. Varicella zoster virus ocular disease.

Int Ophthalmol Clin 1998;38:149-60.

2. Marsh RJ. Herpes zoster ophthalmicus. J R Soc Med 1997;90:

670-4.

3. Harding SP, Lipton JR, Wells JC. Natural history of herpes

zoster ophthalmicus: predictors of postherpetic neuralgia and

ocular involvement. Br J Ophthalmol 1987;71:353-8.

4. Zaal MJ, Volker-Dieben HJ, D’Amaro J. Prognostic value of

Hutchinson’s sign in acute herpes zoster ophthalmicus. Graefes

Arch Clin Exp Ophthalmol 2003;241:187-91.

5. Shaikh S, Ta CN. Evaluation and management of herpes zoster

ophthalmicus. Am Fam Physician 2002;66:1723-30.

6. Kaufman SC. Anterior segment complications of herpes zoster

ophthalmicus. Ophthalmology 2008;115(2 Suppl):S24-32.

7. Liesegang TJ. Herpes zoster ophthalmicus natural history,

risk factors, clinical presentation, and morbidity. Ophthalmology

2008;115(2 Suppl):S3-12.

8. Vrcek I, Choudhury E, Durairaj V. Herpes Zoster Ophthalmicus:

A Review for the Internist. Am J Med. 2017;130:21-6.

9. Phillips AJ, Speedwell L, Makepeace C. CCLRU grading scales

(Appendix D). Contact Lenses. 1997.p.863-7.

10. Standardization of Uveitis Nomenclature (SUN) Working

Group. Standardization of uveitis nomenclature for reporting

clinical data. Results of the First International Workshop. Am

J Ophthalmol 2005;140:509-16.

11. Bak E, Kim N, Khwarg SI, Choung HK. Case Series: Herpes

Zoster Ophthalmicus with Acute Orbital Inammation. Optom

Vis Sci 2018;95:405-10.

12. Chiang E, Bajric J, Harris GJ. Herpes Zoster Ophthalmicus With

Orbital Findings Preceding Skin Rash. Ophthalmic Plast

Reconstr Surg 2018;34:e113-e5.

13. Rousseau A, Bourcier T, Colin J, Labetoulle M. Herpes Zoster

Ophthalmicus—Diagnosis and Management. US Ophthalmic

Review 2013;6:119-24.

14. Adam RS, Vale N, Bona MD, Hasanee K, Farrokhyar F. Triaging

herpes zoster ophthalmicus patients in the emergency department:

Do all patients require referral? Acad Emerg Med 2010;17:

1183-8.

15. Bayu S, Alemayehu W. Clinical profile of herpes zoster

ophthalmicus in Ethiopians. Clin Infect Dis 1997;24:1256-60.

16. Umeh RE. Herpes zoster ophthalmicus and HIV infection in

Nigeria. Int J STD AIDS 1998;9:476-9.

17. Nassaji-Zavareh M, Taheri R, Ghorbani R, Aminian M.

Undiagnosed diabetes mellitus in patients with Herpes Zoster.

Indian J Dermatol 2008;53:119-21.

18. Kaiserman I, Kaiserman N, Nakar S, Vinker S. Herpetic eye

disease in diabetic patients. Ophthalmology 2005;112:2184-8.

19. Zaal MJ, Volker-Dieben HJ, D’Amaro J. Visual prognosis in

immunocompetent patients with herpes zoster ophthalmicus.

Acta Ophthalmol Scand 2003;81:216-20.

20. Albrecht MA. Clinical manifestations of varicella-zoster virus

infection: Herpes zoster. UpToDate, UpToDate, Inc., Waltham,

MA. 2014.

21. Szeto SK, Chan TC, Wong RL, Ng AL, Li EY, Jhanji V. Prevalence

of Ocular Manifestations and Visual Outcomes in Patients

with Herpes Zoster Ophthalmicus. Cornea 2017;36:338-42.

22. Van Dyk M, Meyer D. Hutchinson’s sign as a marker of ocular

involvement in HIV-positive patients with herpes zoster

ophthalmicus. S Afr Med J 2010;100:172-4.

Thamphithak et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

370

Original Article

SMJ

Kitirat Techatraisak, M.D., Ph.D.*, Nattamon Jiwornlerk, M.D.**

*Department of Obstetrics and Gynaecology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, **Division of Obstetrics and

Gynaecology, Charoen Krung Pracha Rak Hospital, Bang Ko Lam, Bangkok 10120, ailand.

Association Between Body Mass Index and

Moderate-to-Severe Vasomotor Symptoms in Thai

Postmenopausal Women

ABSTRACT

Objective: Body mass index (BMI) has been controversially reported to be positively, inversely, or not associated

with vasomotor symptoms (VMS) in postmenopausal women. is study analysed the association between BMI

and moderate-to-severe VMS in ai population.

Methods: A cross-sectional study was conducted among 178 ai postmenopausal women aged 30-60 years. ey

were categorized into two groups by BMI. e two groups (normal BMI <25.0 kg/m

and high BMI ≥25.0 kg/m

)

were compared. Modied menopausal rating scale (MRS) was used to assess the severity of VMS. e association

between BMI, types and duration of menopause, and VMS were analyzed by chi-square test and t-test analysis.

Multiple logistic regression analysis was used to evaluate the associations between BMI, other variables, and severity

of VMS.

Results: Each group comprised 89 consecutive postmenopausal women. e percentage of women with moderate-

to-severe VMS was signicantly higher in the normal BMI group than in the high BMI group (60.9% vs 39.1%,

respectively; p= 0.007). e association between normal BMI and moderate-to-severe VMS was statistically signiﬁcant

aer adjustment for other confounding variables that included age, medical illness, duration or type of menopause

and current medications (adjusted odd ratios [OR]: 2.378; 95% condence interval [CI]: 1.273, 4.447).

Conclusion: Interestingly, ai postmenopausal women with a normal BMI experience a higher rate of moderate-

to-severe VMS than those with a high BMI. e nding indicates that BMI is a risk factor for VMS among ai

postmenopausal women. Healthcare providers should focus on normal BMI ai postmenopausal women when

assessing for and treating VMS.

Keywords: BMI; VMS; postmenopausal women; ai (Siriraj Med J 2019; 71: 370-376)

Corresponding author: Kitirat Techatraisak

E-mail: kitirat.tec@mahidol.ac.th

Received 2 November 2018 Revised 1 July 2019 Accepted 5 August 2019

ORCID ID: http://orcid.org/0000-0001-8449-9379

http://dx.doi.org/10.33192/Smj.2019.56

INTRODUCTION

Menopause is recognized aer 12 months’ absence

of menstruation. The typical age range for onset of

menopause is 47 to 50 years.

Menopausal symptoms

such as vasomotor symptoms (VMS: hot ushes and

night sweats), sleeplessness, urinary frequency, vaginal

dryness, anxiety and depression, adversely aect quality

of life.

2,3

In general, VMS persists for six months to two

years, but they may last for more than 10 years.

4-6

e

frequency and severity of VMS are dierent in each

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

371

Techatraisak et al.

woman. e prevalence of VMS varies widely across

populations and ethnicities. Approximately 40-70% of

postmenopausal women experience VMS.

VMS aects quality of life by causing insomnia,

depression, anxiety and fatigue. ere are many methods

for treating VMS such as lifestyle modications, taking

vitamin supplements, and/or using herbs.

4,6-8

Hormonal

therapies (e.g., estrogen and tibolone) and non-hormonal

therapies (e.g., clonidine, gabapentin, and some serotonin

and serotonin–norepinephrine reuptake inhibitors),

are also used for treating VMS. However, the easiest

and most widely available treatment method is change

in lifestyle. Women with moderate-to-severe VMS are

indicated for hormonal therapy.

e pathophysiology of VMS is not well understood.

e rst hypothesis is that a central event that is initiated

in the hypothalamus drives an increase in core body

temperature, metabolic rate and skin temperature. Another

hypothesis is that estrogen withdrawal or deciency causes

VMS. is hypothesis is supported by the eectiveness

of estrogen therapy for relieving VMS.

e physical

and financial burdens of VMS are considerable. If

practitioners know the risk factors for VMS, they can

advise postmenopausal women regarding the behaviors

to avoid or adopt in order to relieve the symptoms of

VMS, and reduce the probability of developing VMS.

Lower BMI was reported to be a risk factor for VMS.

10,11

e large SWAN cross-sectional study found BMI to be

positively correlated with self-reported hot ashes and/

or night sweats.

One recent study reported that obese

women have a more symptomatic menopausal transition

compared to normal weight women.

However, some

previous studies did not nd association between BMI

and VMS among postmenopausal women.

14-16

If we could identify postmenopausal women at high

risk for VMS, we could develop appropriate preventive

strategies including lifestyle modications and surveillance

in postmenopausal women. Accordingly, the aim of this

study was to investigate the association between BMI

(normal vs. overweight) and moderate-to-severe VMS

in ai postmenopausal women.

MATERIALS AND METHODS

is retrospective cross-sectional study, was conducted

during 2012-2013 among consecutive new menopausal

patients who attended the Gynecologic Endocrinology

Unit, of the Department of Obstetrics and Gynecology,

Faculty of Medicine Siriraj Hospital, Mahidol University,

Bangkok, ailand during 2008-2011. e protocol for

this study was approved by the Ethic Committee of

Faculty of Medicine Siriraj Hospital (Si 659/2011).

e target population included naturally, surgically

and prematurely menopausal women aged 30–60 years.

Women currently using of hormonal therapies or drugs

that aect VMS (e.g., SSRIs, clonidine, and gabapentin)

were excluded. Detailed menopausal symptoms were

obtained from subject response to structured questions

(modied menopausal rating scale, modied MRS).

e scale was originally developed and validated over

several years by a research network consisting of many

institutions located in Germany during the 1990s. [ai

questionnaire Version 5.0 (validation with permission),

was used for the research “Menopause-related symptoms

and quality of life in peri- and postmenopausal women

in ailand: multicenter study” that was published in

Asian Biomedicine 2017;11:207-15]. The objectives

of MRS were to measure the severity of menopausal

symptoms, changes before and aer treatment with

hormone replacement therapy, and their impact on

health-related quality of life. e patients rst read a

menopausal symptoms list, including VMS. ey then

ranked the level of severity from 0-4 (no symptom,

mild, moderate and severe symptoms, respectively).

Scores of 0 and 1 were classied as mild symptoms, and

scores of 2, 3, or 4 were classied as moderate-to-severe

symptoms.

17,18

e sample size was calculated based on a pilot

study of 20 patients who were divided into two groups:

normal weight (BMI <25kg/m

) and overweight (BMI

≥25kg/m

) according to World Health Organization’s

denitions.

BMI was calculated as weight in kilograms

(kg) divided by height in meter squared (m

). e results

revealed that 4 of 10 patents in the normal weight group

experienced moderate-to-severe VMS. In the overweight

group, only 2 of 10 patients experienced moderate-to-

severe VMS. e estimated sample size (type I error of

5 %, and type II error of 20%) was 81 patients in each

group. e estimated sample size was increased by 10%

to compensate for missing or incomplete data. Patients

were consecutively recruited from 2008 until the total

sample size of 89 participants per group was achieved,

as shown in Fig 1.

Retrieved demographic and clinical data included

age, body weight, height, medical history, health behaviors,

such as smoking and alcohol use, type of menopause,

current medication, hormone replacement therapy, and

severity of menopausal symptoms.

Statistical analysis

e study variables were compared among groups

using the SPSS 18.0 statistical package. Chi-square test and

Fisher’s exact test were used for categorical data. T-test

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

372

Original Article

SMJ

was used for analyzing continuous data. Multiple logistic

regression analysis was used to estimate the association

between each of the independent variables, and for

adjusting for potential confounders. e adjustments

were used for independent variables, including BMI,

age, medical illness, duration of menopause, type of

menopause, and current medication. Signicant level

was dened as p-value <0.05.

RESULTS

One hundred and seventy-eight postmenopausal

women that were included in this study. Characteristics of

the normal weight and overweight groups are presented

in Table 1. In addition, both groups of participants were

non-smoking and non-alcohol drinking. e demographic

and clinical characteristics were similar between groups

except for age, medical illness, and antihypertensive drug

use. e mean duration of menopause was not signicantly

dierent between groups. Aside from hypertensive drug

use, there were no dierences between groups for current

medications.

The associations between BMI and severity of

VMS, type of menopause, duration of menopause and

other variables are shown in Table 2. Postmenopausal

women with BMI <25 kg/m

were signicantly more

likely to report moderate-to-severe VMS than higher

BMI postmenopausal women (p =0.004). No signicant

association was observed between severity of VMS and

type or duration of menopause, age, medical illness, or

current medication.

Aer adjustment for covariates using multiple logistic

regression, the normal weight group was signiﬁcantly

associated with greater odds of reporting moderate to

severe VMS as shown in Table 3. Only the association

between normal BMI and moderate to severe VMS

remained signiﬁcant aer adjustment for age, medical

illness, duration of menopause, type of menopause and

current medication (adjusted odd ratios [OR]: 2.378;

95% condence interval [CI]: 1.273, 4.447).

DISCUSSION

One of the most commonly studied menopausal

symptoms associated with obesity is VMS. Obesity was

reported to be a protective factor for VMS.

20,21

On the

contrary, the nding from this study in ai population

showed a dierent result. ai postmenopausal women

with a BMI <25 kg/m

were more likely to experience

moderate - to - severe VMS than women with a higher

BMI aer adjusting for confounding variables.

The result of this study may be explained by

estrone, which is a predominant form of estrogen in

postmenopausal women that is produced in adipose

tissue via the conversion of androstenedione. We assumed

that women with lower BMI might have less adipose

494 new postmenopausal women aged 30-60 years who registered

at the Gynecologic Endocrinology Unit with available BMI and

VMS rating during 2008-2011 were enrolled.

Women on menopausal hormone therapy

were excluded (58 cases).

Women on drug aecting VMS

were excluded (37 cases).

89 normal BMI women ( <25 kg/m

)

were consecutively retrieved and reviewed

for analysis

89 high BMI women ( ≥25 kg/m

)

were retrieved and reviewed for

analysis

Fig 1. Flow diagram of the subject selection process.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

373

TABLE 1. Characteristics of study population.

BMI <25 kg/m

 BMI≥25kg/m

P-value

(N=89) (N=89)

Age (Mean±SD, yr) 51.87 ± 5.35 53.47 ± 4.22 0.028

Type of menopause (case, %) 0.332

Natural menopause 60 (67.4%) 56 (62.9%)

Surgical menopause 29 (32.6%) 31 (34.8%)

Premature menopause 0 (0%) 2 (2.2%)

Duration of menopause (Mean±SD, yr) 4.84 ± 4.63 5.34 ± 4.53 0.467

Medical illness (case, %) 24 (26.7%) 36(40.9%) 0.044

Hypertension 12 (13.3%) 23 (26.1%) 0.032

Diabetes mellitus 4 (4.4%) 7 (8%) 0.331

Dyslipidemia 10 (11.1%) 13 (14.8%) 0.467

Chronic kidney disease 1 (1.1%) 0 (0%) 1

Thyroid disease 3 (3.3%) 2 (2.3%) 1

Allergy 1 (1.1%) 1 (1.1%) 1

Gout 2 (2.2%) 0 (0%) 0.497

CA breast 2 (2.2%) 0 (0%) 0.497

Cerebrovascular disease 1 (1.1%) 0 (0%) 1

Osteoporosis 2 (2.2%) 1 (1.1%) 1

Current medications (case, %) 19 (21.1%) 27 (30.7%) 0.145

Antihypertensive drug 9 (10%) 20 (27%) 0.022

Anti-hyperglycemic drugs 3 (3.3%) 5 (5.7%) 0.494

Hypolipidaemic agents 7 (7.8%) 11 (12.5%) 0.296

Thyroid drug 2 (2.2%) 0 (0%) 0.497

Allopurinol 2 (2.2%) 0 (0%) 0.497

Tamoxifen 2 (2.2%) 0 (0%) 0.497

Bisphosphonate 1 (1.1%) 1 (1.1%) 1

tissue, so they might have lower estrone and a higher

risk of developing VMS. A lower level of estrogen is

also one of the possible causes of VMS. ere are many

reports that support the results of this study. A group of

postmenopausal women was divided into two subgroups

and estrogen levels were measured and compared. e

results showed that the severe VMS group had lower

levels of estrone, estradiol, and body weight than the other

group that never experienced VMS (p <0.05).

e result

of our study supported one former study of naturally

postmenopausal women that found that overweight

women suered from hot ushes and perspiration less

than thinner women.

Another later study also found

the risk of VMS to be higher in heavier women. A cross-

sectional study of women aged 40-60 years that were

classied according to BMI (≤24.9, 25.0–29.9, or ≥30 kg/m

)

Techatraisak et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

374

Original Article

SMJ

TABLE 2. e association between BMI, type of menopause, duration of menopause, age, medical illness, current

medications and severity of VMS.

TABLE 3. Multivariable logistic regression model for moderate to severe VMS.

Vasomotor symptoms

Mild (N=91) Moderate - to - severe (N=87) P-value

BMI 0.004

< 25 kg/m

36 (39.6%) 53 (60.9%)

 ≥25kg/m

55 (60.4%) 34 (39.1%)

Type of menopause 0.372

Natural menopause 58 (63.7%) 58 (66.7%)

Surgical menopause 31 (34.1%) 29 (33.3%)

Premature menopause 2 (2.2%) 0 (0%)

Duration of menopause (Mean±SD, yr) 5.42 ± 4.47 4.74 ± 4.69 0.326

Age (Mean±SD, yr) 52.75 ± 4.63 52.59 ± 5.35 0.829

Medical illness 33 (35.9%) 27 (31.4%) 0.528

Current medications 25 (27.2%) 21 (24.4%) 0.675

Unadjusted odds ratio Adjusted odds ratio P-value

( 95 % CI ) ( 95 % CI )

BMI

< 25 kg/m

2.38 (1.307-4.359) 2.38 (1.273-4.447) 0.007

 ≥25kg/m

1 1

Type of menopause

Natural menopause 1.033 0.868 0.7

Surgical menopause 1 1

Premature menopause NA NA

Age 0.996 (0.938-1.058) 1.024 (0.954-1.099) 0.512

Duration of menopause 0.97 (0.909-1.035) 0.973 (0.901-1.052) 0.496

Medical illness 0.818 (0.439-1.526) 0.938 (0.333-2.64) 0.904

Current medications 0.86 (0.442-10697) 1.033 (0.344-3.1) 0.953

Abbreviations: CI = Conﬁdence Interval, NA = not applicable

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

375

investigated the relationship between lifestyle factors

(e.g., BMI and smoking) and moderate-to-severe VMS.

e result showed that BMI ≥30 kg/m

was associated

with an increased risk of moderate-to-severe hot ashes

when compared with BMI ≤24.9 kg/m

(adjusted OR: 2.1,

95% CI: 1.5-3.0) in premenopausal or perimenopausal

women.

is might be due to the fact that adipose tissue

produces hormones (e.g., leptin, tumor necrosis factor)

that inhibit ovarian steroid production. us, there was

only a slight level of estrogen in obese women, and this

led to the development of severe VMS. One other study

of women aged 47–59 years demonstrated that a greater

percentage of body fat was signicantly associated with

reporting of VMS.

A possible explanation for this may be

that when the body’s adipose tissue volume was increased,

it provided a higher level of insulation, that was followed

by higher body temperature and increasing VMS. In

other words, aromatization was suppressed from high

body temperature.

Another study of perimenopausal

women aged 42-52 years examined the relationship

between VMS and several factors, including hormone

level measurement. e result showed that VMS was

more common in women with high BMI.

ere was

no increase in estrogen level in obese women while

increasing body fat provided more insulation which

would increase core body temperature and thus might

increase in hot ushes, and prevent aromatization in

adipose tissue. A randomized trial in women aged 40-

65 years showed that perimenopausal women with a

BMI ≥25 kg/m

were more likely to report hot ushes

than women with a BMI <25 kg/m

(OR: 2.00, 95% CI

1.28-3.12).

However, BMI was not associated with hot

ushes in postmenopausal women. Some later studies

reported similar results that obesity is associated with

higher VMS prevalence during the midlife.

24-28

Interestingly, some studies reported no association

between BMI and VMS in postmenopausal women.

15,16

On the contrary, a cohort study of both premenopausal

and postmenopausal women reported that the median

duration of moderate to severe VMS was 10.2 years.

at study also found that thinner women had longer

duration of VMS than heavier weight women (BMI >30

kg/m

, p =0.003).

at group postulated that women

with heavier weight have higher estradiol levels than

thinner women. Another later cross-sectional study of

natural menopause women aged 40-60 years in factors

related to VMS reported that thin and smoking women

during the premenopausal period were more likely to

experience VMS than obese women (p =0.03). However,

from that study only BMI could not be the only factor

that aected VMS.

e results of this study showed results dierent

from many previous studies. Perception of VMS may

vary among ethnicities and countries, and the structured

questions used to evaluate VMS may vary among studies.

e design of the study (cross-sectional vs. cohort) and

subject selection (perimenopausal vs. postmenopausal)

may also aect the result since VMS may be perceived

mostly during the perimenopausal period.

Limitation

VMS is subjective, self-reported, and there is no

standard for measuring the level of VMS in individuals.

Dierent patients perceive dierent levels of VMS. In

other words, an individual’s perception of VMS may not

be estimated as the same level between reports, which

makes it dicult to compare the level of symptoms.

In order to answer our questionnaire, patients had to

recall their recent symptoms during their rsts visit to

our menopause clinic.

CONCLUSION

In this study population, ai menopausal women

with a normal BMI were signicantly more likely to

report moderate-to-severe VMS than postmenopausal

women with higher BMI. This finding supports the

hypothesis that heavier ai women may have a higher

level of estrogen due to aromatization in adipose tissue.

ese results suggest the need for more careful screening

of under or normal weight women for VMS during

menopause.

ACKNOWLEDGMENTS

We would like to thank Ms. Julaporn Pooliam,

statistician, for statistical analyzed and Ms. Chongdee

Dangrat, scientist, for cases collected. is research was

supported by a Siriraj Grant for Research and Development

(grant no. R015531024).

REFERENCES

1. Delavar MA, Hajiahmadi M. Age at menopause and measuring

symptoms at midlife in a community in Babol, Iran. Menopause

2011;18:1213-8.

2. Fallahzadeh H. Age at natural menopause in Yazd, Islamic

Republic of Iran. Menopause 2007;14:900-4.

3. Nelson DB, Sammel MD, Patterson F, Lin H, Gracia CR, Freeman

EW. Eects of reproductive history on symptoms of menopause:

a brief report. Menopause 2011;18:1143-8.

4. urston RC, Sowers MR, Chang Y, Sternfeld B, Gold EB,

Johnston JM, et al. Adiposity and reporting of vasomotor

symptoms among midlife women: the study of women’s health

across the nation. Am J Epidemiol 2008;167:78-85.

5. Gold EB, Block G, Crawford S, Lachance L, FitzGerald G,

Miracle H, et al. Lifestyle and demographic factors in relation

Techatraisak et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

376

Original Article

SMJ

to vasomotor symptoms: baseline results from the Study of

Women’s Health Across the Nation. Am J Epidemiol 2004;159:

1189-99.

6. Freeman EW, Sammel MD, Lin H, Liu Z, Gracia CR. Duration

of menopausal hot ushes and associated risk factors. Obstet

Gynecol 2011;117:1095-104.

7. Whiteman MK, Staropoli CA, Langenberg PW, McCarter

RJ, Kjerul KH, Flaws JA. Smoking, body mass, and hot ashes

in midlife women. Obstet Gynecol 2003;101:264-72.

8. Borelli F, Earnst E. Black cohosh (Cimicifuga racemosa) for

menopausal symptoms: a systematic review of its ecacy.

Pharmacol Res 2008;58:8-14.

9. Berek JS, Novak E. Berek & Novak’s gynecology. 14th ed.

Philadelphia, Penn.: Lippincott Williams & Wilkins; 2007.

10. Campagnoli C, Morra G, Belforte P, Belforte L, Prelato Tousijn

L. Climacteric symptoms according to body weight in women

of dierent socio-economic groups. Maturitas 1981;3:279-87.

11. Schwingl PJ, Hulka BS, Harlow SD. Risk factors for menopausal

hot ashes. Obstet Gynecol 1994;84:29-34.

12. Gold EB, Sternfeld B, Kelsey JL, Brown C. Mouton C, Reame

N, et al. Relation of demographic and lifestyle factors to symptoms

in a multi-racial/ethnic population of women 40-50 years of

age. Am J Epidemiol 2000;17:718-26

13. Fernandez-Alonso AM, Caudros JL, Cherdraui P, Mendoza

M, Caudros AM, Perez-Lopez FE. Obesity is related to increased

menopausal symptoms among Spanish women. Menopause

Int 2010,16:105-10.

14. Whiteman MK, Staropoli CA, Langenberg P, McCarter RJ,

Kjerul KH, Flawa JA. Smoking, body mass, and hot ashes

in midlife women. Obstet Gynecol 2003;101:264-72.

15. Sabia S, Fournier A, Mesrine S, Boutron-Ruault MC, Clavel-

Chapelon F. Risk factors for onset of menopausal symptoms:

results from a large cohort study. Maturitas 2008;60:108-21.

16. urston RC, Santoro N, Matthews KA. Adiposity and hot

ashes in midlife women: a modifying role of age. J Clin

Endocrinol Metab 2011;96:E1588-E1595.

17. Heinemann LA, DoMinh T, Strelow F, Gerbsch S, Schnitker

J, Schneider HP. e Menopause Rating Scale (MRS) as outcome

measure for hormone treatment? A validation study. Health

Qual Life Outcomes 2004;2:67.

18. Dinger J, Zimmermann T, Heinemann LA, Stoehr D. Quality

of life and hormone use: new validation results of MRS scale.

Health Qual Life Outcomes 2006;4:32.

19. Appropriate body-mass index for Asian populations and

its implications for policy and intervention strategies. Lancet

2004;363(9403):157-63.

20. Randolp JF Jr, Sheng H, Sowers MR, Crandall C, Crawford

S, Gold EB, et al. Change in follicular-stimulating hormone

and estradiol across the menopausal transition.: Eect of age at

the nal menstrual period. J Clin Endocrinol Metab 2011;96:

746-54.

21. Freeman EW, Sammel MD, Lin H, Gracia CR. Obesity and

reproductive hormone levels in the transition to menopause.

Menopause 2010;17:718-26.

22. Erlik Y, Meldrum DR, Judd HL. Estrogen levels in postmenopausal

women with hot ashes. Obstet Gynecol 1982;59:403-7.

23. Hyde Riley E, Inui TS, Kleinman K, Connelly MT. Dierential

association of modiable health behaviors with hot ashes

in perimenopausal and postmenopausal women. J Gen Intern

Med 2004;19:740-6.

24. Galicchio L, Visvanathan K, Miller SR, Babus J, Lewis LM,

Zacur H, et al. Body mass, estrogen levels, and hot ahes in

midlife women. Am J Obstet Gynecol 2005;193:1353-60.

25. Kumari M, Staord M, Marmot M. e menopausal transition

was associated in a prospective study with decreased health

functioning in women who report menopausal symptoms. J

Clin Epidemiol 2005,58:719-27.

26. Gold EB, Colvin A, Avis N, Boomberger J, Greendale GA,

Powell L, et al. Longtitudinal analysis of the association between

vasomotor symptoms and race/ethnicity across the menopausal

transition: study of women’s health across the nation. Am J

Public Health 2006;96:1226-35.

27. urston RC, Sowers MR, Sutton-Tyrrell K, Everson-Rose

SA, Lewis TT, Edmundowicz D, et al. Abdominal obesity and

hot ashes among midlife women. Menopause 2008,15:429-34.

28. urston RC, Chang Y, Mancuso P, Matthews KA. Adipokines,

adiposity, and vasomotor symptoms during the menopause

transition: ndings from the Study of Women’s Health Across

the Nation. Fertil Steril 2013;100:793-800.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

377

Vannipa Vathanophas, M.D., Kotchakarn Supachavalit, M.D., Phawin Keskool, M.D., Sunun Ongard, M.D.,*

Department of Otorhinolaryngology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.

Factors Associated with Xerostomia in

Non-Radiated Patients

ABSTRACT

Objective: To identify factors signicantly associated with xerostomia in non-radiated patients.

Methods: Patients who attended the outpatient otolaryngology clinic at Siriraj Hospital (Bangkok, ailand) with

complaints of dry mouth were invited to join this study. Collected data included age, gender, body mass index,

smoking status, alcohol use, underlying disease, and previous medication and/or therapy. Irradiated patients

were excluded. Participants were classied into either the diseased or xerostomia group by abnormal oral cavity

examination and symptoms, or the no xerostomia group, which was dened as dry mouth symptoms with no

presence of abnormal physical ndings.

Results: Two hundred and two participants with a history of dry mouth were consecutively enrolled. ere were

86 patients with physical ndings compatible with xerostomia, and 116 symptomatic patients without xerostomia.

Multivariate analysis revealed age over 50 years (adjusted odds ratio [aOR]: 3.1, 95% condence interval [CI]: 1.3-7.9;

p=0.012), analgesic and muscle relaxant intake (aOR: 3.6, 95% CI: 1.3-9.7; p=0.012), psychotherapeutic medication

(aOR: 7.8, 95% CI: 2.6-23.7; p<0.001), and radioactive iodine therapy (aOR: 3.7, 95% CI: 1.2-11.8; p=0.015) to be

independent predictors of xerostomia.

Conclusion: Xerostomia is a condition that can adversely aect quality of life. e results of this study revealed older

age (≥50 years), analgesics and muscle relaxants, psychotherapeutic medications, and radioactive iodine therapy to

be signicantly associated with xerostomia. A thorough understanding of the symptoms, diagnosis, relevant risk

factors, and eective management is essential for improving outcomes among patients with xerostomia.

Keywords: Xerostomia; underlying disease; medication; non-radiation patients (Siriraj Med J 2019; 71: 377-384)

Corresponding author: Sunun Ongard

E-mail: sunun.ong@mahidol.ac.th

Received 23 August 2018 Revised 1 May 2019 Accepted 14 May 2019

ORCID ID: http://orcid.org/0000-0003-4972-984X

http://dx.doi.org/10.33192/Smj.2019.57

INTRODUCTION

Saliva plays an important role in oral health by

helping to prevent infection, and by facilitating chewing,

swallowing, and speaking. Xerostomia and salivary gland

hypofunction are two terms that highlight the important

relationship between saliva and oral health. However,

no signicant association between these two conditions

has been reported. Some patients have salivary gland

hypofunction without having xerostomia. However,

it was reported that individuals with xerostomia may

have abnormal or low ow of saliva.

1,2

Stimulated and

non-stimulated saliva ow rates of <0.1 ml/min and <0.7

ml/min, respectively, are diagnosed as salivary gland

hypofunction.

Xerostomia is a common subjective

complaint of dryness in the mouth.

Dening xerostomia

can be problematic and potentially confusing, because

there are different questionnaires that use different

criteria to identify xerostomia. Among the available

questionnaire-based tools for diagnosing xerostomia,

the Xerostomia Inventory is the most frequently used

tool in a research setting. e Xerostomia Inventory is an

11-item tool that quanties the severity of xerostomia.

5,6

Vathanophas et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

378

Original Article

SMJ

Saliva has various roles and functions. First, it provides

a defense against bacterial pathogens via enzymes, such

as immunoglobulin A and lysozyme.

7,8

Second, it is an

emulgent that aids enzymes that assist in the digestion

of food. ird, saliva provides a physical glycoprotein

and mucoid coating to prevent unfavorable substances

from attaching to the teeth. is coating also aids in oral

lubrication and swallowing.

Fourth, the bicarbonate and

phosphate buers in saliva help to maintain a neutral

oral pH.

erefore, patients that have xerostomia with

either low or abnormal salivary ow are at high risk

for developing oral health-related problems, such as

dental caries, mucosal ulceration, oral candidiasis, and

dysphagia – all of which can adversely impact quality of

life.

It is, therefore, necessary for clinicians to understand

the multidimensional aspects of this condition so that

proper diagnosis can be made and proper treatment can

be given in order to reduce complications and improve

patient outcomes.

Xerostomia aects millions of people around the

world. It is dicult to determine the exact prevalence of

this condition; however, the prevalence was reported to

range from 12% to 30%.

e most commonly reported

cause of xerostomia is radiation to the head and neck

region. Other possible causes that have been reported

include medications and some specic diseases.

11-13

Improved understanding of the factors that signicantly

associate with xerostomia will improve our understanding

of this condition, and improve diagnosis, treatment,

and outcomes. Accordingly, the aim of this study was to

identify factors signicantly associated with xerostomia

in non-radiated patients that presented with dry mouth

symptoms at Siriraj Hospital – ailand’s largest national

tertiary referral hospital.

MATERIALS AND METHODS

Study design and population

e protocol for this prospective cross-sectional

study was approved by the Siriraj Institution Review

Board (SIRB) of the Faculty of Medicine Siriraj Hospital,

Mahidol University, Bangkok, ailand (Si 512/2016). We

consecutively recruited two hundred and two participants

aged >18 years with symptomatic dry mouth from the

outpatient clinic of the Department of Otorhinolaryngology

during the September 2016 to December 2017 study

period. Patients satisfying one or more of the following

criteria were excluded: 1) declined to participate in the

study; 2) unable to answer the questionnaire; 3) unable

to undergo the physical examination; and/or, 4) had

history of radiation therapy or chemotherapy at the

head and/or neck.

Data collection

Patients that complained of dry mouth were asked

to take the ve-question Xerostomia Inventory – Dutch

Version.

5,6

If the patient reported having one or more of

the ve listed symptoms, the patient was invited to join

the study. Written informed consent was obtained from

all participating patients. Clinical history to taken and

physical examination was performed. Information that

was collected included age, gender, body mass index,

smoking status, alcohol use, medications, and current

medical status. Using this information and the results

of the physical examination, participants were allocated

to either the diseased or xerostomia group

12,14

(dened

as patients who had history of dry mouth with physical

examination that revealed abnormal oral cavity) or the

no xerostomia group (dened as having symptoms of

dry mouth, but with no presence of abnormal physical

ndings) (Fig 1).

Statistical analysis

SPSS statistics version 18 (SPSS, Inc., Chicago,

IL, USA) was used to perform all statistical analyses.

Descriptive statistics were used to summarize patient

demographic and clinical data. Data are presented as

frequency and proportion. Univariate analysis using chi-

square test of independence was performed to evaluate

association between investigated variables and xerostomia.

Variables with a p-value <0.2 in univariate analysis were

included in multivariate analysis using logistic regression

model. e results of multivariate analysis are presented

as odds ratio (OR) with 95% condence interval (CI)

and adjusted OR (aOR) with 95% CI. A p-value <0.05

indicates statistical signicance.

RESULTS

e study population comprised 202 participants

with complaint of dry mouth, and the age range of

patients was 19-81 years. Eighty-six participants were

allocated to the xerostomia group, and 116 participants

were assigned to the no xerostomia group. e mean

age of patients was 63.30±11.26 years in the xerostomia

group, and 57.44±14.7 years in the no xerostomia group.

Among the entire study cohort, 57 (28.2%) participants

were male and 145 (71.8%) were female.

As shown in Table 1, signicantly more patients in

the xerostomia group acknowledged having the symptoms

listed on the self-report questionnaire than patients in

the no xerostomia group for all 5 of the listed symptoms

(all p<0.001). “My mouth feels dry” (87%) and “My lips

feel dry” (77%) were the two most commonly reported

symptoms.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

379

Fig 1. Flow diagram.

TABLE 1. Patient-reported symptoms.

Variable Number (%) Odds Ratio P-value

Xerostomia No-Xerostomia (95%CI)

(N=86) (N=116)

My mouth feels dry 75 (87.2) 75 (64.7) 3.7 (1.8-7.8) < 0.001**

My mouth feels dry when eating a meal 15 (17.4) 2 (1.7) 12.0 (2.7-54.2) < 0.001**

Ihavedifcultyineatingdryfoods 21(24.4) 3(2.6) 12.2(3.5-42.4) <0.001**

Ihavedifcultiesswallowingcertainfoods 26(30.2) 7(6.0) 6.7(2.8-16.5) <0.001**

My lips feel dry 66 (76.7) 39 (33.6) 6.5 (3.5-12.3) < 0.001**

**P<0.05: statistical signicant

Fig 2 illustrates the results of various physical

examinations in the xerostomia group. e most common

ndings from physical examination in the xerostomia

group were lack of a saliva pool in the oor of the mouth

(61.6%); presence of erythematous, ssured, and dry

tongue with atrophy of the liform papillae, and a pebbled

cobblestone appearance (58.1%); dry, cracked, and peeling

lips (50.0%); and, dry and sticky oral mucosa (48.8%).

Univariate analysis revealed the prevalence of

xerostomia to be signicantly increased in the ≥50 age

group compared to the 0-49 age group (odds ratio [OR]:

2.9, 95% condence interval [CI]: 1.3-6.7; p=0.006). ere

were no signicant dierences between the xerostomia

and no xerostomia groups for gender, body mass index,

smoking status, or alcohol consumption. Univariate analysis

for association between xerostomia with underlying

Announcing volunteers who self-reported symptoms of dry mouth at outpatient

and inpatient clinics, Department of Otolaryngology, Siriraj Hospital

Questions to ask to contribute to the research

1. My mouth feels dry.

2. My mouth feels dry when eating a meal.

3. I have diculty in eating dry foods.

4. I have diculties swallowing certain foods.

5. My lip feel dry.

e physical examination showed the relationship between the patients with xerostomia.

1. Dry, cracked and peeling lips. 5. Bad breath.

2. Absence of a pool pf saliva in the oor of mouth. 6. Atypical pattern of dental caries.

3. Dry and sticky oral mucosa. 7. Oral candidiasis.

4. Erythematous, ssured and dry tongue with 8. Angular cheilitis.

Atrophy of the liform papillae and a pebbled,

Cobblestone appearance.

Diseased Cases

Patients who have one to ve symptoms and one or

more of the above-mentioned physical examinations.

Controlled Cases

Patients who have one to ve symptoms, but no

abnormalities were detected.

Both groups will be interviewed about

• Age, gender and body mass index.

• Habit of smoking and alcohol consumption.

• Underlying disease.

• Medication and therapy

Vathanophas et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

380

Original Article

SMJ

disease, medications and therapies is presented in Table 3.

Factors with a p-value <0.2 in univariate analysis and

factors of interest were included in multivariate analysis.

Multivariate analysis revealed age ≥50 years (adjusted

OR [aOR]: 3.1, 95% CI: 1.3-7.9; p=0.012), analgesic

and muscle relaxant medication (NSAIDs, gabapentin)

Fig 2. Prevalence of physical examination in xerostomia group.

(aOR: 3.6, 95% CI: 1.3-9.7; p=0.012), psychotherapeutic

medication (antidepressant, anxiolytic, antipsychotic)

(aOR: 7.8, 95% CI: 2.6-23.7; p<0.001), and radioactive

iodine therapy (131I) (aOR: 3.7, 95% CI: 1.2-11.8; p=0.015)

to be independent predictors of xerostomia (Table 4).

TABLE 2. Univariable analysis of associations with xerostomia (Age group, Gender, BMI, Smoking and Alcohol

consumption).

Variable Number (%) Odds Ratio P-value

Xerostomia (N=86) No-Xerostomia (N=116) (95%CI)

Age category

0-49 9 (10.5) 30 (25.9) 1.0 0.006*

 ≥50 77(89.5) 86(74.1) 2.9(1.3-6.7)

Gender

Male 18 (20.9) 39 (33.6) 1.0 0.058*

Female 68 (79.1) 77 (66.4) 1.9 (1.0-3.6)

BMI

0-22.99 41 (47.7) 53 (45.7) 1.0 0.887

 ≥23 45(52.3) 63(54.3) 0.9(0.5-1.6)

Smoking

Never/former 85 (98.8) 114 (98.3) 1.0 1.000

Current 1 (1.2) 2 (1.7) 0.7 (0.1-7.5)

Alcohol consumption

Never/former 85 (98.8) 110 (94.8) 1.0 0.242

Current 1 (1.2) 6 (5.2) 0.2 (0.0-1.8)

*P-value less than 0.2 will be selected for multivariable analysis

Abbreviation: BMI= Body mass index

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

381

TABLE 3. Univariable analysis of associations with xerostomia (underlying disease, medication and therapy).

Variable Number (%) Odds Ratio P-value

Xerostomia No-Xerostomia (95%CI)

(N=86) (N=116)

Diabetes mellitus 20 (23.3) 19 (16.4) 1.5 (0.8-3.1) 0.279

Hypertension 41 (47.7) 45 (38.8) 1.4 (0.8-2.5) 0.250

Cardiovascular disorders 6 (7.0) 3 (2.6) 2.8 (0.7-11.6) 0.174

Thyroid disorders 19 (22.1) 35 (30.2) 0.7 (0.3-1.3) 0.260

Psychological disorders 13 (15.1) 1 (0.9) 20.5 (2.6-159.9) < 0.001**

Neurological disorders 6 (7.0) 8 (6.9) 1.0 (0.3-3.0) 1.000

Rhinitis 13 (15.1) 25 (21.6) 0.6 (0.3-1.3) 0.278

Autoimmune disorders 4 (4.7) 2 (1.7) 2.8 (4.5-15.5) 0.405

Respiratory disorders 3 (3.5) 7 (6.0) 0.6 (0.1-2.2) 0.522

Dyslipidemia 33 (38.4) 41 (35.3) 1.1 (0.6-2.2) 0.768

Skeletal disorders 12 (14.0) 7 (6.0) 2.5 (0.9-6.7) 0.086

Hematologic disorders 0 (0.0) 2 (1.7) 0.6 (0.5-0.6) 0.509

Extraesophageal symptoms 5 (5.8) 4 (3.4) 1.7 (0.5-6.6) 0.500

Renal disorders 1 (1.2) 3 (2.6) 0.4 (0.0-4.3) 0.638

Gastrointestinal disorders 5 (5.8) 4 (3.4) 1.7 (0.4-6.6) 0.500

Antihypertensive medication 41 (47.7) 49 (42.2) 1.2 (0.7-2.1) 0.476

Endocrinologic medication 38 (44.2) 42 (36.2) 1.4 (0.8-2.5) 0.309

Analgesic and muscle relaxant medication 16 (18.6) 7 (6.0) 3.6 (1.4-9.1) 0.007*

(NSAIDs, gabapentin)

Antihistamine 14 (16.3) 24 (20.7) 0.7 (0.4-0.5) 0.471

Psychotherapeutic medication 18 (20.9) 5 (4.3) 5.9 (2.1-16.1) < 0.001*

(antidepressant, anxiolytic, antipsychotic)

Antiplatelet (aspirin, clopidogrel) 16 (18.6) 8 (6.9) 3.1 (1.3-7.6) 0.015*

Anticoagulation 1 (1.2) 3 (2.6) 0.4 (0.0-4.3) 0.638

Respiratory medication 4 (4.7) 3 (2.6) 1.8 (0.4-8.4) 0.462

Antidyslipidemia 34 (39.5) 37 (31.9) 1.4 (2.8-2.5) 0.298

Gastrointestinal medication (prokinetic, laxative) 5 (5.8) 2 (1.7) 3.5 (0.7-18.6) 0.138*

Proton pump inhibitor 13 (15.1) 6 (5.2) 2.8 (1.1-7.3) 0.041*

Nutritional medication 21 (24.4) 23 (19.8) 1.3 (0.7-2.6) 0.492

Neurological medication 2 (2.3) 2 (1.7) 1.4 (0.2-9.8) 1.000

Cardiovascular medication 2 (2.3) 1 (0.9) 2.7 (0.2-30.7) 0.576

Immunosuppressant medication 1 (1.2) 1 (0.9) 1.3 (0.1-21.9) 1.000

Radioactive Iodine therapy (

131

I) 9 (10.5) 6 (5.2) 2.1 (0.7-6.3) 0.181*

*P-value less than 0.2 will be selected for multivariable analysis, **P<0.05: statistical signicant

Vathanophas et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

382

Original Article

SMJ

TABLE 4. Factors associated with xerostomia, using multivariate analysis.

Factors Crude Odds Ratio Adjusted Odds ratio P-value

(95%CI) (95%CI)

Age (years)

0-49 1.0 1.0 0.012**

 ≥50 2.9(1.3-6.7) 3.1(1.3-7.9)

Gender

Male 1.0 1.0 0.075

Female 1.9 (1.0-3.6) 1.9 (0.9-4.0)

Analgesic and muscle relaxant medication 3.6 (1.4-9.1) 3.6 (1.3-9.7) 0.012**

(NSAIDs, gabapentin)

Psychotherapeutic medication 5.9 (2.1-16.1) 7.8 (2.6-23.7) < 0.001**

(antidepressant, anxiolytic, antipsychotic)

Antiplatelet (aspirin, clopidogrel) 3.1 (1.3- 7.6) 2.6 (0.9-7.2) 0.064

Gastrointestinal medication (prokinetic, laxatives) 3.5 (0.7-18.6) 2.8 (0.4-18.2) 0.283

Proton pump inhibitor 2.8 (1.1-7.3) 2.5 (0.7-8.3) 0.117

Radioactive Iodine therapy (

131

I) 2.1 (0.7-6.3) 3.7 (1.2-11.8) 0.015**

**P<0.05: statistical signicant

DISCUSSION

is study investigated the risk factors associated

with xerostomia in non-radiated patients. A systemic

review conducted by Tanasiewicz M, et al.

revealed a

prevalence of xerostomia in population-based studies that

ranged from 12% to 30%, with the wide range explained

by dierences in the assessment methods used among

studies. ere was no universal assessment available to

compare xerostomia among those studies. Several dierent

questionnaires for diagnosing xerostomia have been

developed. A frequently used questionnaire in research is

the 11-item Xerostomia Inventory.

1,2,15,16

e Xerostomia

Inventory-Dutch Version is an abbreviated version of

the original version that includes only the following 5

items: “My mouth feels dry”, “My mouth feels dry when

eating a meal”, “I have diculty in eating dry foods”, “I

have diculties swallowing certain foods”, and “My lips

feel dry”. e fact that the modied version has targeted

statements/questions improved the validity of this version

compared to the original version.

5,6

In the present study,

we used this 5-item version and clinical examination to

diagnose xerostomia. e positive oral ndings include

abnormal lips (dry, cracked, and peeling), absence of

a saliva pool in the oor of mouth, dry and sticky oral

mucosa, abnormal tongue (erythematous, ssured, and

dry with atrophy of the liform papillae, and a pebbled,

cobblestone appearance), bad breath, atypical pattern of

dental caries, oral candidiasis, and angular cheilitis. We

found that signicantly more patients in the xerostomia

group reported each of the ve items when compared

to the no xerostomia group (all p<0.001). In addition,

the following two items were reported by more 80% of

patients: “My mouth feels dry when eating a meal” and

“I have diculty in eating dry foods”.

In contrast to salivary gland hypofunction, which can

be objectively evaluated using sialometry, the measurement

of xerostomia remains problematic. Despite no distinctive

relationship with hyposalivation, xerostomia is a critical

symptom in oral health among clinicians, because it aects

both oral health, general health, and quality of life.

Our

opinion that xerostomia can signicantly adversely aect

patient quality of life was our motivation to investigate

the factors that signicantly associate with this important

condition.

Oral mucosal lesions that were commonly found in

this study were dry, cracked, and peeling lips. We also

observed no presence of a saliva pool in the oor of mouth,

dry and sticky oral mucosa, erythematous, and ssured

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

383

and dry tongue with atrophy of the liform papillae and a

pebbled, cobblestone appearance. Changes of the tongue

surface may have both local (salivary gland hypofunction,

inammation, and changes in microcirculation) and

systemic (disease and medication intake) pathogenic

factors in common.

3,12

In this study, we found significant association

between age ≥50 years and xerostomia in multivariate

analysis. is same nding was reported from several

previous studies.

7-9,14,17

ese results may also be caused

by higher medication intake among older adults, with

acini atrophy oen found in patients with decreased

output function.

15,18

e average age that menopause

begins among females in ailand was reported to be

49.5±3.6 years.

A higher prevalence of xerostomia was observed in

females than in males in the present study, but there was no

statistically signicant dierence between genders, which

is similar to the results reported from other studies.

20,21

However, signicant association between gender and

xerostomia was observed in most studies. Menopause in

women could also aect the amount and characteristics

of saliva.

9,22,23

It is also likely that females have to endure

a higher level of pain intensity, and they may have more

illness conditions that could lead to the development of

xerostomia.

Other factors, including body mass index,

tobacco use, and alcohol consumption, were found not

to be associated with xerostomia.

Of note, xerostomia tends to markedly increase in

patients with psychological disorders. Unstable emotions

and changes in personality may play a role in xerostomia,

with changes that can inuence the nervous, immune,

and endocrine systems. Moreover, anti-sialogic eects

were observed in those taking drugs.

17,25

In this study, multivariate analysis revealed signicant

associations between xerostomia and medications or

therapy, including analgesic NSAIDs, gabapentin, muscle

relaxants, psychotherapeutic drugs, and radioactive iodine

therapy. Despite the fact that some of the mechanisms

remain unknown, medication-induced xerostomia is

commonly caused by the altering of neural pathways that

stimulate the secretion of saliva.

Villa A, et al.

showed

that patients taking analgesic medication (NSAIDs), and

Scully C, et al.

reported that muscle relaxant medication,

which blocks sodium and calcium channels, could be

associated with signicant xerostomia.

Psychotherapeutic medications in our study

included antidepressants (tricyclic antidepressant and

selective serotonin-reuptake inhibitor), anxiolytics, and

antipsychotics. Tricyclic antidepressant (TCA) acts as a

noradrenaline and serotonin stimulator by blocking their

reuptake at the neuronal membrane. e block is also to

histaminic, a1-adrenergic, and muscarinic cholinergic

receptors, which can cause unfavorable drug reactions

like xerostomia. Even though xerostomia can result

from selective serotonin-reuptake inhibitor (SSRIs),

they are less likely to cause anticholinergic side eects

than TCA.

22,27

Anxiolytics and antipsychotics are also

prone to cause xerostomia.

We also found radioactive iodine therapy to be a factor

signicantly associated with xerostomia. Accumulation

of radioactive iodine (131I) inicts damage to the salivary

gland, which can cause long-term complications. Qualitative

and quantitative scintigraphy of salivary gland can be

helpful for assessment and follow-up of salivary gland

dysfunction aer radioactive iodine therapy.

28-30

Future studies should investigate whether drug

dosage plays a role in the development of xerostomia,

and comparative studies with a larger sample size would

be required to make a clear distinction. Study in animals

may also help us to elucidate the pathophysiology of

some common medications and their association with

xerostomia (e.g., NSAIDs).

CONCLUSION

Xerostomia is a condition that can adversely aect

quality of life. e results of this study revealed older

age, analgesics and muscle relaxants, psychotherapeutic

medications, and radioactive iodine therapy to be

signicantly associated with xerostomia. A thorough

understanding of the symptoms, diagnosis, relevant

risk factors, and eective management is essential for

improving outcomes among patients with this condition.

ACKNOWLEDGMENTS

e authors gratefully acknowledge the individuals

who agreed to participate in this study, Ms. Ngamrat

Treerassapanich for assistance with manuscript

preparation, and Ms. Julaporn Pooliam of the Division of

Clinical Epidemiology, Department of Research, Faculty

of Medicine Siriraj Hospital, Mahidol University for

assistance with statistical analysis.

Conict of interest declaration: All authors declare

no personal or professional conicts of interest, and no

nancial support from the companies that produce and/

or distribute the drugs, devices, or materials described

in this report.

Funding disclosure: is study was funded by a grant

from the Faculty of Medicine Siriraj Hospital, Mahidol

University, Bangkok, ailand (grant no. R016032008).

Vathanophas et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

384

Original Article

SMJ

REFERENCES

1. Hopcra MS, Tan C. Xerostomia: an update for clinicians.

Aust Dent J 2010;55:238-44.

2. omson WM. Dry mouth and older people. Aust Dent J

2015;60(Suppl 1):54-63.

3. Lynge Pedersen AM, Nauntoe B, Smidt D, Torpet LA. Oral

mucosal lesions in older people: relation to salivary secretion,

systemic diseases and medications. Oral Dis 2015;21:721-9.

4. Abdullah MJ. Prevalence of xerostomia in patients attending

Shorish dental speciality in Sulaimani city. J Clin Exp Dent

2015;7:e45-53.

5. van der Putten GJ, Brand HS, Schols JM, de Baat C. The

diagnostic suitability of a xerostomia questionnaire and the

association between xerostomia, hyposalivation and medication

use in a group of nursing home residents. Clin Oral Investig

2011;15:185-92.

6. omson WM, van der Putten GJ, de Baat C, Ikebe K, Matsuda

K, Enoki K, et al. Shortening the xerostomia inventory. Oral

Surg Oral Med Oral Pathol Oral Radiol Endod 2011;112:

322-7.

7. Tanasiewicz M, Hildebrandt T, Obersztyn I. Xerostomia of

Various Etiologies: A Review of the Literature. Adv Clin Exp

Med 2016;25:199-206.

8. Ahmad MS, Bhayat A, Zafar MS, Al-Samadani KH. e Impact

of Hyposalivation on Quality of Life (QoL) and Oral Health in

the Aging Population of Al Madinah Al Munawarrah. Int J

Environ Res PublicHealth2017;14(4).pii: E445.

9. Cassolato SF, Turnbull RS. Xerostomia: clinical aspects and

treatment. Gerodontology 2003;20:64-77.

10. Villa A, Nordio F, Gohel A. A risk prediction model for

xerostomia: a retrospective cohort study. Gerodontology

2016;33:562-8.

11. Tan ECK, Lexomboon D, Sandborgh-Englund G, Haasum Y,

Johnell K. Medications at Cause Dry Mouth As an Adverse

Eect in Older People: A Systematic Review and Metaanalysis.

J Am Geriatr Soc 2018;66:76-84.

12. Glore RJ, Spiteri-Staines K, Paleri V. A patient with dry mouth.

Clin Otolaryngol 2009;34:358-63.

13. Villa A, Abati S. Risk factors and symptoms associated with

xerostomia: a cross-sectional study. Aust Dent J 2011;56:290-5.

14. Al-Dwairi Z, Lynch E. Xerostomia in complete denture

wearers: prevalence, clinical findings and impact on oral

functions. Gerodontology 2014;31:49-55.

15. da Silva L, Kupek E, Peres KG. General health inuences episodes

of xerostomia: a prospective population-based study. Community

Dent Oral Epidemiol 2017;45:153-9.

16. Murray omson W, Chalmers JM, John Spencer A, Slade

GD, Carter KD. A longitudinal study of medication exposure

and xerostomia among older people. Gerodontology 2006;23:

205-13.

17. Cho MA, Ko JY, Kim YK, Kho HS. Salivary ow rate and

clinical characteristics of patients with xerostomia according

to its aetiology. J Oral Rehabil 2010;37:185-93.

18. Niklander S, Veas L, Barrera C, Fuentes F, Chiappini G, Marshall

M. Risk factors, hyposalivation and impact of xerostomia on

oral health-related quality of life. Braz Oral Res 2017;31:e14.

19. Chompootweep S, Tankeyoon M, Yamarat K, Poomsuwan P,

Dusitsin N. e menopausal age and climacteric complaints

in ai women in Bangkok. Maturitas 1993;17:63-71.

20. Lee E, Lee YH, Kim W, Kho HS. Self-reported prevalence and

severity of xerostomia and its related conditions in individuals

attending hospital for general health examinations. Int J Oral

Maxillofac Surg 2014;43:498-505.

21. Murray TW,Poulton R, Mark BJ, Al-Kubaisy S. Xerostomia and

medications among 32-year-olds. Acta Odontol Scand 2006;64:

249-54.

22. Delli K, Spijkervet FK, Kroese FG, Bootsma H, Vissink A.

Xerostomia. Monogr Oral Sci 2014;24:109-25.

23. Flink H, Bergdahl M, Tegelberg A, Rosenblad A, Lagerlof

F. Prevalence of hyposalivation in relation to general health,

body mass index and remaining teeth in dierent age groups

of adults. Community Dent Oral Epidemiol 2008;36:523-31.

24. Unruh AM. Gender variations in clinical pain experience.

Pain 1996;65:123-67.

25. Veerabhadrappa SK, Chandrappa PR, Patil S, Roodmal SY,

Kumarswamy A, Chappi MK. Evaluation of Xerostomia in

Dierent Psychological Disorders: An Observational Study. J

Clin Diagn Res 2016;10:ZC24-ZC7.

26. Scully C, Bagan JV. Adverse Drug Reactions in the Orofacial

Region. Crit Rev Oral Biol Med2004;15:221-39.

27. Scully C. Drug eects on salivary glands: dry mouth. Oral Dis

2003;9:165-76.

28. Jeong SY, Kim HW, Lee SW, Ahn BC, Lee J. Salivary gland

function 5 years aer radioactive iodine ablation in patients

with dierentiated thyroid cancer: direct comparison of pre-

and postablation scintigraphies and their relation to xerostomia

symptoms. yroid 2013;23:609-16.

29. Nabaa B, Takahashi K, Sasaki T, Okizaki A, Aburano T. Assessment

of salivary gland dysfunction aer radioiodine therapy for

thyroid carcinoma using non-contrast-enhanced CT: the

signicance of changes in volume and attenuation of the glands.

AJNR Am J Neuroradiol 2012;33:1964-70.

30. Upadhyaya A, Meng Z, Wang P, Zhang G, Jia Q, Tan J, et al.

Eects of rst radioiodine ablation on functions of salivary

glands in patients with dierentiated thyroid cancer. Medicine

(Baltimore) 2017;96:e7164.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

385

Suwimon Wonglaksanapimon, M.D., Panadda Jittawannarat, M.D., Nisa Muangman, M.D.

Department of Radiology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.

Correlation Between CT Findings of Invasive

Pulmonary Aspergillosis and Severity of Neutropenia

ABSTRACT

Objective: e purpose of this study was to investigate the association between CT ndings of invasive pulmonary

aspergillosis and neutrophils level.

Methods: is retrospective study included the patients diagnosed with invasive pulmonary aspergillosis by

computed tomography (CT) at Siriraj Hospital (Bangkok, ailand) during May 2006 to May 2010. e patients

were classied into two groups: group I = neutrophils <500 cells/mm

, and group II = neutrophils ≥500 cells/mm

Patient demographic and clinical data were collected. CT ndings were compared between groups, including

consolidation, pulmonary nodule, pulmonary mass, centrilobular nodule, CT halo sign, CT hypodense, air crescent,

central cavity, lymphadenopathy, and pleural eusion.

Results: Of the 43 patients that were included, 14 patients were assigned to group I, and 29 patients were allocated to

group II. e mean age of patients was 40.2 years, and 48.8% were male. e most common CT nding of invasive

pulmonary aspergillosis was pulmonary nodules (83.7%), followed by CT halo sign (76.7%) and consolidation (69.7%).

e CT nding that showed signicantly more commonly in patients in group I and group II was consolidation

(p=0.02) and central cavity (p=0.03), respectively.

Conclusion: Consolidation was the CT nding signicantly most commonly observed in invasive pulmonary

aspergillosis with neutrophils <500 cell/mm

, and central cavity was the nding signicantly most commonly found

in patients with neutrophils ≥500 cell/mm

with non-specic distribution.

Keywords: Aspergillosis; neutropenia; CT scan; lung; pulmonary (Siriraj Med J 2019; 71: 385-391)

Corresponding author: Suwimon Wonglaksanapimon

E-mail: wpsuwimon@gmail.com

Received 24 September 2018 Revised 9 July 2019 Accepted 7 August 2019

ORCID ID: http://orcid.org/0000-0001-7541-1988

http://dx.doi.org/10.33192/Smj.2019.58

INTRODUCTION

Aspergillus spp. are lamentous fungi that are acquired

via the inhalation of spores. Pulmonary aspergillosis has 4

types, including allergic bronchopulmonary aspergillosis

(ABPA), aspergilloma, semi-invasive pulmonary

aspergillosis, and invasive pulmonary aspergillosis.

Each type of pulmonary aspergillosis depends on host

immunity.

1,2

Invasive pulmonary aspergillosis is a major

complication in immunocompromised patients. Major

risk factors for invasive pulmonary aspergillosis infection

include prolonged neutropenia (<500 cells/mm

for >10

days) or neutrophil dysfunction, post-transplantation,

prolonged (>3 weeks) and high-dose corticosteroid

therapy, hematological malignancy (e.g., leukemia),

cytotoxic therapy, and advanced AIDS.

1,3

Neutrophils are the dominant host defense against

fungi by their action as natural killer cells that are recruited

to the lungs by chemokines. ey aggregate around conidia

in the airways to prevent their germination.

4,5

Neutropenia

for more than 10 days is the most important risk for

invasive pulmonary aspergillosis. In addition, invasive

pulmonary aspergillosis is related to the duration and

Wonglaksanapimon et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

386

Original Article

SMJ

degree of neutropenia. e risk of invasive pulmonary

aspergillosis in neutropenic patients was estimated to

be 1% per day for the rst 3 weeks.

Diagnosis of invasive pulmonary aspergillosis is made

by clinical ndings, physical examination, laboratory

markers (polymerase chain reaction, galactomannan

testing), sputum or bronchoalveolar lavage uid culture,

histopathology, and radiologic ndings. e clinical

presentation of invasive pulmonary aspergillosis is usually

non-specic, such as fever, clinical unresponsive to

antibiotics, chest pain, hemoptysis, and productive cough.

6,7

Early diagnosis of invasive pulmonary aspergillosis is

necessary in these patients because the outcome depends

on early treatment with an appropriate antifungal drug.

Computed tomography (CT) chest imaging is an important

diagnostic tool for early diagnosis.

Chest radiograph is not suciently sensitive for the

diagnosis of invasive pulmonary aspergillosis. Plain chest

radiograph may exhibit non-specic nodular opacities

or air-space inltration in early phase. However, the

air crescent sign and cavitation may become visible on

chest radiographs in later stage.

2,6

oracic CT scan is the most sensitive investigation

for early detection of invasive pulmonary aspergillosis.

oracic CT should be performed in neutropenic patients

with fever that are not responding to antibiotics.

e

gold standard investigation for diagnosis of invasive

pulmonary aspergillosis is histological or pathological

examination of lung tissue from needle aspiration or

biopsy. However, these procedures are invasive, and

they are associated with an increased risk of bleeding.

As a result, thoracic CT scan is considered to be much

more useful method for early diagnosis, and it has a

prognostic impact on the outcome by detection of early

disease.

CT ndings of invasive pulmonary aspergillosis

are variable. Findings that might be observed include

CT halo sign (consisting of a central nodule, mass, or

consolidation surrounded by a zone of hemorrhage that

is found in the early phase of disease), crescentation or

central cavity sign (develops days to weeks aer the initial

presentation, or in the resolving phase of the infection

when neutropenia in the peripheral blood is increasing),

CT hypodense sign (consisting of central low-density

area within nodule or consolidation caused by vascular

obstruction with secondary lung infarction), and non-

specic nodular opacities or air space inltration in the

early stage of infection.

2,8,9

e aim of this study was to identify association

between thoracic CT ndings of invasive pulmonary

aspergillosis and serum neutrophil level.

MATERIALS AND METHODS

Subjects

All patients aged greater than 15 years with diagnosis

of invasive pulmonary aspergillosis who underwent thin-

section thoracic CT scan at the Division of Diagnostic

Radiology, Department of Radiology, Faculty of Medicine

Siriraj Hospital, Mahidol University, Bangkok, ailand

during May 2006 to May 2010 were retrospectively

reviewed. Patients diagnosed with proven, probable, or

possible invasive pulmonary aspergillosis according to

the European Organization for Research and Treatment

of Cancer/Invasive Fungal Infections Cooperative Group

and the National Institute of Allergy and Infectious

Diseases Mycoses Study Group (EORTC/MSG) Consensus

Group were included.

e protocol for this study was

approved by the Siriraj Institutional Review Board (Si

180/2011).

Serum neutrophil level and clinical data (age,

underlying diseases, presenting symptoms, laboratory

investigation, time interval from clinical onset to CT

ndings, and histological or pathological report) of all

individuals were collected from our center’s electronic

medical record database. Regarding serum neutrophil level,

patients were classied into either group I (neutrophils

<500 cells/mm

) or group II (neutrophils ≥500 cells/

Imaging and imaging interpretation

CT examinations were obtained with a 64-slice multi-

detector row CT (GE or Siemens). e GE parameters

were 120 kVp, 250 mA, and pitch = 1.375 craniocaudal,

and the Siemens parameters were 120 kVp, 250 mA, and

pitch = 0.8 craniocaudal. Spiral CT scan was obtained

during full inspiration with or without contrast media

administration.

e CT ndings were assessed by two radiologists,

and conclusions were reached by consensus. e pattern

of pulmonary abnormality ndings, associated ndings,

and distribution was analyzed. e pulmonary abnormality

ndings were consolidation (area of increase in pulmonary

parenchymal attenuation that obscures the margin of vessels

and airway wall), pulmonary nodule (round opacity lesion,

≤3 cm in diameter), pulmonary mass (round pulmonary

opacity >3 cm in diameter), and centrilobular nodule

(small opacity in the center of a secondary pulmonary

lobule).

e associated ndings were “CT halo sign”

(ground-glass attenuation surrounding a pulmonary

nodule or consolidation) (Fig 1), CT hypodense sign

(the dierences between lung opacity in the peripheral

zone of the lesion and lower density of necrotic central

area >15 HU in mediastinal window setting) (Fig 2), air

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

387

Fig 1. A 40-year-old male with underlying acute aplastic anemia and

neutrophils 28 cells/mm

presented with fever and cough. Axial

thoracic CT scan demonstrated pulmonary consolidation surrounded

with ground-glass opacity at right upper lobe that represented CT

halo sign. e nal diagnosis was invasive pulmonary aspergillosis.

Fig 2. A 32-year-old male with underlying acute myeloblastic

leukemia and neutrophils 1,761 cells/mm

presented with fever. (a,

b) axial unenhanced thoracic CT scan demonstrated two consolidation

areas at le upper lobe. One of them showed dierences in lung

opacity in the peripheral zone of the lesion (43.3 HU in a), and lower

density of necrotic central areas (6.7 HU in b) that represented CT

hypodense sign. (c) axial post-contrast-enhanced thoracic CT conrmed

CT hypodense sign (arrow). e nal diagnosis was invasive pulmonary

aspergillosis.

crescent sign (crescent shaped area of air density within

pulmonary nodule or mass) (Fig 3), central cavity (a

lucent area of air density within an area of consolidation,

nodule, or mass) (Fig 4), lymphadenopathy, and pleural

eusion.

Fig 3. A 27-year-old female with underlying acute myeloblastic

leukemia and neutrophils 3,171 cells/mm

presented with fever.

Axial thoracic CT scan demonstrated a crescent shaped area within

an area of consolidation that represented air crescent sign at right

upper lobe. Ground-glass nodules at le upper lobe were also noted.

e nal diagnosis was invasive pulmonary aspergillosis.

Fig 4. A 53-year-old male with underlying acute myeloblastic leukemia

and neutrophils 1,159 cells/mm

with history of fever and cough.

Axial thoracic CT scan demonstrated a lucent area within nodule

that represented central cavity at right upper lobe. e nal diagnosis

was invasive pulmonary aspergillosis.

Statistical analysis

e age of patients and duration from imaging to

CBC examination are expressed as mean. All qualitative

data, including gender, dierent criteria for diagnosis

of invasive pulmonary aspergillosis, and CT ndings,

are described as frequency and percentage. Chi-square

test was used to analyze the radiographic ndings, and

to examine for correlation between 2 groups of CT

ndings and neutrophil levels. A p-value less than 0.05

was considered to be statistically signicant. Statistical

analysis was performed using a statistical soware package

(SPSS Statistics version 13.0; SPSS, Inc., Chicago, IL, USA).

Wonglaksanapimon et al.

(c)

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

388

Original Article

SMJ

RESULTS

Patient characteristics

e characteristics of 43 patients with diagnosis

of invasive pulmonary aspergillosis by proven (6.9%),

probable (20.9%), and possible (72.0%) criteria are

summarized in Table 1. Patients were classied into 2

groups. Fourteen patients (32.6%) were classied in group

I (serum neutrophil level <500 cell/mm

), and 29 patients

(67.4%) were classied in group II (serum neutrophil

level ≥500 cell/mm

). e mean age of patients in group

I (40.7 years) was not signicantly dierent from that

of group II (39.9 years). e gender distribution was

similar in each of the 2 groups (group I: 7 males and

7 females vs. group II: 14 males and 15 females). e

mean time interval from the date of thoracic CT scan

to CBC was 0.7 days in group I, and 0.6 days in group

II (p>0.05). e mean time interval from clinical onset

to CT ndings was 9.9 days in group I, and 13.2 days

in group II (p>0.05). e underlying diseases in group

I were aplastic anemia (21.4%), acute lymphoblastic

leukemia (21.4%), acute myeloid leukemia (35.7%), chronic

myeloid leukemia (14.3%), and unspecied neutropenia

(7.1%). e underlying diseases in group II were acute

lymphoblastic leukemia (20.7%), acute myeloid leukemia

(58.6%), chronic myeloid leukemia (3.4%), lymphoma

(13.8%), and systemic lupus erythematosus (3.4%).

CT protocol

ere were 38 CT studies (87.7%) performed with

contrast medium administration, and 5 studies (12.3%)

performed without contrast medium administration.

Regarding slice thickness of the CT scan, 0.6 mm was

used in 5 studies (10.8%), 0.63 mm in 7 studies (16.9%),

1.25 mm in 16 studies (36.9%), 1.5 mm in 13 studies

(30.8%), and 4 mm in 2 studies (4.6%).

Imaging interpretation

CT ndings of invasive pulmonary aspergillosis

and associated ndings were assessed for the following

radiological patterns; consolidation, pulmonary nodule,

pulmonary mass, centrilobular nodule, CT halo

sign, CT hypodense sign, air crescent, central cavity,

lymphadenopathy, and pleural eusion. e prevalence

of each of these nding was compared between groups

(Table 2).

TABLE 1. Patient demographic and clinical characteristics.

Characteristics Neutrophils <500 cells/mm

 Neutrophils≥500cells/mm

Total

Number, n (%) 14 (32.6%) 29 (67.4%) (N=43)

Age (mean) 40.7 years 39.9 years 40.2 years

Gender, n (%)

Male 7 14 21 (48.8%)

Female 7 15 22 (51.2%)

Criteria diagnosis, n (%)

Proven IPA 1 (7.1%) 2 (6.9%) 3 (7.0%)

Probable IPA 4 (28.6%) 5 (17.2%) 9 (20.9%)

Possible IPA 9 (64.3%) 22 (75.9%) 31 (72.1%)

Duration from imaging to CBC (mean) 0.7 days 0.6 days 0.7 days

Underlying conditions, n (%)

Aplastic anemia 3 (21.4%) 0 (0.0%) 3 (7.0%)

ALL 3 (21.4%) 6 (20.7%) 9 (20.9%)

AML 5 (35.7%) 17 (58.6%) 22 (51.2%)

CML 2 (14.3%) 1 (3.4%) 3 (7.0%)

Lymphoma 0 (0.0%) 4 (13.8%) 4 (9.3%)

SLE 0 (0.0%) 1 (3.4%) 1 (2.3%)

 Unspeciedneutropenia 1(7.1%) 0(0.0%) 1(2.3%)

Abbreviations: IPA=invasive pulmonary aspergillosis; CBC=complete blood count; ALL=acute lymphoblastic leukemia; AML=acute

myeloid leukemia; CML=chronic myeloid leukemia, SLE=systemic lupus erythematosus

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

389

ere was no signicant dierence in the pattern of

pulmonary nodule (78.6% in group I vs. 86.2% in group

II, p=0.52), pulmonary mass (0% in group I vs. 3.4% in

group II, p=0.48), centrilobular nodule (7.1% in group

I vs. 6.9% in group II, p=0.97), CT halo sign (69.9% in

group I vs. 69.0% in group II, p=0.08), CT hypodense

sign (35.7% in group I vs. 44.8% in group II, p=0.57),

air crescent sign (7.1% in group I vs. 27.6% in group II,

p=0.12), lymphadenopathy (11.5% in group I vs. 13.8%

in group II, p=0.96) and pleural eusion (71.4% in group

I vs. 41.4% in group II, p=0.96), as shown in Table 2.

Consolidation was the CT nding signicantly more

commonly found in group I than in group II (92.9%

vs. 58.6%, respectively; p=0.02). Central cavity was the

CT nding signicantly more commonly observed in

group II than in group I (48.3% vs. 14.3%, respectively;

p=0.03).

e distribution of lesions was, as follows: right

upper lobe (RUL) (83.7%), le upper lobe (LUL) (72.1%),

right lower lobe (RLL) (69.8%), le lower lobe (LLL)

(67.4%), and right middle lobe (RML) (48.8%).

DISCUSSION

Invasive pulmonary aspergillosis is a major cause of

mortality in immunocompromised patients. Neutrophils

and alveolar macrophages play important roles in killing

conidia and developing hyphae. If abnormalities exist in

these defense mechanisms, the conidia undergo germination

and turn to invasive hyphae. As such, neutropenia status is

a major risk factor for invasive pulmonary aspergillosis.

2,5

e clinical signs and symptoms of invasive pulmonary

aspergillosis include fever, unresponsiveness to antibiotics,

productive cough, dyspnea, haemoptysis, and pleuritic chest

pain. Diagnosis of invasive pulmonary aspergillosis can

be made by combining clinical, radiographic, laboratory,

and histopathologic data. oracic CT scan has become

an important tool for diagnosing invasive pulmonary

aspergillosis due its non-invasive nature and its high

sensitivity (>80%) and specicity (60-98%).

Pathologically, Aspergillus spp. invades pulmonary

blood vessels and cause hemorrhagic infarction.

Characteristic early lesions consist of central areas of

necrosis surrounded by a zone of hemorrhage, which

TABLE 2. Computed tomography (CT) ndings and associated ndings of invasive pulmonary aspergillosis

compared between neutrophil levels.

CTndings Neutrophils<500 Neutrophils≥500 Total P-value

cells/mm

Consolidation 13 (92.9%) 17 (58.6%) 30 (69.7%) 0.02

Nodule 11 (78.6%) 25 (86.2%) 36 (83.7%) 0.52

Mass 0 (0.0%) 1 (3.4%) 1 (2.3%) 0.48

Centrilobular nodule 1 (7.1%) 2 (6.9%) 3 (6.9%) 0.97

Associatedndings Neutrophils<500 Neutrophils≥500 Total P-value

cells/mm

Peripheral GGO (halo sign) 13 (69.9%) 20 (69.0%) 33 (76.7%) 0.08

CT hypodense sign 5 (35.7%) 13 (44.8%) 18 (41.8%) 0.57

Air crescent sign 1 (7.1%) 8 (27.6%) 9 (37.2%) 0.12

Central cavity 2 (14.3%) 14 (48.3%) 16 (37.2%) 0.03

Lymphadenopathy 3 (11.5%) 4 (13.8%) 7 (16.2%) 0.96

Pleural effusion 10 (71.4%) 12 (41.4%) 22 (51.1%) 0.06

A p-value <0.05 indicates statistical signicance

Abbreviation: GGO=ground glass opacication

Wonglaksanapimon et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

390

Original Article

SMJ

corresponds with the CT halo sign.

11,12

e process of

cavitation characteristically results in the air crescent sign,

which histologically represents necrotic lung mixed with

hyphae and granulocytic inltrate that is surrounded by

a thin rim of air.

2,11,13

Hyung Jin Won, et al. described CT ndings in 5 patients

with neutropenia (<500 cells/mm

) and proven invasive

pulmonary aspergillosis. Most common manifestation

of invasive pulmonary aspergillosis was segmental area

of consolidation with ground-glass appearance (i.e., CT

halo sign) (n=4, 20%) and at least one nodule surrounded

by a halo (n=2, 40%).

Kuhlman, et al. found the typical

CT ndings of invasive pulmonary aspergillosis to be

multiple nodules or uy masses or consolidation with

surrounding halos of low attenuation. Some form of

cavitation or air crescent formation was found at or near

the time of granulocyte recovery.

Horger, et al. found

the CT hypodense sign in 13 of 43 patients (30.2%) with

invasive pulmonary aspergillosis. Eleven patients (84.6%)

had neutrophils <500 cells/mm

and 2 patients (15.4%)

had neutrophils ≥500 cells/mm

, and the CT hypodense

sign was more common in patients with neutrophils

<500 cells/mm

. e CT hypodense sign showed an area

of central necrosis and proved to be a precursor of the

crescent sign within a range of 2-19 days.

Logan, et al.

assessed the CT ndings and pathological ndings in

invasive aspergillosis of the airways in 9 patients. eir

result showed CT ndings in patients with neutrophil level

<500 cells/mm

(n=3) to be peribronchial consolidation,

CT halo sign, and centrilobular micronodules. e CT

ndings in patients with neutrophil level ≥500 cells/

(n=6) were centrilobular micronodules, diuse

bronchiectasis, peribronchial consolidation, and diuse

ground-glass attenuation.

However, no previous study

reported signicant dierence in the CT ndings of

invasive pulmonary aspergillosis patients compared

between those with neutrophil level <500 cells/mm

and those with neutrophil level ≥500 cells/mm

e present study found the most common CT nding

of invasive pulmonary aspergillosis to be pulmonary

nodules, followed by CT halo sign and consolidation. We

classied patients into 2 groups (group I = neutrophil

level <500 cells/mm

, and group II = neutrophil level ≥500

cells/mm

) and compared ndings between groups. We

found consolidation to be signicantly more commonly

observed in patients with neutrophil level <500 cell/mm

is is because the extent and severity of infection in

patients depends on the degree of immunodeciency

and duration of neutropenia.

Invading fungi are more

expanded in individuals with profound granulocytopenia,

which results in a greater extent of infected infarction

(consolidation). Central cavity was found signicantly

more oen in patients with neutrophil level ≥500 cells/

. is might be explained by the cavitating process

of fungal nodules, which relies on granulocytic response.

Fungal nodules in patients with granulocytic function

defect or in patients who remain neutropenic do not

cavitate.

2,15

e most common CT nding of invasive pulmonary

aspergillosis in the present study was presence of pulmonary

nodule (36 of 43 cases [83.7%]), and the second most

common nding was CT halo sign (33 of 43 cases [76.7%]).

However, neither of these ndings were statistically

signicantly dierent between groups. e other CT

ndings that showed no signicant dierence between

groups were centrilobular nodule, pleural effusion,

pulmonary mass, CT hypodense sign, air crescent sign,

and lymphadenopathy. ese lesions in this study showed

no obvious prominence in any pulmonary lobes, which

indicates non-specic distribution.

Limitations

is study has some mentionable limitations. First,

the retrospective nature of our study made it dicult to

control factors, such as CT protocol (with or without

contrast, slice thickness) and duration from date of CT

scan to date of CBC examination. Second, there were

only 3 proven cases of invasive pulmonary aspergillosis

by histology or cytopathology. is can be explained

by the fact needle aspiration and lung biopsy are both

invasive procedures that are associated with increased

risk of bleeding in patients with a hematologic problem.

ird, the size of our study population was relatively small

and there was substantial asymmetry between groups.

It is possible that these factors could have adversely

inuenced the results of our statistical analyses.

CONCLUSION

Consolidation was the CT nding signicantly most

commonly observed in invasive pulmonary aspergillosis

with neutrophils <500 cell/mm

, and central cavity was the

nding signicantly most commonly found in patients with

neutrophils ≥500 cell/mm

with non-specic distribution.

REFERENCES

1. Zmeili OS, Soubani AO. Pulmonary aspergillosis: a clinical

update. QJM 2007;100:317-34.

2. ompson BH, Stanford W, Galvin JR, Kuribara Y. Varied

radiologic appearances of pulmonary aspergillosis. Radiographics

1995;15:1273-84.

3. Pauw BD, Walsh TJ, Donnelly JP, Stevens DA, Edwards JE,

Calandra T, et al. Revised denitions of invasive fungal disease

from the European organization for research and treatment of

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

391

cancer/invasive fungal infections cooperative group and the

national institute of allergy and infectious diseases mycoses study

group (EORTC/MSG) consensus group. Clin Infect Dis

2008;46:1813-21.

4. Segal BH. Aspergillosis. N Engl J Med 2009;360:1870-84.

5. Ben-Ami R, Lewis RE, Kontoyiannis DP. Enemy of the

(immunosuppressed) state: an update on the pathogenesis of

Aspergillus fumigatus infection. Br J Haematol 2010; 150:

406-17.

6. Reichenberger F, Habicht JM, Gratwohl A, Tamm M. Diagnosis

and treatment of invasive pulmonary aspergillosis in neutropenic

patients. Eur Respir J 2002;19:743-55.

7. Cordonnier C, Botterel F, Amor RB, Pautas C, Maury S,

Kuentz M, et al. Correlation between galactomannan antigen

levels in serum and neutrophil counts in haematological patients

with invasive aspergillosis. Clin Microbiol Infect 2009;15:

81-86.

8. Horger M, Einsele H, Schumacher U, Wehrmann M, Hebart

H, Lengerke C, et al. Invasive pulmonary aspergillosis: frequency

and meaning of the “hypodense sign” on unenhanced CT. Br

J Radiol 2005;78:697-703.

9. Won HJ, Lee KS, Cheon JE, Hwang JH, Kim TS, Lee HG,

et al. Invasive pulmonary aspergillosis: prediction at thin-section

CT in patients with neutropenia--a prospective study. Radiology

1998;208:777-82.

10. Hansell DM, Bankier AA, MacMahon H, McLoud TC, Muller

NL, Remy J. Fleischner society: glossary of terms for thoracic

imaging. Radiology 2008;246:697-719.

11. Kuhlman JE, Fishman EK, Burch PA, Karp JE, Zerhouni EA,

Siegelman SS. CT of invasive pulmonary aspergillosis. AJR

Am J Roentgenol 1988;150:1015-20.

12. Kuhlman JE, Fishman EK, Siegelman SS. Invasive pulmonary

aspergillosis in acute leukemia: characteristic ndings on CT,

the CT halo sign, and the role of CT in early diagnosis. Radiology

1985;157:611-4.

13. Shibuya K, Ando T, Hasegawa C, Wakayama M, Hamatani S,

Hatori T, et al. Pathophysiology of pulmonary aspergillosis. J

Infect Chemother 2004;10:138-45.

14. Logan PM, Primack SL, Miller RR, Muller NL. Invasive aspergillosis

of the airways: radiographic, CT, and pathologic ndings.

Radiology 1994;193:383-8.

15. Geer WB, Albelda SM, Talbot GH, Gerson SL, Cassileth PA,

Miller WT. Invasive pulmonary aspergillosis and acute leukemia:

limitations in the diagnostic utility of the air crescent sign.

Radiology 1985;157:605-10.

Wonglaksanapimon et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

392

Original Article

SMJ

Phanitra Maneeratprasert, M.D., Pavit Sutchritpongsa, M.D., Dittakarn Boriboonhirunsarn, M.D., M.P.H., Ph.D.

Department of Obstetrics and Gynecology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.

Success Rate of Cervical Cerclage at Siriraj

Hospital

ABSTRACT

Objective: To determine the success rate of cervical cerclage at Siriraj Hospital and the associated factors.

Methods: e study included 90 pregnant women who received cervical cerclage procedures at Siriraj Hospital

during 2007-2016. Clinical information and the cerclage procedures were evaluated. Successful cervical cerclage

was dened as delivery at 34 weeks’ gestation or more.

Results: e mean maternal age was 31.8 ± 5 years. e most common indication was history of second trimester

abortion (76.7%). e mean gestational age (GA) at cervical cerclage was 18.0 ± 4.3 weeks. e mean cervical

length was 25.2 ± 12.0 mm. Most of the patients had no cervical dilatation (76.7%). Almost all the cervical cerclage

procedures were performed using McDonald’s technique (98.9%). Of the 90 women, 66 (73.3%) delivered at >34

weeks’ gestation, while 59.9% delivered at >37 weeks’ gestation. e mean birth weight was 2404.9 ± 991.7 grams.

e success rate of cervical cerclage increased signicantly among women with greater cervical length (28.2±10.5

vs. 18.0±14.9 mm., p=0.002), without cervical dilatation (83.3% vs. 58.3%, p=0.046), without bulging of membranes

(92.5% vs. 66.7%, p=0.002), and prophylactic operations (89.4% vs. 58.3%, p=0.001). Maternal complications were

signicantly lower in women with successful cervical cerclage (9.1% vs. 45.8%, p<0.001).

Conclusion: e success rate of cervical cerclage at Siriraj Hospital was 73.3%. Possible associated factors included

cervical length, cervical dilatation, bulging of membranes, and the prophylactic procedure.

Keywords: Cervical cerclage; success rate (Siriraj Med J 2019; 71: 392-398)

Corresponding author: Phanitra Maneeratprasert

E-mail: phanitra_numphon@hotmail.com

Received 10 October 2018 Revised 22 April 2019 Accepted 14 May 2019

ORCID ID: http://orcid.org/0000-0002-2991-9470

http://dx.doi.org/10.33192/Smj.2019.59

INTRODUCTION

Preterm delivery is the most signicant obstetric

problem worldwide and is a major cause of death aer birth.

Surviving newborns have high morbidity and mortality

rates, both short-term and long-term complications such

as neurological developmental abnormalities, respiratory

distress, and visual and hearing problems. e issue incurs

signicant costs for healthcare services. In previous study,

the incidence of preterm births in developed countries

was found to be 5-7%, but this was higher in developing

countries.

In ailand, the incidence of preterm births

was 12%,

similar to the rate at Siriraj Hospital (12.89%).

Cervical insuciency is a risk factor for preterm

labour and is defined as the inability of the uterine

cervix to retain a pregnancy in the absence of signs and

symptoms of clinical contraction, labour, or both in the

second trimester.

e incidence of cervical insuciency

is 0.12 - 2.0% in normal pregnancies, but 15% in habitual

abortions.

e cause of cervical insuciency is still

unknown, although it is suspected to be related to cervical

operative procedures such as conization, dilatation, and

curettage. ere is no criteria for diagnosis of cervical

insuciency. e diagnosis is dependent on cervical

progression in the second trimester in the absence of

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

393

labour pain or other causes such as premature rupture of

membrane, bleeding from vagina, or infection. Transvaginal

ultrasonography for cervical length measurement is now

used to co-evaluate this condition. However, diagnosis of

cervical insuciency is not only based on short cervical

length. Surgical and non-surgical procedures can be used

to treat cervical insuciency. For non-surgical treatments,

physical activity restrictions, bedrest, and abstinence from

sexual intercourse have been proposed but without strong

supporting evidence.

Surgical treatment is in the form of

the cervical cerclage procedure, which can be performed

either transvaginally or transabdominally. McDonald

and Shirodkar’s techniques are considered standard

cervical cerclage techniques, and there is currently no

signicant data to prove which technique is superior.

7-9

Transabdominal cervical cerclage may be used in cases

of failed transvaginal cervical cerclage or if there is a

history of trachelectomy.

The American College of Obstetricians and

Gynecologists (ACOG) provide a number of recommended

indications of cervical cerclage.

e rst is dependent upon

history (history of unexplained second trimester delivery

without labour or abruptio placenta). Second, a physical

examination (painless cervical dilation in the second

trimester). While the third is based on ultrasonographic

information (short cervical length <25 mm.) with a history

of prior preterm birth. A previous study reported that

cervical cerclage in women with cervical incompetence

signicantly decreased the incidence of preterm delivery

prior to 33 weeks’ gestation.

Limited data is available for

rescue cervical cerclage and the success rate is unpredictable.

Prior observational studies suggest that rescue cervical

cerclage may be able to improve the neonatal outcomes,

although there is insucient data.

At Siriraj Hospital, cervical cerclage is used as a

surgical treatment for women with cervical incompetence.

Yet there is limited clinical information about the success

rate of cervical cerclage and neonatal outcomes. e

primary objective of this study was subsequently to

assess the success rate of cervical cerclage procedures at

Siriraj Hospital. Meanwhile, pregnancy outcomes and

potential associated factors were also evaluated.

MATERIALS AND METHODS

Aer approval from the Siriraj Institutional Review

Board (Si 555/2017), this retrospective descriptive study

was conducted at the Department of Obstetrics and

Gynaecology, Faculty of Medicine Siriraj Hospital. e

sample size was calculated based on an estimated success

rate of 70%. At a 95% condence level and with a 10%

acceptability error, a sample size of at least 81 cases were

required. A total of 90 pregnant women who had undergone

a cervical cerclage procedure at the hospital based on

any indications between 2007 and 2016 were included

in this study. At Siriraj Hospital, the cervical cerclage

procedure is performed based on the recommendations of

the American College of Obstetricians and Gynecologists.

This study evaluated the clinical information,

characteristics of the cerclage procedure, and pregnancy

outcomes. Women with detected fetal anomalies, multifetal

pregnancy, or who had terminated their pregnancies

due to maternal or fatal indications were excluded. Data

was obtained from medical records, which included the

baseline clinical information, characteristics of the cerclage

procedures, and the pregnancy outcomes. Successful

cervical cerclage was dened as delivery at > 34 weeks’

gestation. Non-viable delivery before 24 weeks’ gestation

was dened as abortion.

Aer performing the cervical cerclage procedure,

all the patients were admitted for medical sta to closely

monitor uterine contraction. If uterine contractions were

detected, a tocolytic drug was administered intravenously.

Daily vaginal progesterone (Utrogestan 200 mg) was

given in some cases until 36 weeks’ gestation. An oral

nonsteroidal anti-inammatory drug (ibuprofen 800-1,200

mg per day) was given three days aer the operation.

In most cases, an intravenous antibiotic (Cefoxitin 1g)

was administered intraoperation. And then an oral

cephalosporin (Cephalexin 2g per day) was administered

for seven days. Post-operative complications such as

uterine contraction, vaginal bleeding, cervical laceration,

rupture of membrane, and chorioamnionitis were recorded

among the study sample population.

Descriptive statistics, including the mean, median,

standard deviation, number, and percentage were used

to describe the various characteristics as appropriate.

Comparisons between the groups were performed using

the Student’s t-test and the chi square test as appropriate.

P values less than 0.05 were considered statistically

signicant.

RESULTS

A total of 90 pregnant women who had undergone

a cervical cerclage procedure were included in this study.

Table 1 shows the patient’s baseline characteristics.

e mean maternal age was 31.8 ± 5 years, while the

majority of the patients were nulliparous (61.1%), and

88.9% had previously undergone at least one abortion.

A history of second trimester abortion was found in 76.7%.

Meanwhile, 7.8 % had a history of preterm delivery and

15.6% had previous cervical cerclage. Cervical cerclage

was indicated by a physical examination in 18.9% of the

Maneeratprasert et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

394

Original Article

SMJ

women, while an ultrasonographic examination was

used to indicate cervical cerclage in 34.4% of the women.

Among these women, 9 cases of cervical cerclage were

indicated by both history and a physical examination.

Finally, 26 cases had cervical cerclage indicated by both

history and ultrasonographic examination.

e characteristics of the cerclage procedure is

demonstrated in Table 2. e mean gestational age at

the cervical cerclage procedure was 18.0 ± 4.3 weeks

and the mean cervical length was 25.2 ± 12.0 mm.

Most of the patients had no cervical dilatation (76.7%),

while 6.7% had a cervical dilatation of >3 cm. Cervical

cerclage was performed as a prophylactic procedure in

81.1% of the women, while 18.9% underwent rescue

operations. Almost all the cervical cerclage procedures

were performed using McDonald’s technique, while only a

single woman underwent the Shirodkar technique following

the cervical intraepithelial neoplasia III status post loop

electrosurgical excision procedure (CIN III S/P LEEP).

Antibiotic prophylaxis was used during the procedure

in 91.1% of the cases, and progesterone supplement

was used in 50% of the cases. Vaginal progesterone

was administered as a supplement aer the cerclage

procedure until 36 weeks’ gestation.

TABLE 1. Baseline characteristics of the patients (N=90).

Characteristics Mean ± SD

Mean age ± SD (year) 31.8 ± 5.0

N (%)

Parity

0 55 (61.1)

1 30 (33.3)

2 5 (5.6)

Abortion

0 10 (11.1)

1 26 (28.9)

 ≥2 54(60)

Underlying disease

No 70 (77.8)

Diabetes mellitus 5 (5.6)

Hypertension 4 (4.4)

Cervical pathology* 3 (3.3)

Others** 8 (8.9)

History-indicated cerclage

History of abortion in 2

trimester 69 (76.7)

History of preterm delivery 7 (7.8)

Previous cervical cerclage 14 (15.6)

Physical examination-indicated cerclage 17 (18.9)

Ultrasound-indicated cerclage 31 (34.4)

Previous cervical operation 27 (30)

No 63 (70)

Dilatation and curettage 24 (26.7)

LEEP*** 3 (3.3)

* Cervical pathology: Cervical intraepithelial neoplasia, **Others: asthma, hyperthyroid, chronic hepatitis B, mitral stenosis , ***LEEP =

Loop electrosurgical procedure

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

395

TABLE 2. Characteristics of cerclage procedure (N=90).

Characteristics Mean ± SD

Mean gestational age at cerclage (weeks) 18.0 ± 4.3

Mean cervical length (mm) (N=70) 25.2 ± 12.0

N (%)

Cervical dilatation at cerclage

No dilatation 69 (76.7)

1-2 cm 15 (16.7)

 ≥3cm 6(6.7)

Bulging of membranes 13 (14.4)

Type of cervical cerclage

Prophylactic procedure 73 (81.1)

Rescue procedure 17 (18.9)

Antibiotic prophylaxis 82 (91.1)

Progesterone treatment 45 (50.0)

Table 3 shows the pregnancy outcomes of the women

in this study. e mean gestational age (GA) at delivery

was 33.9 ± 6.2 weeks. Meanwhile, 66 (73.3%) of the

women delivered at > 34 weeks’ gestation and 58.9%

delivered at > 37 weeks’ gestation. All newborns in this

group survived and had good outcomes, with APGAR

scores > 7 at 1 and 5 minutes aer birth, meanwhile

there were no NICU admissions. e median interval

from cervical cerclage to delivery was 125 days. e

mean birth weight was 2404.9 ± 991.7 g. Among the

24 women who delivered before 34 weeks’ gestation,

12 had abortions, 2 were stillbirth, and 10 of this group

of newborns survived. e lowest GA of the surviving

newborns was 27 weeks with a birth weight of 810 g. In

the prophylactic cervical cerclage group, 59 out of the

73 women (80.8%) delivered at > 34 weeks’ gestation,

while 64.3% delivered at > 37 weeks’ gestation. In the

rescue cervical cerclage group, 7 out of the 17 (41.2%)

women delivered at > 34 weeks’ gestation and 35.3%

delivered at >37 weeks’ gestation. Term newborns who

were delivered from a rescue operation had cervical

dilatation varying between 2-3 cm. e mean interval

from cervical cerclage to delivery was very dierent

between the prophylactic and the rescue groups. In the

prophylactic group, the median interval was 143 days

whereas it was 60 days for the rescue group.

Table 4 displays a comparison between the various

characteristics of women who delivered at ≥ 34 and < 34

weeks’ gestation. ere were no signicant dierences

between the two groups in terms of age and gestational

age at cerclage. e mean cervical length in the women

who delivered at ≥ 34 weeks’ gestation was signicantly

higher than those who delivered at <34 weeks’ gestation

(28.2 ± 10.5 vs. 18.0 ± 14.9 mm., p=0.002). Cervical

dilatation was signicantly less among those who delivered

at ≥ 34 weeks’ gestation, with 83.3% having no cervical

dilatation compared to only 58.3% with cervical dilation

in the other group (p=0.046), while bulging of membranes

was also signicantly less common in those delivered

at ≥34 weeks’ gestation (7.6% vs. 33.3%, p=0.002). e

prophylactic cerclage procedure was signicantly more

common in those delivered at ≥34 weeks’ gestation

(89.4% vs. 58.3%, p=0.001). ere were no signicant

dierences found between the use of antibiotic prophylaxis

and progesterone supplementation. Finally, maternal

complications were signicantly higher among those

delivered at <34 weeks’ gestation, including uterine

contractions, vaginal bleeding, rupture of membrane

and other complications (p<0.001).

DISCUSSION

Cervical cerclage is a surgical treatment used for

patients with cervical incompetence which aims to reduce

the risks of miscarriage and preterm birth. Cervical cerclage

procedures are performed to increase the strengthening

of the cervix either to maintain pregnancy until term

Maneeratprasert et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

396

Original Article

SMJ

TABLE 3. Pregnancy outcomes of the women in the study (N=90).

*Others: cervical tear, chorioamnionitis

Pregnancy outcome Mean ± SD

Gestational age at delivery (weeks) 33.9 ± 6.2

N (%)

Gestational age at delivery

<24 weeks 12 (13.3)

24-27

weeks 5 (5.6)

28-33

weeks 7 (7.8)

34-36

weeks 13 (14.4)

 ≥37weeks 53(58.9)

Mean birth weight (grams) (N=87) 2404.9 ± 991.7

< 1,000 13 (14.9)

1,000-1,999 13 (14.9)

2,000-2,999 33 (37.9)

3,000-3,999 28 (32.1)

Maternal complication

Uterine contraction 11 (12.2)

Vaginal bleeding 2 (2.2)

Rupture of membrane 2 (2.2)

Others** 2 (2.2)

Median (IQR)

Interval from cerclage to delivery (days) 125 (63.8, 155.8)

or to prolong the pregnancy for as long as possible.

Based on previous studies,

12,13

there is limited clinical

information about the procedures’ ecacy and success

rate, as well as regarding which patient groups benet

from the operation and the neonatal outcomes. e

overall success rate in both the prophylaxis and rescue

cervical cerclage groups in this study was 73.3% and

there were good neonatal outcomes. e mean cervical

length in the successful group (delivery at GA >34 weeks)

was signicantly higher than the other group. Cervical

dilatation and bulging of membranes were signicantly

less in the successful group, similar to previous ndings.

11,14

ere were no signicant dierences between the cervical

cerclage techniques, the use of antibiotic prophylaxis, and

progesterone supplementation. Many studies support a

similar ecacy between the McDonald and Shirodkar

techniques.

7,9,15

McDonald’s technique is found to be

preferable since it is easier to place and remove the

sutures. From a previous study, the addition of vaginal

progesterone in the rescue cervical cerclage procedure

was associated with a reduction of spontaneous preterm

births.

Maternal complications were signicantly higher

among those delivered at <34 weeks’ gestation, including

uterine contractions, vaginal bleeding, and rupture of

membranes.

Prophylactic cervical cerclage was performed in

history-indicated and ultrasound-indicated patients.

e results show that this group’s success rate was high

(80.8%) and with low levels of complications. For the

rescue cervical cerclage procedure, the success rate was

difficult to predict and usually had poor pregnancy

outcomes. Patients who had a cervical dilatation exceeding

4 cm with bulging of fetal membrane into the vagina had

a higher chance of cerclage failure.

In this study, the

success rate in the rescue cervical cerclage procedure was

only 41.2%. ere was a case report of rescue cervical

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

397

TABLE 4. Comparison of characteristics between women delivered ≥ 34 and <34 weeks’ gestation (N=90).

Characteristics Deliveryat<34weeks Deliveryat≥34weeks Pvalue

N=24 N=66

Mean ± SD Mean ± SD

Age (years) 31.5 ± 5.8 31.9 ± 4.4 0.769

Gestational age at cerclage (weeks) 18.6 ± 4.2 17.7 ± 4.4 0.422

Cervical length (mm) (N=70) 18.0 ± 14.9(N=17) 28.2 ± 10.5(N=53) 0.002

N (%) N (%)

Cervical dilatation 0.046

No dilatation 14 (58.3%) 55 (83.3%)

1-2 cm 7 (29.2%) 8 (12.1%)

>3 cm 3 (12.5%) 3 (4.5%)

Bulging of membranes 8 (33.3%) 5 (7.6%) 0.002

Type of cervical cerclage 0.001

Prophylactic 14 (58.3%) 59 (89.4%)

Rescue 10 (41.7%) 7 (10.6%)

Progesterone treatment 9 (37.5%) 36 (54.5%) 0.153

Antibiotic prophylaxis 21 (87.5%) 61 (92.4%) 0.435

Maternal complications 11 (45.8%) 6 (9.1%) <0.001

Uterine contraction 6 (25%) 5 (7.6%)

Vaginal bleeding 1 (4.2%) 1 (1.5%)

Rupture of membrane 2 (8.3%) 0

Others* 2 (8.3%) 0

*Others: cervical tear, chorioamnionitis

cerclage in a patient who had 7 cm. of cervical dilatation

and bulging of membranes, and delivered at 33 weeks

and 4 days of gestation with good neonatal outcomes.

e success rate of the prophylactic group was higher

than the rescue group, similar to previous studies (78%

vs 53%).

18,19

For clinical applications, e success rate can

apply to counsel the pregnant women who has cervical

incompetence and plan to do the cervical cerclage. Early

identication of women at risk of cervical incompetence

and early treatment will improve outcomes.

A strength of the present study is that almost all

of the cervical cerclage procedures were performed by a

single experienced obstetrician, meaning that there was

no variation in the surgeon skill or technique.

Nonetheless, this study has a number of limitations.

e recommendations for cervical cerclage was changed

aer receiving new research data.

4,20

History indicated

that cervical cerclage had changed from three or more

second trimester pregnancy losses to only one or more

previous second trimester pregnancy losses. e data

used in this study used the previous criteria to diagnose

cervical incompetence in some cases, meaning that some

cases of cervical incompetence may have been missed.

At present, transvaginal ultrasounds are available in all

hospitals. Nowaday, we uses cervical length to co-evaluate

with any history of prior preterm births. Recent meta-

analysis

shows that cervical cerclage in patients with a

history of prior preterm births and short cervical length

Maneeratprasert et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

398

Original Article

SMJ

(<25 mm.) is eective at reducing preterm births, and

also avoids unnecessary cervical cerclage based solely

on history. In this study, charts from the past 10 years

were reviewed, but some cases provided no information

about cervical length. Future additional research and

RCTs are therefore still require.

CONCLUSION

e success rate of cervical cerclage procedures

in the 10 year period at Siriraj Hospital was 73.3%. e

success rate in the prophylactic group was 80.8%, while

it was 41.2% in the rescue cervical cerclage procedure

group. Possible associated factors included cervical

length, cervical dilatation, bulging of membranes, and

prophylactic procedure.

Author contributions: All authors contributed signicantly

for the study. PM, PS, and DB together plan and design

the study. Conduct of the study and data collection were

mainly performed by PM. DB performed data analysis.

PM draed the manuscript. PS and DB were responsible

for editing the nal manuscript. All authors have reviewed

and agree with the nal version of the manuscript.

Conicts of interest: e authors declare no conicts

of interest.

REFERENCES

1. Beck S, Wojdyla D, Say L, Betran AP, Merialdi M, Requejo JH,

et al. e worldwide incidence of preterm birth: a systematic review of

maternal mortality and morbidity. Bull World Health Organ

2010;88:31-8.

2. Blencowe H, Cousens S, Oestergaard MZ, Chou D, Moller AB,

Narwal R, et al. National, regional, and worldwide estimates of

preterm birth rates in the year 2010 with time trends since 1990 for

selected countries: a systematic analysis and implications.

Lancet 2012;379:2162-72.

3. Chawanpaiboon S, Sutantawibul A. preterm birth rate in Siriraj

Hospital:a seven-year review (2002-2008 BE). ai J Obstet

Gynecol 2009;17:204-14.

4. American College of Obstetricians and Gynecologists. ACOG

Practice Bulletin No.142: Cerclage for the management of

cervical insuciency. Obstet Gynecol 2014;123:372-9.

5. Wang SW, Ma LL, Huang S, Liang L, Zhang JR. Role of cervical

cerclage and vaginal progesterone in the treatment of cervical

incompetence with/without preterm birth history. Chin Med

J (Engl) 2016;129:2670-5.

6. Sciscione AC. Maternal activity restriction and the prevention

of preterm birth. Am J Obstet Gynecol 2010;202:232 e1-5.

7. Wood SL, Owen J. Cerclage: Shirodkar, McDonald, and

Modications. Clin Obstet Gynecol 2016;59:302-10.

8. Hume H, Rebarber A, Saltzman DH, Roman AS, Fox NS.

Ultrasound-indicated cerclage: Shirodkar vs. McDonald. J

Matern Fetal Neonatal Med 2012;25:2690-2.

9. Odibo AO, Berghella V, To MS, Rust OA, Althuisius SM,

Nicolaides KH. Shirodkar versus McDonald cerclage for the

prevention of preterm birth in women with short cervical

length. Am J Perinatol 2007;24:55-60.

10. Final report of the Medical Research Council/Royal College

of Obstetricians and Gynaecologists multicentre randomised

trial of cervical cerclage. MRC/RCOG Working Party on

Cervical Cerclage. Br J Obstet Gynaecol 1993;100:516-23.

11. Naqvi M, Barth WH, Jr. Emergency cerclage: outcomes,

patient selection, and operative considerations. Clin Obstet

Gynecol 2016;59:286-94.

12. Alrevic Z, Owen J, Sharp AN, Szychowski JM, Goya M.

Vaginal progesterone, cerclage or cervical pessary for preventing

preterm birth in asymptomatic singleton pregnant women

with a history of preterm birth and a sonographic short cervix.

Ultrasound Obstet Gynecol 2013;41:146-51.

13. Hezelgrave NL, Watson HA, Ridout A, Diab F, Seed PT,

Chin-Smith E, et al. Rationale and design of Support: a multi-

centre randomised controlled trial to compare three treatments:

cervical cerclage, cervical pessary and vaginal progesterone,

for the prevention of preterm birth in women who develop a

short cervix. BMC Pregnancy Childbirth 2016;16:358.

14. Sutchritpongsa P, Wanitpongpan P. Successful rescue cervical

cerclage in patient with advanced cervical dilatation in second

trimester : case report. Siriraj Med J 2016;68:183-6.

15. Figueroa R, Crowell R, Martinez A, Morgan M, Wakeeld D.

McDonald versus Shirodkar cervical cerclage for the prevention

of preterm birth: impact of body mass index. J Matern Fetal

Neonatal Med2019;32:3408-14.

16. Jung EY, Oh KJ, Hong JS, Han BR, Joo JK. Addition of adjuvant

progesterone to physical-exam-indicated cervical cerclage to

prevent preterm birth. J Obstet Gynaecol Res 2016;42:1666-72.

17. Abu Hashim H, Al-Inany H, Kilani Z. A review of the

contemporary evidence on rescue cervical cerclage. Int J

Gynaecol Obstet 2014;124:198-203.

18. Liu Y, Ke Z, Liao W, Chen H, Wei S, Lai X, et al. Pregnancy

outcomes and superiorities of prophylactic cervical cerclage

and therapeutic cervical cerclage in cervical insufficiency

pregnant women. Arch Gynecol Obstet 2018;297:1503-8.

19. Harger J. Comparison of success and morbidity in cervical

cerclage procedures. Obstet Gynecol 1980;56:543-8.

20. Lee KN, Whang EJ, Chang KH, Song JE, Son GH, Lee KY.

History-indicated cerclage: the association between previous

preterm history and cerclage outcome. Obstet Gynecol Sci

2018;61:23-9.

21. Berghella V, Mackeen AD. Cervical length screening with

ultrasound-indicated cerclage compared with history-indicated

cerclage for prevention of preterm birth: a meta-analysis.

Obstet Gynecol 2011;118:148-55.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

399

anyaporn Rakbangboon, M.Sc.*, Manunchaya Samala, M.Ed.*, Sunee Bovonsunthonchai, Ph.D.**, Gary

Guerry, Ph.D.*, Kasun Kithsiri Rupasinghe, PO.B.*, Loganayaki Ramakrishnan, PO.B. *, Stephney Weerasinghe,

PO.B.*, Jeewantha Sasika, PO.B.*, Pham i Hoa, PO.B.* anatat Charatrungolan, M.Econ.*

*Sirindhorn School of Prosthetics and Orthotics, Faculty of Medicine Siriraj Hospital, Mahidol University, **Faculty of Physical erapy, Mahidol

University, Bangkok 10700, ailand.

The Vertical Ground Reaction Force and Temporal-

Spatial Parameters of Transfemoral Amputees

Wearing Three Prosthetic Knee Joints Available

in Thailand: a Pilot Study

ABSTRACT

Objective: To examine the temporal-spatial characteristics of transfemoral amputees using three prosthetic knees

available in ailand. In addition, the estimated vertical Ground Reaction Force (vGRF) was explored, in particular

the graphical dierences in the M-shape of the vGRF pattern amongst each of the knees and the sound limb.

Methods: ree transfemoral amputees were tted with three dierent prosthetic knee joints (Chulalongkorn

University (CU) Polycentric Knee Joint, Prosthesis Foundation Knee, Otto Bock 3R20) and performed walking

trials while the vGRF and temporal-spatial parameters were collected for all participants.

Results: Similarities existed amongst GRF metrics across all prosthetic knees. Stance and swing time in the CU

Polycentric Knee Joint was similar to that of the sound limb. Walking speeds were highest in the Otto Bock 3R20

and lowest in the Prosthesis Foundation Knee.

Conclusion: is preliminary pilot testing revealed similarities amongst all three prosthetic knees. Future research

with more participants and additional analysis could further elucidate characteristics of these prosthetic knees.

Keywords: Prosthetic knee joint; transfemoral amputee gait; vertical ground reaction, temporal-spatial parameters

(Siriraj Med J 2019; 71: 399-404)

Corresponding author: Manunchaya Samala

E-mail: manunchaya.saa@mahidol.edu

Received 21 September 2017 Revised 17 May 2018 Accepted 27 May 2019

ORCID ID: http://orcid.org/0000-0002-6244-7209

http://dx.doi.org/10.33192/Smj.2019.60

INTRODUCTION

Lower extremity amputations are occurring at an

increasing rate in South East Asia.

e provision of

functional lower limb prosthesis is common place throughout

the world and the advancement of technologies has led

to an improved quality of life for persons aicted with

transfemoral amputations.

2,3

Still, for the transfemoral

amputee, the loss of the anatomical knee joint presents a

plethora of problems.

e biggest being the inability to

provide anatomical exion and extension of the prosthesis.

e able-bodied person has an intact knee joint and

therefore a greater control over placement of the foot

during walking. e above knee amputee has increased

metabolic cost and a disturbed sensory control because

of the absence of normal joint and muscular functions.

ey must make muscle adaptions in their non-aected

limb musculature to compensate for these residual limb

issues.

Furthermore, there are lowered joint torques and

Rakbangboon et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

400

Original Article

SMJ

mechanical energy requirements for the transfemoral

prosthesis limb and an increased demand placed on the

sound limb compared to able-bodied persons.

Fortunately,

eorts have been made to reduce the inherent reduced

swing phase duration that occurs with the prosthetic knee

joint, through either mechanical or hydraulic knees.



e prosthetist creating a transfemoral prosthesis

has a number of suitable prosthetic knee options available

to them. e single axis hinged knee joints that were

commonly prescribed several decades ago are no longer

a standard of care for the ambulating amputee. Modern

knee joints have integrated hydraulic mechanisms for

added functionality.

Prosthetic knees which take into

consideration direction of the instantaneous center of

rotation (ICR) while attempting to mimic the sliding

motion of the anatomical knee joint, are more commonly

provided to users.

A commonly used knee in ailand is

the Otto Bock 3R20 (Otto Bock, Duderstadt, Germany),

which is a 4-bar linkage knee joint with a mechanical

swing phase control. e kinematics of transfemoral

amputees walking in this knee have been explored before.

e 3R20 knee has been shown to allow for decreased

mean values of stride length and increased walking speed

than its lower end model the 3R15.

10,11

However, there has been an increased push by the

United Nations for aordable locally developed rehabilitation

technologies over recent years.

ailand has answered

the call by developing its own prosthetic knee joints. e

CU Polycentric Knee Joint and the Prosthesis Foundation

Knee were both designed and developed with a desire

to provide locally made prosthetic technology to ai

amputees. e CU Polycentric Knee Joint is designed

to closely mimic the motion of the anatomical knee

joint through utilization of a unique sliding apparatus

and the Prosthesis Foundation Knee is a 4-bar linkage

polycentric knee joint. ere is a dearth of objective data

and scholarship for both of these knee joints as they are

still relatively new to the market. Of unique interest to

researchers and prosthetists are the performance features

provided by each of the aforementioned knee joints.

Objectively measuring knee characteristics and

stability can be performed through use of instrumented

motion analysis exploring kinematic and kinetic

parameters, although this procedure requires the use of

resource demanding laboratory equipment and training.

Exploring stability through alternative means such as

plantar pressure distribution analysis is another viable

option.

Pressure mat systems are useful for measuring

pressure patterns occurring under the foot and to help

understand temporal-spatial parameters such as stance

and swing variability and cadence. By combining the

calculated simultaneous estimations of vertical ground

reaction force (vGRF) and center of pressure (COP), the

pressure mat system becomes a useful tool for identifying

features of the aforementioned prosthetic knee joints.

e objective of this study was to examine the temporal-

spatial characteristics of transfemoral amputees using

three prosthetic knees available in ailand. In addition,

the estimated vGRF was explored, and in particular the

graphical dierences in the M-shape of the vGRF pattern

amongst each of the knees and the sound limb.

MATERIALS AND METHODS

is study protocol was approved by Siriraj Institutional

review Board (Si 013/2014).

Participants

All participants provided written informed consent

prior to completing the Siriraj Hospital Faculty of Medicine

Mahidol University SIRB protocol. Inclusion criteria

were restricted to unilateral transfemoral amputees and

patients of the Prosthetic and Orthotic clinic at Siriraj

Hospital. ree eligible participants were recruited in

this study. Each participant was 2-years post amputation,

had a medium stump length, at least a grade 4 out of 5

on the manual muscle test (MMT) as determined by the

study prosthetist. Participants’ were previously familiar

with walking with their transfemoral prostheses for at

least one year. An Amputee Mobility Predictor (AMPro)

activity level of either 3 or 4 (MFCL K Level) was required

for all participants.

Prosthesis

Each participant performed walking trials in a

newly provided prosthesis which received the Otto

Bock 3R20, CU Polycentric Knee Joint or Prosthesis

Foundation Knee (Fig 1). For the entirety of the study,

all participants received a quadrilateral socket design

with passive suction suspension and were then randomly

assigned to receive either of the study knees, a solid-ankle

cushion heel (SACH) foot which was appropriate for

persons requiring stability and comfort during walking,

and a 1 cm heel height shoe. Participants performed an

hour walking accommodation period at their preferred

walking speed along an indoor walkway prior to all data

collection. Furthermore, each of the prosthesis was

properly aligned by the study prosthetist according to

manufacturer guidelines.

Walking trials

All participants performed outcome measures in their

prosthesis under each prosthetic knee conguration for

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

401

a total of three trials. e Zebris Gait Mat System (zebris

Medical GmbH, Germany) and FDM Soware were used

to collect and analyze temporal-spatial parameters as well

as vGRF. Participants were instructed to walk along the

designated walkway at their preferred walking speed.

Data analysis

Data analysis from the Zebris Gait Mat System

was performed using Microso Excel Analysis Toolpak

(Microso, US). Key parameters of interest were walking

velocity, percentages of stance and swing phases of

both limbs, and step length. Moreover, with respects to

vGRF, the visible dierences between the sound side and

prosthesis side rst peak (weight acceptance), second peak

(push-o) and valley of vGRF were analyzed. e GRF

data was reported in Newton’s (N) and was normalized

to body weight with each step time being normalized

to 100% of the gait cycle, where a complete cycle was

dened as two consecutive heel contacts on a single foot.

RESULTS

Participant demographics and activity classication

are provided in Table 1. Slight dierences were observed

for comparisons of the rst peak and valley of the mean

GRF for each of the study knees and sound limb (Fig 2).

Similarities existed amongst GRF metrics across all

prosthetic knees. Furthermore, across all prosthetic knees,

the stance phase was shorter on the prosthetic side than

on the sound side. When prosthetic and sound sides

were compared, the curves were mostly dissimilar along

the second peak and valley. e valley of vGRF of the

sound limb were lowest and highest when participants

wore the Prosthesis Foundation Knee (Fig 3). Stance and

swing phase dierences amongst prosthetic knees and

the sound limb showed that the CU Polycentric Knee

Joint exhibited stance (57%) and swing (43%) times

closer to the sound limb (70%) and (30%) respectively

(Figs 4 and 5). e preferred walking velocities showed

a higher velocity when participants used the 3R20, and

lowest velocity when using the Prosthesis Foundation

Knee (Table 2). With respects to spatial parameters, the

step lengths of participants using the CU Polycentric

Knee Joint (68.33 cm) and Prosthesis Foundation Knee

(68.33 cm) were closer in length to the sound side limb

(51.88 cm), whereas the longest step length was observed

when participants wore the 3R20 (72.33 cm).

Fig 1. Image of three prosthetic knees tested: From le to right; Chulalongkorn University (CU) Polycentric Knee Joint, Prosthesis Foundation

Knee, Otto Bock 3R20

TABLE 1. Demographic data and MCFL classication for three study participants.

Participant Amputation Cause K-Level Age Height (kg) Weight (cm) BMI

Participant 1 Traumatic 3 17 168 52 18.42

Participant 2 Traumatic 3 46 173 68 22.72

Participant 3 Traumatic 3 53 164 55 20.44

*MCFL; Medicare Functional Classication Level, K-Level; MCFL, BMI = kg/m

Rakbangboon et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

402

Original Article

SMJ

Fig 2. Example of one participant’s vGRF curve comparisons of vertical ground reaction force for three prosthetic knees and sound side

limb. (k1:CU Polycentric Knee Joint, k2:Prosthesis Foundation Knee, k3:Otto Bock 3R20)

Fig 3. Comparisons of vertical ground reaction force of three prosthetic knees and sound side limb at TVF (valley of the force in the vertical

direction). Bars represent average vGRF for %BW (k1: CU Polycentric Knee Joint, k2: Prosthesis Foundation Knee, k3: Otto Bock 3R20)

Fig 5. Mean swing phase duration times for three prosthetic knees and sound side during a walking trial (k1: CU Polycentric Knee Joint,

k2: Prosthesis Foundation Knee, k3: Otto Bock 3R20).

Fig 4. Mean stance phase duration times for three prosthetic knees and sound side during a walking trial (k1: CU Polycentric Knee Joint,

k2: Prosthesis Foundation Knee, k3: Otto Bock 3R20).

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

403

DISCUSSION

e purpose of this pilot study was to explore GRF

and temporal-spatial characteristics of three prosthetic

knees available to the ai transfemoral amputee. Overall,

the investigation revealed small dierences in valley vGRF

curve magnitudes between prosthetic knees and the

sound limb. e CU Polycentric Knee Joint was capable

of stance and swing times that were closer to that of the

sound limb and increased preferred walking speed was

seen in participants wearing the 3R20. Able bodied gait

is characterized by roughly 60% of gait spent in stance

and 40% of time spent in swing.

Previous investigations

have observed the mechanical swing phase control 3R20

to be favorable by users because of a greater sense of

stability over a pneumatic swing phase control knee.

e CU Polycentric Knee Joint’s ability to provide

a more identical swing and stance time is an interesting

nding which might be as a result of its lowered Instant

Centre of Rotation (ICR) than the two 4-bar linkage

knees. e CU Polycentric Knee Joint and Prosthesis

Foundation Knee had step lengths that more closely

resembled sound limb step lengths. e increased walking

speed and larger step length seen in the 3R20 might be

a result of a decreased dependence on compensatory

mechanisms of the participants,

although our data

does not allow us to determine that entirely. e lower

limb amputee lowers walking speed to reduce the Rate

of Metabolic Energy Expenditure (RMEE) (Vo2, mL/kg

per min).

16,17

Walking speed and RMEE could be explored

to determine the relationship between speed of walking

and metabolic cost when tted with these knees.

is preliminary investigation was not without

its limitations. e small participant sample of three

transfemoral prosthesis users makes generalizing the results

of this study dicult. Our decision to analyze data from

three trials might have resulted in variance in amputee

walking performance. Although we noticed no marked

dierences amongst trial metrics, an increase in walking

trials over several days could reveal the chance of trial-

to-trial and day-to-day dierences. Future investigations

which recruit a larger sample size and able bodied matched

control group should be pursued. Also these knee joints

should be compared with others commonly used in

prosthetic practice. We used a pressure measurement

system, which is capable of quantifying center of pressure,

but unable to directly measure a force vector applied.

Other investigations have utilized coecients collected

from calibration trials to delineate gait phase and in

doing so correct GRF and COP data collected from

pressure mats. is in turn allows for a better estimation

of forces in each foot during the double limb support

period.

is study focused on vGRF and temporal-spatial

parameters, so a more robust description of walking with

these knees including kinematic, electromyography and

energy consumption should be conducted. ere are a

number of additional GRF measurements that were not

explored in this current study that should be in future

work, such as weight acceptance and push-o as well

as impulses and joint kinematics as well.

Ultimately,

the lower limb prosthesis user, especially those living in

resource limited environments, will be required to walk

and traverse in a variety of terrain. erefore, research

exploring the performances of amputees using these

and other new prosthetic technologies during free-living

activity should be explored.

CONCLUSION

is study presents a preliminary investigation

on the gait characteristics of transfemoral prosthesis

wearers using three prosthetic knees available to the ai

amputee. Special importance was given to the parameters

that indicate walking ability. e CU Polycentric Knee

Joint should be recognized for its ability to promote

a more symmetrical gait and the 3R20, for its potential

to increase walking speeds.

TABLE 2. Temporal-spatial parameters for three prosthetic knees.

Prosthetic knee Velocity (m/s) Step length (cm)

CU Polycentric Knee Joint 0.814 68.33

Prosthesis Foundation Knee 0.792 68.33

Otto Bock 3R20 0.892 72.33

Sound Limb N/A 51.88

Rakbangboon et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

404

Original Article

SMJ

ACKNOWLEDGMENTS

e authors would like to express the gratitude to Ms.

Cathy McConnell for her valuable suggestion and support

for this research work as well as the Sirindhorn School of

Prosthetics & Orthotics (SSPO) for the research funding and

thank Assistant Professor Dr. Pairat Tangpornprasert and

Assistant Professor Dr. Chanyaphan Virulsri, Department

of Mechanical Engineering, Chulalongkorn University

for providing the CU Polycentric knee joints in the study

as well as for the valuable suggestions and support.

REFERENCES

1. Yoon KH,Lee JH,Kim JW,Cho JH,Choi YH,Ko SH,et al.

Epidemic obesity and type 2 diabetes in Asia. Lancet 2006;368:

1681-8.

2. Eide A, Øderud T. Disability & International Development.

(Maclachlan M, Swartz L, eds.). New York, NY: Springer US;

2009.p.149-60.

3. Powell BA, Mercer SW, Harte C. Measuring the impact of

rehabilitation services on the quality of life of disabled people

in Cambodia. Disasters 2002;26:175-91.

4. Winter DA. Biomechanics and Motor Control of Human

Movement. Hoboken, NJ, USA: John Wiley & Sons, Inc.;

2009.p.176-99.

5. Seroussi RE, Gitter A, Czerniecki JM, Weaver K. Mechanical

work adaptations of above-knee amputee ambulation. Arch

Phys Med Rehabil 1996;77:1209-14.

6. Hale SA. Analysis of the swing phase dynamics and muscular

eort of the above-knee amputee for varying prosthetic shank

loads. Prosthet Orthot Int 1990;14:125-35.

7. Hicks R, Tashman S, Cary JM, Altman RF, Gage JR. Swing

phase control with knee friction in juvenile amputees. J Orthop

Res 1985;3:198-201.

8. Radclie CW. Four-bar linkage prosthetic knee mechanisms:

kinematics, alignment and prescription criteria. Prosthet

Orthot Int. 1994;18:159-73.

9. Boonstra AM, Schrama JM, Eisma WH, Hof AL, Fidler V.

Gait analysis of transfemoral amputee patients using prostheses

with two dierent knee joints. Arch Phys Med Rehabil 1996;77:

515-20.

10. Farahmand F, Rezaeian T, Narimani R, Hejazi Dinan P. Kinematic

and Dynamic Analysis of the Gait Cycle of Above-Knee Amputees.

Sci Iran 2006;13:261-71.

11. Taheri A, Karimi MT. Evaluation of the gait performance of

above-knee amputees while walking with 3R20 and 3R15 knee

joints. J Res Med Sci 2012;17:258-63.

12. United Nations Committee on the Rights of Persons with

Disabilities. Convention on the Rights of Persons with Disabilities;

2008.p.18-20

13. Muro-de-la-Herran A, Garca-Zapirain B, Mndez-Zorrilla A.

Gait Analysis Methods: An Overview of Wearable and Non-

Wearable Systems, Highlighting Clinical Applications. Sensors

2014;14:3362-94.

14. Gailey RS,Roach KE,Applegate EB,Cho B,Cunnie B,Licht

S,et al. e Amputee Mobility Predictor: An instrument to

assess determinants of the lower-limb amputee’s ability to

ambulate. Arch Phys Med Rehabil 2002;83:613-27.

15. Pitkin M. Eects of Design Variants in Lower-Limb Prostheses

on Gait Synergy. J Prosthetics Orthot 1997;9:113-22.

16. Waters RL, Perry J, Antonelli D, Hislop H. Energy cost of

walking of amputees: the inuence of level of amputation. J

Bone Joint Surg Am 1976;58:42-46.

17. Czerniecki JM, Morgenroth DC. Metabolic energy expenditure

of ambulation in lower extremity amputees: what have we

learned and what are the next steps? Disabil Rehabil 2017;39:

143-51.

18. Perry J, Burneld J. Gait Analysis: Normal and Pathological

Function. 2

ed. Slack Incorporated; 2010.p.215-318.

19. Johansson JL, Sherrill DM, Riley PO, Bonato P, Herr H. A

clinical comparison of variable-damping and mechanically

passive prosthetic knee devices. Am J Phys Med Rehabil2005;84:

563-75.

20. Lythgo N,Marmaras B,Connor H. Physical function, gait, and

dynamic balance of transfemoral amputees using two mechanical

passive prosthetic knee devices. Arch Phys Med Rehabil2010;91:

1565-70.

21. Boonstra AM,Schrama J,Fidler V,Eisma WH. Energy cost

during ambulation in transfemoral amputees: a knee joint with

a mechanical swing phase control vs a knee joint with a pneumatic

swing phase control. Scand J Rehabil Med1995;27:77-81.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

405

Leena Chularojanamontri, M.D.*, Natchaya Junsuwan, M.D.*, Narumol Silpa-archa, M.D.*, Chanisada Wongpraparut,

M.D.*, Praveena Chiowchanwisawakit, M.D.**

*Department of Dermatology, **Department of Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.

Utility of the Siriraj Psoriatic Arthritis Screening

Tool, the Thai Psoriasis Epidemiology Screening

Tool, and the Early Arthritis for Psoriatic Patients

Questionnaire to Screen for Psoriatic Arthritis in

an Outpatient Dermatology Clinic Setting, and

IdenticationofFactorsSignicantlyAssociated

with Psoriatic Arthritis

ABSTRACT

Objective: To assess the clinical utility of the Psoriasis Epidemiology Screening Tool (PEST), the Early Arthritis

for Psoriatic Patients (EARP) questionnaire, and the Siriraj Psoriatic Arthritis Screening Tool (SiPAT) as screening

tools for psoriatic arthritis (PsA), and to identify factors signicantly associated with PsA.

Methods: is cross-sectional study included adult psoriasis patients who attended the outpatient clinic at Siriraj

Hospital and had not been diagnosed with PsA during 1 March 2017 to 28 February 2018. Participants completed

the EARP, PEST, and SiPAT, aer which musculoskeletal history was taken, and examination and radiography

were performed. Diagnosis of PsA was based on Classication Criteria for Psoriatic Arthritis. Receiver operator

characteristic (ROC) curves, sensitivity, and specicity were used to determine assessment tool performance. Logistic

regression analysis was used to identify factors associated with PsA.

Results: Eighty-seven patients with a mean age of 45.90±14.75 years were enrolled. Twenty-six (29.88%) patients

were diagnosed as PsA. According to ROC values, EARP had the best discriminative power (0.83) for distinguishing

between psoriatic patients with and without PsA (SiPAT: 0.78, PEST: 0.77). SiPAT had the highest sensitivity (92.3%),

followed by EARP (84.6%) and PEST (50.0%); whereas, PEST had the highest specicity (82.0%), followed by EARP

(62.3%) and SiPAT (54.1%) for detecting PsA. Multivariate analysis revealed body surface area involvement >10%

to be the only independent predictor of PsA (OR: 2.99, 95% CI: 1.09-8.21).

Conclusion: SiPAT is an eective and simple to use tool for screening PsA in psoriasis patients. Body surface area

involvement >10% is a signicant predictor of PsA.

Keywords: Factors associated with PsA; Psoriasis Epidemiology Screening Tool (PEST); screening tools for psoriatic

arthritis; the Early Arthritis for Psoriatic Patients (EARP) questionnaire; the Siriraj Psoriatic Arthritis Screening

Tool (SiPAT) (Siriraj Med J 2019; 71: 405-413)

Corresponding author: Praveena Chiowchanwisawakit

E-mail: preveena.chi@mahidol.ac.th

Received 28 January 2019 Revised 17 April 2019 Accepted 14 May 2019

ORCID ID: http://orcid.org/0000-0002-4253-9229

http://dx.doi.org/10.33192/Smj.2019.61

Chularojanamontri et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

406

Original Article

SMJ

INTRODUCTION

Psoriatic arthritis (PsA) is a multidimensional

musculoskeletal inammatory disease that is associated

with psoriasis (Ps). e prevalence of PsA in Ps was

reported to range from 6 to 42%.

Early detection and

intervention is important for good long-term outcome

in terms of achieving remission, stopping and reversing

structural change, and regaining and preserving function.

Most patients present with Ps before developing PsA.

us, dermatologists and general practitioners play an

important role in the identication of PsA. However, up to

40% of patients with Ps had undiagnosed PsA.

A simple

and eective PsA screening tool is, therefore, needed.

Currently available screening tools include the Psoriatic

Arthritis Screening and Evaluation (PASE) questionnaire

the Psoriasis Epidemiology Screening Tool (PEST)

and the Early Arthritis for Psoriatic Patients (EARP)

questionnaire

for patients with Ps, and the Toronto

Psoriatic Arthritis Screen II (ToPAS II) questionnaire

for both Ps patients and the general population. e

reported variation in the performance of these tools may

be due to patient musculoskeletal manifestations.

8-9

PsA-

related musculoskeletal manifestations are divided into

the following 3 groups: spinal inammation, peripheral

arthritis, and enthesitis. e Siriraj Psoriatic Arthritis

Screening Tool (SiPAT), which focuses on the evaluation

of musculoskeletal manifestations, is a composite tool that

was developed via the adoption of 2 questions from the

EARP questionnaire, and 1 question from PEST.

e

formal permission for these assessments to be translated

into ai language and validated.

A previous study

showed the sensitivity and specicity to be 91% and 69%

for SiPAT; 83% and 79% for EARP; and, 72% and 90%

for PEST, respectively (all ai versions).

However,

there are two notable reasons why the aforementioned

sensitivities may have been overestimated. First, that

study included both diagnosed and undiagnosed PsA

patients. Second, it was performed in a psoriasis clinic

where the majority of patients would have severe skin

disease and/or musculoskeletal symptoms. Accordingly,

the aim of this study was to investigate the clinical utility

of the SiPAT, the PEST, and the EARP questionnaire

to screen for PsA in an outpatient dermatology clinic

setting, and to identify factors signicantly associated

with PsA.

MATERIALS AND METHODS

Adult psoriasis patients aged 18 years or older

who attended the outpatient dermatology clinic, Siriraj

Hospital during 1 March 2017 to 28 February 2018, and

that had never been diagnosed with PsA were asked

to voluntarily participate in this cross-sectional study.

Siriraj Hospital is ailand’s largest university-based

national tertiary referral center. e protocol for this

study was approved by the Siriraj Institutional Review

Board, Faculty of Medicine Siriraj Hospital, Mahidol

University, Bangkok, ailand (Si 659/2012), and written

informed consent was obtained from all participants.

is study complied with the principles set forth in the

1964 Declaration of Helsinki and all of its subsequent

amendments.

All enrolled patients were asked to individually

complete (i) the validated ai version of PEST, (ii) the

validated ai version of EARP, and (iii) the SiPAT.

ese three questionnaires were provided not only as

three separate forms, but they were also given in random

order to avoid completion bias. Table 1 shows all of the

questions contained in the three questionnaires. e EARP

and PEST contain 10 and 5 questions, respectively, with

a score of 1 for each positive answer and a score of 0 for

each negative answer. A score of 3 or higher for each

of these two questionnaires was considered to indicate

that the patient was positive for PsA. e SiPAT consists

of 3 questions designed to elicit information about the

presence of inammatory back pain, peripheral arthritis,

and heel enthesitis. A positive answer was scored as 1,

and a negative answer was scored as 0. A SiPAT score of

1 or higher was considered to indicate that the patient

was positive for PsA.

An evaluation of patient medical records and

physical examinations were then performed. Current

non-steroidal anti-inammatory drugs (NSAIDs) use

was dened as current use or drug discontinuation for

less than 2 weeks before recruitment. Musculoskeletal

examination was performed independently by one

expert rheumatologist (PC). Rheumatoid factor (RF),

and radiography of the cervical spine, lumbar spine,

pelvis, hands, and feet were performed. All radiography

results were interpreted by one expert rheumatologist

(PC). e expert rheumatologist (PC) who examined

patients and evaluated patient data was blinded to patient

questionnaire data from the EARP, PEST, and SiPAT.

PsA was diagnosed by a rheumatologist according to the

Classication Criteria for Psoriatic Arthritis (CASPAR).

Patterns of PsA, including peripheral arthritis, axial

inammation, and enthesitis, were dened using patient

history, physical examination of joint/musculoskeletal

pain, and radiography.

Sample size calculation and statistical analysis

Using an estimated specicity of about 60% and

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

407

a 95% confidence interval (CI) of 60% ± 12% from

previous study, a sample size of 65 subjects without

PsA was required for this study.

Based on an estimated

prevalence of PsA of about 25%, a total of 88 subjects

with Ps was calculated.

Descriptive statistics were used to summarize patient

demographic and clinical data. Pearson’s Chi-square test

or Fisher’s exact test was used to compare categorical

data. Independent t-test and Mann-Whitney U test were

used to compare continuous parametric and continuous

non-parametric data, respectively. Receiver operating

characteristic (ROC) curves were generated and analyzed

to evaluate the diagnostic performance of the EARP,

PEST, and SiPAT questionnaires. e area under the

ROC curve (AUC) measured how well each questionnaire

could distinguish between psoriatic patients with and

without PsA. An AUC of 0.5 indicated no distinguishable

dierence between the two conditions, while an AUC of

1 reected a clear and absolute distinguishable dierence

between PsA and no PsA.

Logistic regression was

employed to identify factors signicantly associated

with PsA. Variables with a p-value less than or equal to

0.2 in univariate analysis were entered into multivariate

analysis to identify independent predictors of PsA.

A p-value less than or equal to 0.05 in multivariate

analysis was regarded as being statistically signicant.

Data were analyzed using PASW Statistics version 18

(SPSS, Inc., Chicago, IL, USA).

TABLE 1. e questions asked on the Early Arthritis for Psoriatic Patients (EARP) questionnaire, the Psoriasis

Epidemiology Screening Tool (PEST) questionnaire, and the Siriraj Psoriatic Arthritis Screening Tool (SiPAT).

A ‘yes’ answer receives a score of 1, and a ‘no’ answer receives a score of 0.

Assessment tools Yes No

EARP*

1. Do your joints hurt? 1 0

2.Haveyoutakenanti-inammatorydrugsmorethantwiceaweekforjointpainwithinthelast3months? 1 0

3. Do you wake up at night because of low back pain? 1 0

4. Do you feel stiffness in your hands for more than 30 minutes in the morning? 1 0

5.Doyourwristsandngershurt? 1 0

6.Doyourwristsandngersswell? 1 0

7.Doesonengerhurtandswellformorethan3days? 1 0

8. Does your Achilles tendon swell? 1 0

9. Do your feet or ankles hurt? 1 0

10. Do your elbows or hips hurt? 1 0

PEST**

1. Have you ever had a swollen joint (or joints)? 1 0

2. Has a doctor ever told you that you have arthritis? 1 0

3.Doyourngernailsortoenailshaveholesorpits? 1 0

4. Have you had pain in your heel? 1 0

5.Haveyouhadangerortoethatwascompletelyswollenandpainfulfornoapparentreason 1 0

SiPAT***

1. Do you wake up at night because of low back pain? 1 0

2.Doyourwristsandngersswell? 1 0

3. Have you had pain in your heel? 1 0

*,**Formal permission to validate and translate these two assessment tools into ai language was obtained from the copyright owners. e

validated ai versions of these two questionnaires were used in this study. ***Items contained in the SiPAT questionnaire were adopted

from items 3 and 6 of the EARP, and from item 4 of the PEST. Its validated version was used in this study.

Chularojanamontri et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

408

Original Article

SMJ

RESULTS

Eighty-eight participants were initially enrolled;

however, one participant was excluded before any

investigations were performed due to age less than

18 years. Eighty-seven patients with a mean age of

45.90±14.75 years were enrolled in this study. e most

common underlying diseases were hypertension (30%),

dyslipidemia (27.6%), and diabetes mellitus (13.8%).

Primary, secondary, vocational, and tertiary education

was completed in 16, 21, 9, and 41 patients, respectively.

All patients were able to answer all of the questions on

all three questionnaires (18 questions in total) within

5 minutes. According to CASPAR criteria, 26 (29.9%)

patients had newly established PsA. Of patients with

no PsA, 34 (55.7%) patients had other musculoskeletal

diseases, including mechanical back pain (14 patients),

knee osteoarthritis (10 patients), hand osteoarthritis (5

patients), heel pain related to activity (4 patients), and

gout (1 patient). Six (23.1%) patients with PsA also had

other musculoskeletal diseases, including mechanical

back pain (4 patients), knee osteoarthritis (1 patient), and

coexisting mechanical back pain and knee osteoarthritis

(1 patient). e data in Table 2 shows longer disease

duration, presence of psoriatic nail, and body surface

area (BSA) involvement >10% to be signicantly more

common in patients with PsA. Moreover, the EARP, PEST,

and SiPAT scores obtained from psoriatic patients with

PsA were signicantly higher than the scores obtained

from psoriatic patients without PsA.

e diagnostic performance of the three questionnaires

for detecting PsA and patterns of PsA is shown in Table 3.

According to area under the ROC curve (AUC) values, the

EARP questionnaire demonstrated the best discriminative

power (0.83) for distinguishing between psoriatic patients

with and without PsA (SiPAT: PEST: 0.77). All three

questionnaires performed well in all patterns of PsA, except

in purely axial inammation. e SiPAT had the highest

sensitivity (92.3%), followed by EARP (84.6%) and PEST

(50.0%). Conversely, the PEST had the highest specicity

(82.0%), followed by EARP (62.3%) and SiPAT (54.1%),

regardless of musculoskeletal patterns. For subgroup of

PsA pattern analysis, the sensitivity of EARP and SiPAT

were still good in the three major musculoskeletal patterns

regardless of combinations of other patterns, except pure

axial inammation. Conversely, the sensitivity of PEST

were lower than EARP and SiPAT in the three major

musculoskeletal patterns regardless of combinations of

other patterns, especially axial pattern. In addition, the

PEST could not detect any PsA patients with pure axial

inammation.

Univariate analysis showed disease duration more

than 10 years [odds ratio (OR): 3.03, 95% CI: 1.14-8.03;

p=0.026], nail involvement (OR: 5.02, 95% CI: 1.07-23.52;

p=0.040), and BSA involvement higher than 10% (OR: 3.58,

95% CI: 1.36-9.41; p=0.010) to be signicantly associated

with PsA (Table 4). Multivariate analysis revealed BSA

involvement >10% to be the only independent predictor

of PsA (OR: 2.99, 95% CI: 1.09-8.21; p=0.034) (Table 5).

DISCUSSION

Dermatologists need an eective and simple to

use screening tool that can dierentiate between Ps

patients with and without PsA. Several of the available

PsA screening tools, including PASE (15 questions), EARP

(10 questions), and ToPAS II (13 questions with pictures),

contain at least 10 questions, which may be complicated,

and time-consuming for use in a dermatology outpatient

clinic setting.

4, 6-7

e Psoriatic arthritis UnclutteRed

screening Evaluation-4 (PURE-4) (4 questions) was

recently developed for dermatologists to screen for PsA.

e overall performance of PURE-4 for diagnosing newly

established PsA was good with sensitivity (85.7%) and

specicity (83.6%), and an AUC of 0.88.

e Simple

Psoriatic Arthritis Screening (SiPAS) questionnaire (5

questions), which is another tool that was developed to

screen for PsA, was reported to have a sensitivity of 79%

and a specicity of 87%.

SiPAT was designed as a short PsA screening tool

(3 questions), which is similar to the 4-question PURE-

4 and the 5-question SiPAS. e three questions from

the PEST and EARP assessments that were considered

most relevant to detecting the 3 major patterns of PsA

were adopted for use in the SiPAT questionnaire.

Regardless of musculoskeletal patterns, the specicity

of SiPAT was much lower than that of the PURE-4 and

the SiPAS for detecting PsA. e study population in

a study that investigated the performance of the SiPAS

did not include patients with other rheumatic diseases

that can mimic PsA.

In contrast, our study included all

Ps patients, which included a high proportion of other

musculoskeletal diseases, especially mechanical back

pain. Seven patients without PsA answered ‘yes’ for the

rst question of the SiPAT (Do you wake up at night

because of low back pain?). Of those 7 patients, six had

mechanical back pain, and the one remaining patient

had no evidence of musculoskeletal disease. Low back

pain is a common symptom in general population. To

improve the performance of the SiPAT questionnaire

for detecting spinal inammation, the question used to

detect inammatory back pain should be amended in

the future.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

409

TABLE 2. Demographic and clinical characteristics compared between patients with and without psoriatic arthritis

(PsA) (N=87).

Characteristics No PsA (n=61) PsA (n=26) P-value

Age (yrs), mean±SD 46.0±14.7 45.6±15.3 0.896

Male gender, n (%) 30 (49.2%) 16 (61.5%) 0.290

Age of onset (yrs), n (%) 0.784

<40 38 (62.3%) 17 (65.4%)

 ≥40 23(37.7%) 9(34.6%)

Disease duration (yrs), median (IQR, min-max) 7.6 12.2 0.018

(10.8, 0.6-43.6) (14.4, 0.26-40.5)

Psoriatic nail, n (%) 43 (70.5%) 24 (92.3%) 0.027

Bodymassindex≥25kg/m

, n (%) 31 (53.4%) 17 (68.0%) 0.218

Body surface area (BSA) involvement, n (%)

 ≤10% 46(75.4%) 12(46.2%) 0.008

>10% 15 (24.6%) 14 (53.8%)

PASI, n (%)

 ≤10% 49(80.3%) 18(69.2%) 0.260

>10% 12 (19.7%) 8 (30.8%)

Past and current treatment, n (%)

Only topical treatment 23 (37.7%) 9 (34.6%) 0.784

Systemic treatment + phototherapy 38 (62.3%) 17 (65.4%)

Methotrexate use for psoriasis, n (%)

Current 18 (29.5%) 7 (26.9%) 0.807

NSAIDs use, n (%)

Current 2 (3.3%) 4 (15.4%) 0.063

Past 19 (31.1%) 6 (23.1%) 0.446

Positive rheumatoid factor (RF), n (%) 17/58 (29.3%) 9/25 (36.0%) 0.547

Pattern of PsA*, n (%)

Peripheral arthritis - 17 (65.4%)

 Axialinammation - 15(57.7%)

Enthesitis - 8 (30.8%)

Assessment tools, mean±SD

EARP score 1.9±2.0 4.5±1.9 <0.001

PEST score 1.4±1.2 2.7±1.3 <0.001

SiPAT score 0.6±0.7 1.5±0.8 <0.001

A p-value<0.05 indicates statistical signicance

Abbreviations: SD = standard deviation; IQR = interquartile range; PASI = Psoriasis Area & Severity Index; NSAIDS = non-steroidal

anti-inammatory drugs; EARP = Early Arthritis for Psoriatic Patients questionnaire; PEST = Psoriasis Epidemiology Screening Tool;

SiPAT = Siriraj Psoriatic Arthritis Screening Tool

*One patient could have one more pattern of PsA

Chularojanamontri et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

410

Original Article

SMJ

TABLE 3. Diagnostic performance of the Early Arthritis for Psoriatic Patients (EARP) questionnaire, the Psoriasis

Epidemiology Screening Tool (PEST), and the Siriraj Psoriatic Arthritis Screening Tool (SiPAT).

Tools AUC Sensitivity Specicity PPV NPV LR+ LR-

(95% CI) (%) (%) (%) (%)

All psoriatic arthritis patients (n=26)

EARP 0.83 (0.73-0.92) 84.6% 62.3% 48.9% 90.5% 2.24 0.24

PEST 0.77 (0.67-0.87) 50.0% 82.0% 54.2% 79.4% 2.78 0.61

SiPAT 0.79 (0.69-0.89) 92.3% 54.1% 46.2% 94.3% 2 0.15

Patients with predominant peripheral arthritis (n=17)

EARP 0.92 (0.86-0.98) 100% 62.3% 42.5% 100% 2.65 0

PEST 0.85 (0.75-0.95) 70.6% 82.0% 52.2% 90.9% 3.92 0.36

SiPAT 0.83 (0.74-0.93) 100% 54.1% 37.8% 100% 2.17 0

Patientswithpredominantaxialinammation(n=15)

EARP 0.75 (0.62-0.89) 73.3% 62.3% 32.4% 90.4% 1.94 0.43

PEST 0.70 (0.57-0.83) 33.3% 82.0% 31.3% 83.3% 1.85 0.81

SiPAT 0.75 (0.62-0.89) 86.7% 54.1% 31.7% 94.3% 1.89 0.25

Patients with predominant enthesitis (n=8)

EARP 0.87 (0.74-1.00) 87.5% 62.3% 23.3% 97.4% 2.32 0.20

PEST 0.83 (0.69-0.97) 62.5% 82.0% 31.3% 94.3% 3.47 0.46

SiPAT 0.86 (0.75-0.97) 100% 54.1% 22.2% 100% 2.18 0

Patients with pure peripheral arthritis (n=9)

EARP 0.92 (0.86-0.99) 100% 61.3% 27.3% 100% 2.58 0

PEST 0.84 (0.71-0.96) 66.7% 82.3% 35.3% 94.4% 3.77 0.40

SiPAT 0.81 (0.69-0.93) 100% 51.6% 23.1% 100% 2.07 0

Patientswithpureaxialinammation(n=5)

EARP 0.54 (0.34-0.74) 40.0% 62.3% 8.0% 92.7% 1.06 0.96

PEST 0.56 (0.39-0.73) 0.0% 82.0% 0.0% 90.1% 0 1.22

SiPAT 0.58 (0.31-0.84) 60.0% 54.1% 9.7% 94.3% 1.31 0.74

Patients without PsA (n=61) were included in all subsections of this analysis. ere was no patient with pure enthesitis, so it was not further

analyzed.

Abbreviations: AUC = area under the curve; CI = condence interval; PPV = positive predictive value; NPV = negative predictive value;

LR+ = positive likelihood ratio; LR- = negative likelihood ratio.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

411

TABLE 4. Univariate analysis for factors associated with psoriatic arthritis (PsA).

Patients (%)

Factors No PsA PsA Odds ratio P-value

(95% CI)

Sex

Male 30 (49.2) 16 (61.5) 1.65 (0.65 - 4.22) 0.292

Female 31 (50.8) 10 (29.9) 1

Obesity, BMI (kg/m

)

 ≥25 31(53.4) 17(68.0) 1.85(0.69-4.96) 0.221

< 25 27 (46.6) 8 (32.0) 1

Disease duration (years)

 ≥10 26(42.6) 18(69.2) 3.03(1.14-8.03) 0.026

< 10 35 (57.4) 8 (30.8) 1

Nail involvement

Yes 43 (70.5) 24 (92.3) 5.02 (1.07 - 23.52) 0.040

No 18 (29.5) 2 (7.7) 1

Body surface area (BSA)

> 10% 15 (24.6) 14 (53.8) 3.58 (1.36 - 9.41) 0.010

 ≤10% 46(75.4) 12(46.2) 1

Psoriasis Area and Severity Index (PASI)

> 10% 12 (19.7) 8 (30.8) 1.82 (0.638 - 5.16) 0.264

 ≤10% 49(80.3) 18(69.2) 1

Current treatment

Topical treatment 29 (47.5) 12 (46.2) 1.06 (0.42 - 2.65) 0.906

Systemic treatment + phototherapy 32 (52.5) 14 (53.8) 1

Abbreviations: CI = condence interval, BMI = body mass index

TABLE 5. Multivariate analysis for factors associated with psoriatic arthritis.

Variable Adjusted odds ratio (95%CI) P-value

Disease duration > 10 years 2.63 (0.94-7.37) 0.065

Nail involvement 4.31 (0.87-21.42) 0.075

BSA involvement > 10% 2.99 (1.09-8.21) 0.034

Abbreviations: CI = condence interval, BSA = Body surface area

Chularojanamontri et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

412

Original Article

SMJ

Conversely, the sensitivities of the SiPAT were very

high, with 90-100% sensitivity for predominant enthesitis

and peripheral arthritis in both previous study

and the

current study. However, no sensitivities and specicities

for specic musculoskeletal manifestations were reported

for the PURE-4 or the SiPAS. Moreover, they have not

been evaluated in other populations, which could show

that their diagnostic performance may be decreased

from a development step as the previous report.

be a screening tool for detecting early PsA in psoriasis

population, the sensitivity of the test might be more

important than the specicity.

ese screening tools

are designed to identify Ps patients with probable PsA

that need to be evaluated by a rheumatologist. Detection

of PsA at an early stage and treatment may result in

preventing deformity and good function.

In addition,

the performance of the screening tools may depend on

ability to detect all patterns of PsA.

e sensitivity and

specicity of EARP and SiPAT for detecting PsA were

good and comparable, regardless of PsA patterns in this

study. Conversely, the PEST had much lower performance

in detecting axial inammation than EARP and SiPAT.

It may result from EARP and SiPAT capturing all three

main patterns of PsA while PEST has no question assessing

spinal inammation.

e pathogenesis that contributes to the development

of PsA needs to be further elucidated. Four large cohort

studies revealed increasing BMI, smoker, and excessive

alcohol consumption to be signicantly associated with

the development of PsA.

e diagnosis of PsA in those

four cohort studies was based on medical code and

patient self-report veried by PASE tool. Moreover,

previous case-control studies which PsA were diagnosed

by a rheumatologist using CASPAR criteria reported

scalp psoriasis, intergluteal and/or perianal psoriasis,

>3 aected sites, nail dystrophy, injuries, infections that

required antibiotics, and heavy liing to be signicantly

associated with the development of PsA; however, the

reported associations were inconsistent among studies.

e observed variation in associations may be due to

variations in patient characteristics. In the present

study, we found only severity of Ps (BSA involvement

more than 10%) to be an independent predictor of PsA

in multivariate analysis which the same as Haroon,

et al ’study.

e mentionable limitations of this study

include those associated with studies with a cross-sectional

design, including lack of temporal association (risk

factors and disease are measured at the same point

in time), and a small study population from a single

center compared to the larger populations enrolled in

multicenter cohort studies.

In conclusion, the results of this study revealed the

SiPAT screening tool to have a relatively high sensitivity

for detecting PsA in patients with Ps in an outpatient

dermatology clinic setting. Although the pathogenesis of

PsA and the risk factors for the development of PsA both

need to be further studied and elucidated, the ndings

of our study suggest that treatment of skin psoriasis may

prevent the development of PsA.

ACKNOWLEDGMENTS

is research project was supported by a grant from

the Siriraj Research Fund, Faculty of Medicine Siriraj

Hospital, Mahidol University (grant no. R015935053).

e funder did not inuence the design of this study,

the decision to report the ndings of this study, or the

journal submission decision. e authors gratefully

acknowledge the study subjects for their participation and

cooperation, Dr. Chulaluk Komoltri for her assistance

with statistical analysis, and Mr. Kevin P. Jones for

editing the manuscript.

Conict of interest declaration: All authors declare no

personal or professional conicts of interest relating to

any aspect of this study.

REFERENCES

1. Gladman DD, Antoni C, Mease P, Clegg DO, Nash P. Psoriatic

arthritis: epidemiology, clinical features, course, and outcome.

Ann Rheum Dis 2005;64 Suppl 2:ii14-7.

2. Coates LC. erapy strategies in psoriatic arthritis. Clin Exp

Rheumatol 2015;33:S70-2.

3. Mease PJ, Gladman DD, Papp KA, Khraishi MM, aci D,

Behrens F, et al. Prevalence of rheumatologist-diagnosed

psoriatic arthritis in patients with psoriasis in European/North

American dermatology clinics. J Am Acad Dermatol 2013;69:

729-35.

4. Dominguez P, Husni ME, Garg A, Qureshi AA. Psoriatic

Arthritis Screening and Evaluation (PASE) questionnaire and

the role of dermatologists: a report from the GRAPPA 2009

annual meeting. J Rheumatol 2011;38:548-50.

5. Helliwell PS. Psoriasis Epidemiology Screening Tool (PEST):

a report from the GRAPPA 2009 annual meeting. J Rheumatol

2011;38:551-2.

6. Tinazzi I, Adami S, Zanolin EM, Caimmi C, Confente S,

Girolomoni G, et al. e early psoriatic arthritis screening

questionnaire: a simple and fast method for the identication of

arthritis in patients with psoriasis. Rheumatology (Oxford.

2012;51:2058-63.

7. Tom BD, Chandran V, Farewell VT, Rosen CF, Gladman

DD. Validation of the Toronto Psoriatic Arthritis Screen

Version 2 (ToPAS 2). J Rheumatol 2015;42:841-6.

8. Haroon M, Kirby B, FitzGerald O. High prevalence of psoriatic

arthritis in patients with severe psoriasis with suboptimal

performance of screening questionnaires. Ann Rheum Dis

2013;72:736-40.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

413

9. Coates LC, Walsh J, Haroon M, Fitzgerald O, Aslam T, Al

Balushi F, et al. Development and testing of new candidate

psoriatic arthritis screening questionnaires combining optimal

questions from existing tools. Arthritis Care Res Hoboken

2014;66:1410-6.

10. Chiowchanwisawakit P, Wattanamongkolsil L, Srinonprasert

V, Petcharat C, Siriwanarangsun P, Katchamart W. Developing

the ai Siriraj Psoriatic Arthritis Screening Tool and validating

the Thai Psoriasis Epidemiology Screening Tool and the

Early Arthritis for Psoriatic Patients questionnaire. Rheumatol

Int 2016;36:1459-68.

11. Taylor W, Gladman D, Helliwell P, Marchesoni A, Mease

P, Mielants H, et al. Classication criteria for psoriatic arthritis:

development of new criteria from a large international study.

Arthritis Rheum 2006;54:2665-73.

12. Mandrekar JN. Receiver operating characteristic curve in

diagnostic test assessment. J orac Oncol 2010;5:1315-6.

13. Audureau E, Roux F, Lons Danic D, Bagot M, Cantagrel A,

Dernis E, et al. Psoriatic arthritis screening by the dermatologist:

development and rst validation of the ‘PURE-4 scale’. J Eur

Acad Dermatol Venereol2018;32:1950-3.

14. Sala F, Di Carlo M, Luchetti MM, Di Donato E, Campanati

A, Benfaremo D, et al. A validation study of the Simple Psoriatic

Arthritis Screening (SiPAS) questionnaire to screen psoriasis

patients for psoriatic arthritis. Clin Exp Rheumatol 2018;36:

127-35.

15. Parikh R, Mathai A, Parikh S, Chandra Sekhar G, omas

R. Understanding and using sensitivity, specicity and predictive

values. Indian J Ophthalmol 2008;56:45-50.

16. Haroon M, Gallagher P, FitzGerald O. Diagnostic delay of

more than 6 months contributes to poor radiographic and

functional outcome in psoriatic arthritis. Ann Rheum Dis

2015;74:1045-50.

17. Ogdie A, Gelfand JM. Clinical Risk Factors for the Development

of Psoriatic Arthritis Among Patients with Psoriasis: A Review

of Available Evidence. Curr Rheumatol Rep 2015;17:64.

Chularojanamontri et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

414

Review Article

SMJ

Ployploen Phikulsod, M.D.*, Weerapat Owattanapanich, M.D.*, Marie Bill, M.D.**, Wannaphorn Rotchanapanya,

M.D.***, Marcus C. Hansen**, Marianne Hokland, M.D.****, Peter Hokland, M.D.**

*Department of Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand, **Department of Haematology, Aarhus

University Hospital, Aarhus 8000, Denmark, ***Department of Medicine, Chiangrai Prachanukroh Hospital, Chiang Rai, ailand,****Department

of Biomedicine, Faculty of Health, Aarhus University, Aarhus 8000, Denmark.

The Development of Personalized Medicine: Acute

Myeloid Leukemia as a Model

ABSTRACT

e term personalized medicine has been employed in widely dierent contexts and has acquired status as

one of the most oen used keywords recently. In this review we take it to understand the application of modern

diagnostic medicine and therapeutics to patient with the purpose of eradicating disease or alleviating symptoms in

a manner, where all actions are based on detailed knowledge of the condition of the individual patient. Applying

these concepts should lead to optimization of clinical decision-making and, in its utmost consequence, a substantial

decrease in costs incurred for hospitalization and follow-up. e latter is based on the evidence that for many

disorders “less but more targeted” will mean improved outcome. Acute myeloid leukemia (AML) is the most

common acute leukemia in adults and is a major challenge in terms of diagnosis, care, follow-up and therapy.

us, in population-based analyses, overall survival is only just exceeding 40% with major reasons for treatment

failure. For these reasons, AML has been intensely studied during the recent decades. With the development of

multiparametric ow cytometry, it allows us to get an accurate diagnosis and immunophenotypic proles of AML.

In addition, there is now an abundance of knowledge regarding its cytogenetic and molecular background. ese

enable us to follow the amount of disease down to the minutest quantity with a high resolution of molecular details.

Finally, based on knowledge of these variables in the single patient cytoreduction is now being rened to therapies

targeted to the molecular changes in the patient.

Keywords: Acute myeloid leukemia; developing country; precision medicine; targeted therapy; ailand (Siriraj

Med J 2019; 71: 414-425)

Corresponding author: Ployploen Phikulsod

E-mail: ployploen.phi@mahidol.ac.th

Received 20 June 2019 Revised 15 July 2019 Accepted 8 August 2019

ORCID ID: http://orcid.org/0000-0001-5389-1896

http://dx.doi.org/10.33192/Smj.2019.62

INTRODUCTION

Acute myeloid leukemia (AML) is a highly malignant

blood cancer characterized by rapid accumulation of

abnormally dierentiated immature myeloid cells in

the bone marrow (BM) and peripheral blood (PB),

resulting in disruption of the production of normally

functioning blood cells.

us, patients present with

severe symptoms of BM failure, i.e. anemia, bleeding

and oen life-threatening infections. AML can arise

de novo, secondary to antecedent myelodysplastic syndrome

(MDS) or myeloproliferative neoplasm (MPN), known as

secondary acute myeloid leukemia (sAML), or following

cytotoxic therapy (tAML). e incidence of AML is

generally reported to be 3-5 per 100,000 per year, but

varies according to age and study. Despite the fact that

initial therapy induces complete remission (CR) in up to

80% of patients, the relapse-rates are high in AML other

than acute promyelocytic leukemia (APL) subtype. e

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

415

ve-year survival rate of younger patients (age < 60 years)

is approximately 40% and less than 20% in the elderly.

Unfortunately, data from less developed countries, such

as ailand, showed an even worse outcome. e 2-year

overall survival (OS) rates of younger and elderly groups

in ai AML patients were merely 28.6% and 10.3%,

respectively.

Surprisingly, the ai cohort showed a

lower proportion of patients with good-risk cytogenetics

in comparison to other reports.

e diagnostic tool box in AML

There are several diagnostic tools that should

be included in the investigation of newly-diagnosed

AML patients (Fig 1). Firstly, blast counts from PB

and BM smear are simple assays and instantly provide

the provisional diagnosis of AML. Standard guidelines

recommend counting from 200 total white blood cells

from PB or 500 total nucleated cells from BM smear.

4,5

20% myeloblasts or more in PB or BM is compatible

with AML; however, physicians should keep in mind

that AML can be diagnosed in the patient who has

t(15;17), t(8;21) or inv(16) – even if blast count is less

than 20%.

4,5

In addition, for AML-M4, a total blast counts

must include monoblasts and promonocytes.

Secondly, immunophenotyping from multiparametric

ow cytometry (MFC) is an essential investigation for AML

Fig 1. e diagnostic toolbox in acute myeloid leukemia. Blood counts (A) and morphology of peripheral blood and bone marrow (B)

are still frontline assays, when diagnosing acute myeloid leukemia, whereas as molecular genetics/genomics (C), cytogenetics (D), ow

cytometry (E) are of highly important for subtype classication and prognosis. is information, together with performance score (F), forms

the basis of clinical decision-making and treatment strategy for the individual patient. Cytogenetics are also used in diagnosis as specic

chromosomal aberrations may override the lower threshold of blasts in the peripheral blood or bone marrow.

subtype categorization. However, it is required that the

MFC machine should be well-equipped with ≥2 lasers and

≥6-8 photomultiplier tubes and operated with technical

expertise. Immunophenotypic markers can separate

precursor-, granulocytic-, monocytic-, megakaryocytic-

and erythroid lineages. MFC is an important part of the

diagnostic work up in daily clinical practice, as it provides

a rapid blast count together with immunophenotyping

capable of conrming the myeloid origin of blasts. In

some cases, the immunophenotype can be indicative of

underlying cytogenetic aberrations, e.g. co-expression of

the lymphoid marker CD19 on myeloid blasts in t(8;21)

AML or lack of CD34 and HLA-DR antigens in APL.

Moreover, a baseline immunophenotypic result before

treatment is important for treatment response evaluation

and measurable residual disease (MRD) monitoring.

7-9

Cytogenetic and molecular analyses are also required

for all newly-diagnosed AML to predict patient prognosis

and guide treatment decision.

Besides, these genetic

data are also used to further classied patients into

dierent subtype according to WHO 2016 classication

criteria.

e molecular panel that is mandatory to be screened

by polymerase chain reaction (PCR) based method

includes PML-RARA, CBFβ-MYH11, RUNX1-RUNX1T1

and BCR-ABL1 genes. In addition, other gene mutations

Phikulsod et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

416

Review Article

SMJ

TABLE 1. Diagnostic tool for newly-diagnosed AML patient.

Strongly recommended Turnaround time

Complete blood count with blood smear 1 hour

Bone marrow aspiration for assessment of blast number 1 day

Multiparametricowcytometry 3days

Screening for gene rearrangements including PML-RARA, CBFβ-MYH11, 3 days

RUNX1-RUNX1T1 and BCR-ABL1 genes by RT-PCR

Cytogenetics 7 days

Mutations of NPM1 and FLT3-ITD 3 days

Recommended or optional Turnaround time

SpecicgenemutationsincludingCEBPA, KIT, RUNX1, TP53, ASXL1, 7-14 days

IDH1 and IDH2 genes

Next-generation sequencing for myeloid panel 7-14 days

including NPM1, FLT3-ITD, CEBPA, KIT, RUNX1,

TP53, ASXL1, IDH1 and IDH2 genes are also highly

recommended from the latest standard guidelines as

an initial investigation.

4,5

e turn-around time of these

workups should not exceed more than 3 days for fusion

gene rearrangements, NPM1 mutation and FLT3-ITD

mutation and not more than one week for cytogenetic

analyses.

More advanced technologies, such as next-

generation sequencing (NGS), is also now integrated for

molecular characterization of AML patients.

11,12

With this

method, we can obtain multiple or complete mutational

data in a single assay. In the future, we plan to expand the

molecular work-up for ai patients to see if there are

any population-based dierences from other ethnicity

cohorts. is information may arguably be useful in the

personalized therapy era. In addition, cost-eectiveness

of generalized NGS for routine implementation is to be

validated. Table 1 illustrates diagnostic tools for newly-

diagnosed AML patients.

While previous prognostic classication systems have

primarily been based on cytogenetic ndings

, the revised

risk stratication from the European Leukemia Net (ELN)

incorporates the most signicant prognostic mutations

(Table 2).

Genetic risks along with patient related factors,

e.g. age, comorbidity or performance status, and post-

treatment factors, such as the presence or absence of

MRD

, guide the choice and intensity of therapy, which

will be discussed further.

Conventional treatment and follow up in AML

While targeted therapy is evolving rapidly for

hematological malignancies, the general treatment of

AML has not changed substantially the past 30 years.

For

patients with AML eligible for intensive chemotherapy,

standard regimens consist of 1-2 courses of chemotherapy

to induce complete remission (CR). CR is dened by a

presence of less than 5% blasts in BM accompanied by

a complete hematologic recovery. e most commonly

used regimen for AML other than APL subtype, referred

to as the 7+3 regimen, is comprised of 7 days continuous

intravenous cytarabine infusions and 3-day anthracycline

infusions.

11,15

e response is evaluated two weeks aer

the onset of chemotherapy. Younger and physically t

patients, belonging to intermediate or adverse risk groups,

will be referred for allogeneichematopoietic stem cell

transplantation(allo-HSCT) in rst CR. Patients with

favorable risk or without donor eligibility may continue to

receive further 2-4 cycles of consolidation chemotherapy.

Older patients, not eligible for intensive treatment, will

receive best supportive care, e.g. low-dose cytarabine

(LDAC) or alternatively therapy with a hypomethylating

agent (HMA) such as 5-azacitidine

. With the continuously

poor outcome for AML-patients,

it is clear that new

and better treatment options are warranted in order

to improve survival. In that regard, targeted treatment

strategies are attractive in order to obtain ecient kill of

the leukemic cells without extensive toxicity on normal

cells.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

417

TABLE 2. Risk stratication according to the 2017 ELN guidelines.

Risk category Genetic abnormality

Favorable t(8;21)(q22;q22.1); RUNX1-RUNX1T1

inv(16)(p13.1q22) or t(16;16)(p13.1;q22); CBFB-MYH11

Mutated NPM1 without FLT3-ITD or with FLT3-ITD

low

Bi-allelic mutated CEPBA

Intermediate Mutated NPM1 and FLT3-ITD

high

Wild-type NPM1 without FLT3-ITD or with FLT3-ITD

low

(without adverse-risk genetic lesions)

t(9;11)(p21.3;q23.3); MLLT3-KMT2A

 Cytogeneticabnormalitiesnotclassiedasfavorableoradverse

Adverse t(6;9)(p23;q34.1); DEK-NUP214

t(v;11q23.3); KMT2A rearranged

t(9;22)(q34.1;q11.2); BCR-ABL1

Inv(3)(q21.3q26.2) or t(3;3)(q21.3;q26.2); GATA2, MECOM(EVI1)

-5 or del(5q); -7; -17/abn(17p)

Complex karyotype, monosomal karyotype

Wild-type NPM1 and FLT3-ITD

high

Mutated RUNX1

Mutated ASXL1

Mutated TP53

Modied from

, where further details are available

Fig 2. Chemotherapy versus targeted treatment. Targeted therapy is based on the most optimal individual treatment for specic types or

combinations of molecular aberrations. e current molecular diagnostics toolbox to guide precision medicine includes both next generation

sequencing, digital droplet PCR and ow cytometry, each with dierent sensitivities, strengths and weaknesses.

Phikulsod et al.

Taking aim at personalized therapy in AML: clinical

considerations cost-benet

In contrast to the stagnant situation in conventional

induction cytoreduction in AML, treatment therapies are

now becoming more personalized. In recent years, much

focus has been put on mapping genetic and epigenetic

heterogeneity of the malignancies. Consequently, the

information of specic antigens expressed on leukemic

cells together with somatic mutations have been included

in drug investigations to improve outcomes in AML.

e treatment approaches are summarized in Fig 2 and 3.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

418

Review Article

SMJ

e studies on novel therapeutic targets, including

surface antigen, driver oncogenes and cellular pathways

have been an active eld of AML research in recent

years. us, there are many pathway specic agents

and chemotherapy with new drug delivery system being

approved by the US Food and Drug Administration

(FDA). ese include midostaurin, CPX-351, enasidenib

and gemtuzumab ozogamicin in 2017 and glasdegib,

venetoclax, ivosidenib, and gilteritinib in 2018. Here,

we review the data of these promising agents along with

the benet to the patients.

Approved drugs

FMS-like tyrosine kinase 3 (FLT3) Inhibitors

FLT3 mutations are found in 30% of AML patients,

leading to ligand-independent activation of the receptor,

thereby promoting proliferation, survival, and resistance to

apoptosis. It is known that patients with a FLT3 mutations

have signicantly lower OS.

Several drugs that can inhibit

FLT3 kinases have been tested in AML. Midostaurin

works through inhibition of multiple receptor tyrosine

kinases including the activity of wild-type FLT3 and

kinases with either internal tandem duplications (ITD)

or a mutated tyrosine kinase domain (TKD). In phase

III study, where the drug has been given in addition to

standard chemotherapy and as a maintenance therapy,

OS and event-free survival (EFS) were significantly

higher in comparison to placebo (74.7 vs. 25.6 months

and 8.2 vs. 3 months, respectively).

Due to its survival

benet, US FDA has approved its use in combination

with standard induction and consolidation therapy

in adults with newly diagnosed AML, who are FLT3-

mutation positive. Another approved agent is gilteritinib.

Gilteritinib is a highly potent inhibitor which has been

approved in fast track according to preliminary phase I/II

data.

Nonetheless, the drug as a monotherapy ultimately

showed improved survival in relapsed or refractory

AML patients with FLT3-mutation in comparison to

standard chemotherapy (9.3 VS 5.6 months) in complete

Phase III analysis. However, long term survival was

still uncommon on either treatment arm. e longest

survival in the gilteritinib arm was seen among patients

Fig 3. Examples of targeted therapy in acute myeloid leukemia. Targeted therapy or precision medicine has the purpose of eradicating

leukemic cells with specic molecular aberrations without hitting nonspecic targets, such as normal hematopoietic cells. ese drugs

include monoclonal antibodies, such as antibody-drug conjugate gemtuzumab ozogamicin (Mylotarg), which targets CD33, or selective small

molecule inhibitors, such as venetoclax for older patients that respond poorly to standard therapy. Other promising strategies of precision

medicine include the use of chimeric antigen receptor T cells, where patient T cell are engineered to target the malignant cells (insert).

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

419

who underwent HSCT followed by maintenance therapy.

erefore, ongoing study has been done to see whether

using the drug as a rst-line treatment would give more

survival benet to the patients. Other drugs in this category

including sorafenib, quizartinib and crenolanib, have

shown benet in preclinical and early clinical studies

and awaits for further validation in various settings.

Isocitrate dehydrogenase (IDH) Inhibitors

IDH1 and IDH2 are enzymes that catalyze the

conversion of isocitrate to α-ketoglutarate. IDH1 and

IDH2 mutations occur in 5-10% and 10-15% of adult AML

patients, respectively, with higher frequencies in older

patients. ese result in an abnormal enzyme activity,

lead to the competitive inhibition of α-ketoglutarate

dependent enzymes, target genes hypermethylation

and impaired hematopoietic dierentiation.

So far,

two agents targeting these mutant enzymes have been

approved. e rst drug, enasidenib, which is a selective

IDH2 inhibitor, provided overall response rate (ORR) of

40.3% and impressive median time to rst response of

1.9 months in relapsed and refractory AML patients.

Data also conrmed that enasidenib can salvage a number

of patients (50% ORR) and provided survival benets.

ere is one particular study of enasidenib in newly

diagnosed elderly patients, which showed that there

were relatively low toxicities (cytopenic rate 21%), good

response rate (30.8%) and acceptable OS (11.3 months).

Ivosidenib, on the other hand, inhibits IDH1 mutant

enzyme, demonstrated ORR of 42% and CR of 22% in

relapsed and refractory AML patients.

erefore, both

were approved for treatment of relapsed or refractory

AML with a corresponding IDH mutation. e durations

of responses varied from an average of 6 months to

more than 2 years. Additional studies will give more

information on the benet of these drugs in terms of a

long-term survival as a frontline treatment. Studies has

now been expanded to test these drugs in combination

with chemotherapy in newly diagnosed patients and

gave a promising results.

B-cell leukemia/lymphoma-2 (BCL2) inhibitor

Venetoclax is a selective, oral small-molecule BCL-2

inhibitor, leading to cell apoptosis

(Fig 3). BCL2 is

overexpressed in hematologic malignancies and implicated

in AML cell survival, chemoresistance, and is linked

to poor OS in AML patients.

As a monotherapy, it

demonstrated activity in relapsed refractory or unt

AML (19% overall response rate) with a tolerable safety

prole.

IDH mutational status was correlated with

good responses, however, all patients eventually relapse

despite the initial response. Lately, the studies showed

that rapid resistance may occur from MCL1 and BAX

upregulation. Many agents can reduce MCL1 expression

in vitro including chemotherapy and HMA.

Venetoclax

in combination with HMA in newly diagnosed elderly

AML patients demonstrated 67% CR and median OS of

17.5 months.

e responses occurred in patients with

poor risk mutations; TP53 and FLT3 mutations, similar to

those with IDH and NPM1 mutations. ese impressive

results led to recent approval of drug combination with

HMA or LDAC for frontline treatment of elderly unt

patients. New combination therapy such as MCL1 inhibitor

and IDH inhibitors are lining up to be tested.

Glasdegib

is drug works through inhibition of sonic hedgehog

pathway in leukemic stem cells (Fig 3). In phase II,

randomized, open-label, multicenter study, 132 patients

with AML or high-risk MDS unsuitable for intensive

chemotherapy were evaluated for the ecacy of glasdegib

plus LDAC in comparison to LDAC alone.

It showed

OS benet (8.8 vs. 4.9 months) and clinical ecacy

across patients with diverse genetic proles.

is led

to its approval by US FDA in 2018 as a treatment in unt

AML/MDS patients.

CPX-351

CPX-351 is a dual-drug liposomal encapsulation of

cytarabine and daunorubicin packaged at a synergistic

dose of 5:1 molar ratio.

A nanoliposomal carrier of

the drug leads to prolonged exposure and intracellular

delivery. In comparison to a conventional 7+3 induction

of cytarabine and daunorubicin treatment in high risk

elderly AML patients (60–75 years of age), CPX-351

arm demonstrated higher CR and CR with incomplete

recovery (CRi) (57% vs 40% and 73% vs 52%) and led

to higher number of patients undergoing HSCT. CPX-

351 also improved OS compared with 7+3 (median OS

9.56 vs 5.95 months), regardless of age, therapy-related

or MDS-related subgroup.

Consequently, it has been

approved as a frontline treatment in patients of all ages

with therapy-related AML or AML with MDS-related

changes.

Gemtuzumab ozogamicin

CD33 antigen is a transmembrane receptor and

myeloid dierentiation marker variably expressed on AML

cells in almost all patients.

Gemtuzumab ozogamicin

is a humanized CD33 antibody-toxin conjugate toxic

to CD33-expressing leukemic cells

(Fig 3B). Aer

binding to the antigen on the surface of leukemic blasts

Phikulsod et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

420

Review Article

SMJ

the antibody is internalized and binds to DNA leading to

double-strand break and cell death. It has been previously

withdrawn from American market in 2010 due to its

toxicity and limited benet.

However, with recently more

ecacy data and desperate needs to improve treatment

of elderly AML patients, the drug has been resubmitted

to US FDA regulatory review again.

In newly diagnosis

adult AML patients age 50 to 70 years, a randomized

phase III trial of daunorubicin and cytarabine with or

without gemtuzumab ozugamycin in induction and

consolidation chemotherapies (ALFA-0701) has shown

the benet of increase median EFS of 15.6 months in

the gemtuzumab ozogamicin arm versus 9.7 months in

control arm except in patients with adverse cytogenetic

risk.

In addition, in de novo or sAML patients age >75

years or age 61-75 years with WHO performance status

>2 who are unwilling to receive intensive chemotherapy,

gemtozumab ozogamycin monotherapy resulted in

a signicant improvement in median OS 1.3 months

comparing to best supportive care arm. e toxicities

did not dier between treatment arms.

is agent was

approved in newly diagnosed CD33-positive adult AML

and relapsed or refractory CD33-positive AML in adult

and pediatric AML aged 2 years and older.

New CD33

antibody and drugs targeting CD33 are currently tested

in various studies.

Investigational therapy

Cellular therapy targeting surface targets

Chimeric antigen receptor T cell (CAR-T) therapy

targeting surface marker of AML, e.g. CD33, CD123 and

FLT3, have been developed in preclinical and early clinical

phase with some concerns of serious toxicities.

Many

leukemia associated antigens are dierentially expressed

by MHC on tumor cells and have been investigated in

small clinical studies to be used as a vaccine or dendritic

cell-based therapy. ese targets include Wilm’s tumor 1

(WT1), New York esophageal squamous cell carcinoma 1

(NY-ESO1), PRAME and survivin (BIRC5).

However,

the main challenge of vaccine-based therapy is MHC

restriction. Immune checkpoint inhibitors are also of

great interests and have currently been studied in dierent

combination at various phases of treatment.

Ongoing investigation therapies

Many more inhibitors that targeting dierent pathways

such as DNA repair and epigenetic modication, including

polyadenosine diphosphate ribose polymerase (PARP),

lysine-specic demethylase 1 (LSD1), bromodomain and

extra terminal (BET) and mutant TP53 pathway have

now been tested in clinical studies.

41,42

Innovative approach for precision medicine

As a result of novel targeted therapy discovery,

we may select the proper treatment to the patients

according to their molecular abnormalities. Not only

specic mutation can be targeted directly but rather

several features can also predict drug response. For

instance, co-occurring IDH2 and DNMT3A mutations

result in distinct DNA methylation pattern, leading to

upregulation of RAS signaling and sensitivity to MEK

inhibition.

is proved to be feasible in Beat AML

study which included 273 elderly patients. Ninety-ve

percent of the study group could be assigned to treatment

by mutation stratied algorithm within 7 days.

Apart

from matching genetic data to targeted therapies, drug

sensitivity testing (DST) approaches such as ex vivo

drug sensitivity screening can guide the individualized

treatment more precisely. By using a sample from each

patient tumor as an avatar to evaluate patient-specic

drug sensitivity proles. is method is attractive due

to easy accessibility of malignant cells in the PB or BM

and will be very useful – especially for relapse/refractory

patient. Preliminary data from a pilot study showed

that it can be done within short turnaround time. Even

though initial results have showed a promising outcome,

further studies should be performed.

However, due

to molecular heterogeneity of AML, possible target

mutations may be one of several potential targets. Many

challenging questions still need to be answered including

how to identify the key pathway, whether a combination

of targeted drug would be benecial and how to manage

the toxicities.

Costs of novel treatment in AML and clinical benets

Advanced therapies have changed the way of treatment

especially to patients with high risk features who have low

chance of survival. Most of them are currently approved

to be used as a last resort in patients who fail or deem

unsuitable to conventional treatment. Even though they

provide more opportunities to increase OS the treatment,

however, rarely lead to cure as monotherapy. ese

eective drugs also come with high costs leading to an

increase of the economic burden in AML treatment.

e prices of these drugs are summarized in Table 3.

In countries with limited resources treatment decision

should always be made by evaluating the cost-eectiveness

to patient survival and quality of life. Choosing a drug,

which provides the best response to the right group of

patients at the right time and administered in a brief

duration as a bridging therapy to curative HSCT, would

be the most optimal way to limit costs and provide the

best outcome. In contrast, we can save cost by not giving

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

421

TABLE 3. Novel drug costs in AML: dose recommendations and average prices.

Drug name Dose recommendations Average price*

Midostaurin 50 mg twice daily, days 8–21 5,448 baht per 25-mg tablet

(FLT3 inhibitor) of induction and consolidation

Gilteritinib 120 mg once daily 9,600 baht per 40 mg

(FLT3 inhibitor)

Enasidenib 100 mg once daily 32,953 baht per 100 mg

(IDH2 inhibitor)

Ivosidenib 500 mg once daily 16,713 baht per 250 mg

(IDH1 inhibitor)

Venetoclax Day 1, 100 mg once daily 357 baht per 10 mg

(BCL2 inhibitor) Day 2, 200 mg once daily 1,784 baht per 50 mg

Day 3, 400 mg once daily 3,568 baht per 100 mg

Day 4 and beyond, 400 mg once daily

Glasdegib 100 mg once daily 10,832 baht per 25 mg

CPX-351 Induction, 44 mg/m

–100 mg/m

on days 1, 3, 5 306,528 baht per 44–100-mg vial

Reinduction (if patient not in remission),

44 mg/m

–100 mg/m

days 1, 3

Consolidation, 29 mg/m

–65 mg/m

on days 1 and 3 for 2 cycles

Gemtuzumab ozogamicin Induction (incombination with “7+3” regimen, 314,880 baht per 4.5-mg vial

(humanized CD33 antibody- 3 mg/m

(max 4.5mg/dose) on days 1, 4, 7

toxin conjugated) Consolidation, 3 mg/m

* day 1

*Costs are reported as average wholesale price and calculated from US dollar price (1 US dollar equals 32 ai Baht). ey are not meant

to represent true costs.

Modied from

Phikulsod et al.

drugs to patient with predictive low response prole.

Nowadays, technologies for precision medicine, and

especially next generation sequencing, have become

much more accessible than a decade ago. With more

information and tools, they will guide the clinicians to

the treatment that provides maximum benets – not

only as a salvage but as a frontline therapy. If we can use

these regimens upfront, we may prevent more patients

from relapsing and avoid the side eects and costs from

chemotherapy and HSCT. Nevertheless, novel therapies

give us treatment options, which were previous unavailable,

and possibly with new side-eects. erefore, information

with regards to risks and benets should be given to

the patient and should be able to reach a nal decision

that match their interests and expectations under solid

counselling from the clinician.

Follow-up in the age of personalized medicine

Relapse in AML patients may be caused by a persistence

or an evolution of therapy-resistant leukemic clones

in the bone marrow. To capture low level of cancer

cells, referred to as MRD, high sensitivity laboratory

investigations are required. As discussed above, relapse

is still common even in patients who have reached CR

states. e goal of MRD monitoring is early detection of

patients at risk and guide treatment decisions to minimize

the risk of clinical relapse by intensifying the therapy.

Alternatively, in the era of targeted therapy, we may be

able to target the specic pathway in resistant leukemic

clones. Conventional methods using morphology and

cytogenetic assay have low sensitivities and only allow

the detection of leukemic cells greater than one cell in

20 leukocytes. erefore, even in a CR state, up to 1011

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

422

Review Article

SMJ

residual leukemic cells may still exist.

Currently, two

widely accepted methods to detect MRD are; MFC and

real time quantitative PCR (RT-qPCR) which can detect

MRD leukemic cells in white blood cells component

with a sensitivity down to 10

-3

-6

However, MRD

detection is not a straightforward process, because of

the immunophenotypic and molecular heterogeneity

between patients and clonal heterogeneity that may

emerge as novel resistant clones under selective pressure

– especially in response to single targeted therapy.

49,50

Identifying the appropriate markers that dierentiate

leukemic cells from normal hematopoietic progenitor, to

evaluate disease burden or MRD and to predict relapse, is

still challenging. Follow-up approaches are summarized

in Fig 4.

Multiparametric ow cytometry

By labelling cells with multiple uorescent markers

and determining the light emission from the cells as they

ow against laser beam, MFC can characterize phenotypes

and immunophenotypes of malignant cells down to a

single level. Apart from being used as a diagnostic tool,

MFC can also be used as a disease monitoring method.

MFC has the sensitivity to detect MRD of leukemic

cells down to 10

-3

-10

-5

It can be applied for almost all

patients. However, due to the complexity of leukemic

immunophenotype described previously, the markers

need to be tailored. A large panel of antibodies of more

than 8 colors are recommended for characterizing the

leukemic cells by MFC. e marker should include early

hematopoietic markers (e.g. CD34, CD45, CD117),

myeloid-lineage associated markers (e.g. CD4, CD11b,

CD11c, CD33, CD64), aberrant dierentiation markers

(e.g. CD2, CD7, CD19, CD56).

MRD can be dened by

either tracking leukemia associated immunophenotypes

(LAIP) from the diagnosis or by identifying new emerging

clone that express aberrant dierentiation proles called

dierent-from-normal (DfN) approach. ELN MRD working

party recommends a combination of both approaches

to best dene MRD burden.

Other markers that could

be used to dene leukemic stem cell population, e.g.

CD38, CD123, CD133, are also of interests. ere are

also some technical considerations including the source

of specimen, sample collection and number of collected

cells. For example, MRD level in PB may be one-tenth the

frequency in BM. erefore, a detection of MRD by MFC

in PB is not currently recommended and bone marrow

specimen should be collected from the rst pull to avoid

hemodilution. Because of the complicated process and

interpretation, a validation of the method is essential. It

Fig 4. Detection of minimal/measurable residual disease (MRD). Currently, PCR is still the most sensitive method to detect MRD. Next

generation sequencing (NGS) is being investigated as a potentially highly sensitive method for detection of residual disease, while increasing

the informational content for clonal assessment.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

423

is recommended that there should be a central institute

for an evaluation and a nal interpretation and should

not be done in a single-centered fashion. Various cut-o

values have been applied but MRD level below 0.1% is

the most widely used in the clinical studies, with more

than 500,000 CD45+ cells acquired.

Molecular MRD

Molecular abnormalities of the leukemic cells can

be used as a marker to monitor the level of malignant

cells with the help of PCR-based methods. qRT-PCR is

a laboratory technique based on the PCR to measure

the amplication of a targeted DNA molecule during

the PCR in real-time. It can be used quantitatively or

semi-quantitatively. is approach has a high sensitivity

for detecting one malignant cell in 10

to 10

leukocytes.

It can detect target genes representing malignant

cells and is considered as the gold standard for MRD

detection. However, in contrast to MCF, this approach

is applicable for only around 40% of patients who have

suitable abnormalities. Target genes could be fusion,

recurrent mutated or overexpressed genes. However,

only some veried markers associated strongly with

relapse including NPM1 mutation, RUNX1-RUNX1T1,

CBFB-MYH11 and PML-RARA fusion.

Some other

mutations may persist aer treatments but do not relate

to real disease burden. For example, mutations that

occur in the preleukemic clone which can be found

in aging population e.g. DNMT3A, ASXL1, TET2

52,53

or in germline e.g. RUNX1, GATA2, CEBPA, DDX41,

ANKRD26

. Apart from those genetic markers, FLT3-

ITD, FLT3-TKD, NRAS, KRAS, IDH1, IDH2, MLL-PTD

and the expression of EVI1 are not recommended to use

as a single marker but rather in combination.

Despite

being shown to relate with patient outcome in many

studies, using WT1 expression as a MRD marker is still

controversial due to its low sensivity.

51,55-57

More recent

technologies such as Next generation sequencing (NGS)

and droplet digital PCR (ddPCR) are ideal to be used

in all patients. NGS gives the whole genetic prole of

the patient while ddPCR allows simultaneous multiple

gene expression quantication. e main shortcoming

of NGS is that it has low specicity in low allelic burden

setting.

While ddPCR require the establishment of

specic set of primers according to sequencing data.

Even though promising, more studies is needed to verify

and standardize the methods.

CONCLUSIONS AND PERSPECTIVES

AML is oen regarded as a signature disease in

hematology. us, it is the most common leukemia

in adulthood and though advances have been made,

both in terms of diagnosis, follow-up and therapy, the

long-term survival of patients is still unsatisfactory,

with treatment failure being the reason. e concepts of

personalized and precision medicine have been coined aer

the availability of genome sequencing and the increased

realization that each patient has dierent in genetic

mutational background. A targeted treatment points

to the specic mutation which causes better treatment

outcome compared to conventional therapy.

e standard AML treatment in last several decades

has changed, expecting to improve patient outcomes

and quality of life. However, having more available

choices might lead to additional concerns, such as drug

costs versus clinical signicance. Comprehensive or

complex decision-making in personalized medicine are

highly relevant matters in these situations. However, it

is important to note that not all these techniques and

the clinical decisions following their results need to

increase the costs of treating AML patients. In fact, the

presence of some molecular phenotypes, such as the

CBF subset, will mean that allo-HSCT is not needed

for certain patient groups. In addition, it should be

remembered that current therapy of AML, like the

time-honored 7+3 regimen, while not being expensive

in term of drugs, is accompanied by very severe side

eects, mostly related to the deep immunosuppression

and highly correlated with serious bacterial and fungal

infection. In this setting, more expensive personalized

therapies developed though knowledge about the phenotype

and molecular prole of the given patient’s malignant

cells, might be more ecient and not accompanied by

similar side-eect, resulting in fewer days in hospital

and minimize antibiotic or antifungal usage. is applies

not only to agents targeting molecular drivers, but also

to the recently developed therapeutic options relating

to redirecting cytotoxic NK- and T-cells (CAR cells) or

unmasking immunogenic molecules on the AML cells

in the exciting eld of checkpoint immunotherapy. In

all these situations it is the job of health care providers to

enlighten administrators and funding authorities about

such potentially cost-saving actions.

It is noteworthy that these issues pertain not only

to auent countries, but equally to those with strained

economies. Pharmacoeconomics is to be investigated

with regards to appropriation of novel therapy for AML

patients in each country.

REFERENCES

1. Swerdlow SH, World Health Organization, International

Agency for Research on Cancer. WHO classication of tumours

Phikulsod et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

424

Review Article

SMJ

of haematopoietic and lymphoid tissues. Revised 4

edition.

ed. Lyon: International Agency for Research on Cancer;

2017.p.585.

2. Dohner H, Weisdorf DJ, Bloomfield CD. Acute Myeloid

Leukemia. N Engl J Med 2015;373:1136-52.

3. Owattanapanich W, Utchariyaprasit E, Tantiworawit A,

Rattarittamrong E, Niparuck P, Puavilai T, et al. Improved

Survival of Elderly-t Patients With Acute Myeloid Leukemia

Requiring Intensive erapy: 3-Year Multicenter Analysis

From TALWG. Clin Lymphoma Myeloma Leuk 2018;18:e509-e14.

4. Dohner H, Estey E, Grimwade D, Amadori S, Appelbaum FR,

Buchner T, et al. Diagnosis and management of AML in adults:

2017 ELN recommendations from an international expert

panel. Blood 2017;129:424-47.

5. NCCN NCCN. Acute Myeloid Leukemia (version 2.2019),

Accessed 20 March 2019. 2019.

6. Bene MC, Nebe T, Bettelheim P, Buldini B, Bumbea H, Kern W,

et al. Immunophenotyping of acute leukemia and lymphoproliferative

disorders: a consensus proposal of the European LeukemiaNet

Work Package 10. Leukemia 2011;25:567-74.

7. Hokland P, Ommen HB, Mule MP, Hourigan CS. Advancing

the Minimal Residual Disease Concept in Acute Myeloid

Leukemia. Semin Hematol 2015;52:184-92.

8. Hokland P, Ommen HB. Towards individualized follow-up in

adult acute myeloid leukemia in remission. Blood 2011;117:

2577-84.

9. Roug AS, Hansen MC, Nederby L, Hokland P. Diagnosing

and following adult patients with acute myeloid leukaemia in

the genomic age. Br J Haematol 2014;167:162-76.

10. Arber DA, Orazi A, Hasserjian R, iele J, Borowitz MJ, Le

Beau MM, et al. e 2016 revision to the World Health Organization

classication of myeloid neoplasms and acute leukemia. Blood

2016;127:2391-405.

11. Papaemmanuil E, Gerstung M, Bullinger L, Gaidzik VI, Paschka

P, Roberts ND, et al. Genomic Classication and Prognosis

in Acute Myeloid Leukemia. N Engl J Med 2016;374:2209-21.

12. Bacher U, Shumilov E, Flach J, Porret N, Joncourt R, Wiedemann

G, et al. Challenges in the introduction of next-generation

sequencing (NGS) for diagnostics of myeloid malignancies

into clinical routine use. Blood Cancer J 2018;8:113.

13. Grimwade D, Hills RK, Moorman AV, Walker H, Chatters S,

Goldstone AH, et al. Renement of cytogenetic classication

in acute myeloid leukemia: determination of prognostic

signicance of rare recurring chromosomal abnormalities

among 5876 younger adult patients treated in the United

Kingdom Medical Research Council trials. Blood 2010;116:

354-65.

14. Yates JW, Wallace HJ, Jr., Ellison RR, Holland JF. Cytosine

arabinoside (NSC-63878) and daunorubicin (NSC-83142)

therapy in acute nonlymphocytic leukemia. Cancer Chemother

Rep 1973;57:485-8.

15. Cassileth PA, Katz ME. Chemotherapy for adult acute

nonlymphocytic leukemia with daunorubicin and cytosine

arabinoside. Cancer Treat Rep 1977;61:1441-5.

16. Li S, Garrett-Bakelman FE, Chung SS, Sanders MA, Hricik

T, Rapaport F, et al. Distinct evolution and dynamics of epigenetic

and genetic heterogeneity in acute myeloid leukemia. Nat Med

2016;22:792-9.

17. iede C, Steudel C, Mohr B, Schaich M, Schakel U, Platzbecker

U, et al. Analysis of FLT3-activating mutations in 979 patients

with acute myelogenous leukemia: association with FAB

subtypes and identication of subgroups with poor prognosis.

Blood 2002;99:4326-35.

18. Stone RM, Mandrekar SJ, Sanford BL, Laumann K, Geyer S,

Bloomeld CD, et al. Midostaurin plus Chemotherapy for

Acute Myeloid Leukemia with a FLT3 Mutation. N Engl J Med

2017;377:454-64.

19. Perl AE, Altman JK, Cortes J, Smith C, Litzow M, Baer MR, et al.

Selective inhibition of FLT3 by gilteritinib in relapsed or

refractory acute myeloid leukaemia: a multicentre, rst-in-

human, open-label, phase 1-2 study. Lancet Oncol 2017;18:1061-

75.

20. Short NJ, Kantarjian H, Ravandi F, Daver N. Emerging treatment

paradigms with FLT3 inhibitors in acute myeloid leukemia.

er Adv Hematol 2019;10:2040620719827310.

21. Saygin C, Carraway HE. Emerging therapies for acute myeloid

leukemia. J Hematol Oncol 2017;10:93.

22. Stein EM, DiNardo CD, Pollyea DA, Fathi AT, Roboz GJ,

Altman JK, et al. Enasidenib in mutant IDH2 relapsed or

refractory acute myeloid leukemia. Blood 2017;130:722-31.

23. Largeaud L, Berard E, Bertoli S, Dufrechou S, Prade N, Gadaud

N, et al. Outcome of AML patients with IDH2 mutations in

real world before the era of IDH2 inhibitors. Leuk Res 2019;81:82-7.

24. Popovici-Muller J, Lemieux RM, Artin E, Saunders JO, Salituro

FG, Travins J, et al. Discovery of AG-120 (Ivosidenib): A

First-in-Class Mutant IDH1 Inhibitor for the Treatment of

IDH1 Mutant Cancers. ACS Med Chem Lett 2018;9:300-5.

25. Stein EM, DiNardo CD, Mims AS, Savona MR, Pratz K, Stein

AS, et al. Ivosidenib or Enasidenib Combined with Standard

Induction Chemotherapy Is Well Tolerated and Active in

Patients with Newly Diagnosed AML with an IDH1 or IDH2

Mutation: Initial Results from a Phase 1 Trial. Blood 2017;130:726.

26. Konopleva M, Zhao S, Hu W, Jiang S, Snell V, Weidner D, et al.

e anti-apoptotic genes Bcl-X(L) and Bcl-2 are over-expressed

and contribute to chemoresistance of non-proliferating leukaemic

CD34+ cells. Br J Haematol 2002;118:521-34.

27. Konopleva M, Pollyea DA, Potluri J, Chyla B, Hogdal L,

Busman T, et al. Ecacy and Biological Correlates of Response

in a Phase II Study of Venetoclax Monotherapy in Patients

with Acute Myelogenous Leukemia. Cancer Discov 2016;6:

1106-17.

28. Campos EDV, Pinto R. Targeted therapy with a selective BCL-2

inhibitor in older patients with acute myeloid leukemia. Hematol

Transfus Cell er 2019;41:169-77.

29. DiNardo CD, Pratz K, Pullarkat V, Jonas BA, Arellano M,

Becker PS, et al. Venetoclax combined with decitabine or

azacitidine in treatment-naive, elderly patients with acute

myeloid leukemia. Blood 2019;133:7-17.

30. Cortes JE, Heidel FH, Hellmann A, Fiedler W, Smith BD,

Robak T, et al. Randomized comparison of low dose cytarabine

with or without glasdegib in patients with newly diagnosed

acute myeloid leukemia or high-risk myelodysplastic syndrome.

Leukemia 2019;33:379-89.

31. Brunetti C, Anelli L, Zagaria A, Specchia G, Albano F. CPX-

351 in acute myeloid leukemia: can a new formulation maximize

the ecacy of old compounds? Expert Rev Hematol 2017;10:

853-62.

32. Lancet JE, Uy GL, Cortes JE, Newell LF, Lin TL, Ritchie EK, et

al. CPX-351 (cytarabine and daunorubicin) Liposome for

Injection Versus Conventional Cytarabine Plus Daunorubicin

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

425

in Older Patients With Newly Diagnosed Secondary Acute

Myeloid Leukemia. J Clin Oncol 2018;36:2684-92.

33. Godwin CD, Gale RP, Walter RB. Gemtuzumab ozogamicin

in acute myeloid leukemia. Leukemia 2017;31:1855-68.

34. Petersdorf SH, Kopecky KJ, Slovak M, Willman C, Nevill T,

Brandwein J, et al. A phase 3 study of gemtuzumab ozogamicin

during induction and postconsolidation therapy in younger

patients with acute myeloid leukemia. Blood 2013;121:

4854-60.

35. Jen EY, Ko CW, Lee JE, Del Valle PL, Aydanian A, Jewell C,

et al. FDA Approval: Gemtuzumab Ozogamicin for the Treatment

of Adults with Newly Diagnosed CD33-Positive Acute Myeloid

Leukemia. Clin Cancer Res 2018;24:3242-6.

36. Castaigne S, Pautas C, Terre C, Raoux E, Bordessoule D,

Bastie JN, et al. Eect of gemtuzumab ozogamicin on survival of

adult patients with de-novo acute myeloid leukaemia (ALFA-

0701): a randomised, open-label, phase 3 study. Lancet

2012;379:1508-16.

37. Amadori S, Suciu S, Selleslag D, Aversa F, Gaidano G, Musso

M, et al. Gemtuzumab Ozogamicin Versus Best Supportive Care

in Older Patients With Newly Diagnosed Acute Myeloid

Leukemia Unsuitable for Intensive Chemotherapy: Results of

the Randomized Phase III EORTC-GIMEMA AML-19 Trial.

J Clin Oncol 2016;34:972-9.

38. Yang X, Wang J. Precision therapy for acute myeloid leukemia.

J Hematol Oncol 2018;11:3.

39. Hofmann S, Schubert ML, Wang L, He B, Neuber B, Dreger

P, et al. Chimeric Antigen Receptor (CAR) T Cell erapy in

Acute Myeloid Leukemia (AML). J Clin Med2019;8(2). pii: E200.

40. Pyzer AR, Avigan DE, Rosenblatt J. Clinical trials of dendritic

cell-based cancer vaccines in hematologic malignancies. Hum

Vaccin Immunother 2014;10:3125-31.

41. DeStefano CB, Hourigan CS. Personalizing initial therapy in

acute myeloid leukemia: incorporating novel agents into clinical

practice. er Adv Hematol 2018;9:109-21.

42. Lohse I, Statz-Geary K, Brothers SP, Wahlestedt C. Precision

medicine in the treatment stratication of AML patients:

challenges and progress. Oncotarget 2018;9:37790-7.

43. Glass JL, Hassane D, Wouters BJ, Kunimoto H, Avellino R,

Garrett-Bakelman FE, et al. Epigenetic Identity in AML Depends

on Disruption of Nonpromoter Regulatory Elements and Is

Aected by Antagonistic Eects of Mutations in Epigenetic

Modiers. Cancer Discov 2017;7:868-83.

44. Burd A, Levine RL, Shoben A, Mims AS, Borate U, Stein EM,

et al. Initial Report of the Beat AML Umbrella Study for

Previously Untreated AML: Evidence of Feasibility and Early

Success in Molecularly Driven Phase 1 and 2 Studies. Blood

2018;132:559.

45. Swords RT, Azzam D, Al-Ali H, Lohse I, Volmar CH, Watts JM,

et al. Ex-vivo sensitivity proling to guide clinical decision making

in acute myeloid leukemia: A pilot study. Leuk Res 2018;64:

34-41.

46. Vaughn JE, Shankaran V, Walter RB. Trends in Clinical

Benets and Costs of Novel erapeutics in AML: at What

Price Does Progress Come? Curr Hematol Malig Rep 2019;14:171-78.

47. Ehinger M, Pettersson L. Measurable residual disease testing

for personalized treatment of acute myeloid leukemia. APMIS

2019;127:337-51.

48. Ravandi F, Walter RB, Freeman SD. Evaluating measurable

residual disease in acute myeloid leukemia. Blood Adv 2018;2:

1356-66.

49. Schmitt MW, Loeb LA, Salk JJ. e inuence of subclonal

resistance mutations on targeted cancer therapy. Nat Rev Clin

Oncol 2016;13:335-47.

50. Hirsch P, Zhang Y, Tang R, Joulin V, Boutroux H, Pronier

E, et al. Genetic hierarchy and temporal variegation in the clonal

history of acute myeloid leukaemia. Nat Commun. 2016;7:12475.

51. Schuurhuis GJ, Heuser M, Freeman S, Bene MC, Buccisano

F, Cloos J, et al. Minimal/measurable residual disease in AML:

a consensus document from the European LeukemiaNet MRD

Working Party. Blood 2018;131:1275-91.

52. Shlush LI. Age-related clonal hematopoiesis. Blood 2018;131:

496-504.

53. Jan M, Snyder TM, Corces-Zimmerman MR, Vyas P, Weissman

IL, Quake SR, et al. Clonal evolution of preleukemic hematopoietic

stem cells precedes human acute myeloid leukemia. Sci Transl

Med2012;4:149ra118.

54. Miller LH, Qu CK, Pauly M. Germline mutations in the bone

marrow microenvironment and dysregulated hematopoiesis.

Exp Hematol 2018;66:17-26.

55. Mashima K, Oh I, Ikeda T, Toda Y, Ito S, Umino K, et al.

Role of Sequential Monitoring of WT1 Gene Expression in

Patients With Acute Myeloid Leukemia for the Early Detection

of Leukemia Relapse. Clin Lymphoma Myeloma Leuk

2018;18:e521-e7.

56. Frairia C, Aydin S, Audisio E, Riera L, Aliberti S, Allione B,

et al. Post-remissional and pre-transplant role of minimal

residual disease detected by WT1 in acute myeloid leukemia:

A retrospective cohort study. Leuk Res 2017;61:10-7.

57. Jacobsohn DA, Loken MR, Fei M, Adams A, Brodersen LE,

Logan BR, et al. Outcomes of Measurable Residual Disease in

Pediatric Acute Myeloid Leukemia before and aer Hematopoietic

Stem Cell Transplant: Validation of Dierence from Normal

Flow Cytometry with Chimerism Studies and Wilms Tumor 1

Gene Expression. Biol Blood Marrow Transplant 2018;24:2040-6.

58. Shumilov E, Flach J, Joncourt R, Porret N, Wiedemann G,

Angelillo-Scherrer A, et al. Critical evaluation of current

molecular MRD strategies including NGS for the management

of AML patients with multiple mutations. Hematol Oncol.2019

Mar 6. doi: 10.1002/hon.2603. [Epub ahead of print]

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

426

Case Report

SMJ

Sakanus Vijintanawan, D.D.S.*, Kumar KC, B.D.S.*, Anchisa Aittiwarapoj, D.D.S.*, Natthamet Wongsirichat,

B.Sc., D.D.S.**

*Master degree students international program of Oral Maxillofacial Surgery, Department of Oral Maxillofacial Surgery, Faculty of Dentistry, Mahidol

University, **Department of Oral Maxillofacial Surgery, Faculty of Dentistry, Mahidol University, Bangkok 10400, and Consultant of International

College of Dentistry Committee, Walailuk University International College of Dentistry, Bangkok 10400, ailand.

Hybrid Central Odontogenic Fibroma with Central

Giant Cell Granuloma Like Lesion ; A Case Report

and Review of the Literature

ABSTRACT

e combination of central odontogenic broma and central giant cell granuloma is extremely rare and has

been reported in only 46 cases worldwide in previous articles. e lesion is called a hybrid central odontogenic

broma with central giant cell granuloma like lesion (HCOF-CGCG) due to a histopathological characteristics of

multinucleated giant cell and odontogenic epithelium in bro-cellular background. We report a case of HCOF-

CGCG in a 27-year old ai male patient who was identied as the rst case in ailand. Radiographic examination

showed a well-dened radiolucency lesion at posterior mandibular area. Enucleation and curettage were done and

microscopic study revealed a bro-cellular proliferating tissue containing numerous strands, cord of odontogenic

epithelium, and numerous multinucleated giant cells. is report also included a literature review of the previous

45 cases.

Keywords: Central odontogenic broma; central giant cell granuloma; hybrid (Siriraj Med J 2019; 71: 426-431)

Corresponding author: Natthamet Wongsirichat

E-mail: natthamet.won@mahidol.ac.th, natthamet.wo@mail.wu.ac.th

Received 7 March 2019 Revised 31 July 2019 Accepted 5 August 2019

ORCID ID: http://orcid.org/0000-0003-3005-2680

http://dx.doi.org/10.33192/Smj.2019.63

INTRODUCTION

The central odontogenic fibroma (COF) is an

uncommon disease found in gnathic area.

In 2017, WHO

denes it as an ectomesenchymal tumor with a mature

brous connective tissue lesion, which usually combines

with an odontogenic epithelium.

It can occur in wide-

ranging age distribution and slight female predilection

aecting relatively equal frequency in both jaws.

e lesion in maxilla usually occurs anterior to

rst molar and about half of all mandibular lesions are

found posterior to rst molar.

e small COF is oen

asymptomatic and presents as well-dened unilocular

radiolucency whereas the larger lesion may be associated

with a pain, a bony expansion, a root resorption, a divergence

of adjacent tooth roots, a loosening of teeth and presents

as multilocular radiolucency.

3,4

e histological features of the COF are similar to other

tumors including enlarging dental follicle, odontogenic

myxoma and desmoplastic broma.

5-7

Histopathological

features of COF compose principally of mature collagen

bers along with numerous interspersed broblasts. In

addition, presence of small nests or strands of inactive

odontogenic epithelium are found to be a variable feature.

It also can present with histological variations such

as granular cell, giant cell and pleomorphic broblast

combination.

8,9

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

427

Central giant cell granuloma (CGCG) is a benign

neoplasm of jaw comprising a giant cell reparative

granuloma.

e CGCG occurs in around 10% of jaw

lesions and is reported to occur oen in patients aged

below 20 years with female predilection.

2,10

e lesion

usually develops in the anterior of the mandible with

characteristics of slow growing, asymptomatic, and

expansible lesion in nature.

e radiographic nding

shows the well-defined radiolucency, but it can be

multilocular in advanced lesion. About 30% of CGCGs

exhibit aggressive behavior associated with pain, root

resorption, displacement of adjacent tooth, and cortical

perforation.

Histopathological feature of the CGCG is shown

with a non-encapsulated proliferation of mononuclear

spindle-shaped and polygonal cells with osteoclast-type

multinucleated giant cells in a vascular background with

hemorrhage and haemosiderin pigment.

10,11

Likewise,

there is a similar histological feature between CGCG

and giant cell lesion of bone comprising cherubism,

hyperparathyroidism, aneurysmal bone cyst and Noonan

syndrome.

12-14

In 1992, Allen reported the rst case of hybrid central

odontogenic broma with central giant cell granuloma like

lesion (HCOF-CGCG). e histological nding revealed

the bro-cellular tissue, odontogenic epithelium, and

distribution of the giant multinucleated cells.

HCOF-

CGCG has been reported in only 45 cases worldwide in

history and considered as a rare lesion. is study aimed

to report a rare case and provide a literature review of

HCOF-CGCG.

CASE REPORT

A 27-year-old ai male was referred by general

dentist for management of a radiolucent lesion which

was accidentally found on a radiographic examination

prior to a third molar surgery. According to a history

taking, the patient did not notice the presence of lesion

in his lower le jaw and did not report numbness or

pain in any area. He also underwent a tooth extraction

in this area 3 years ago. Clinical examination found a

buccal expansion with obliteration of mucobuccal fold

and a root fragment of mandibular le rst molar. e

teeth involved in this area were vital in response to an

electric pulp test.

e routine radiograph (Fig 1A) which was obtained

3 years ago did not present any sign of the lesion, but

it presented an endodontic lesion of a rst mandibular

le molar. Preoperative radiograph (Fig 1B) revealed a

well-dened unilocular radiolucent lesion, measuring

approximately 18x18 mm, extending from a le mandibular

second premolar to le mandibular second molar. e

lesion was superimposed with a mental foramen and

inferior alveolar canal. e root fragment was identied

as a radiopaque mass at superior border of the lesion.

e dierential diagnosis included a radicular cyst,

odontogenic keratocyst, and unicystic ameloblastoma.

e treatment plan was enucleation by curettage of the

total lesion. During the surgical procedure, we found

a buccal bone perforation and so tissue mass within

the lesion detached from the surrounding bone. Both

inferior alveolar nerve and mental nerve were exposed

during the surgery, the inferior alveolar artery was also

injured. e bleeding was controlled using bone wax.

e lesion was completely removed. e macroscopic

appearance (Fig 2) of the lesion was red-pink so tissue

mass with hemorrhage. e primary wound closure with

Fig 1A. e routine Panoramic Radiograph 3 years ago.

Fig 1B. e preoperative Panoramic Radiograph.

Fig 1C. e 6-month follow-up Panoramic Radiograph.

Vijintanawan et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

428

Case Report

SMJ

3-0 silk was done. At one-week follow-up, the patient

presented with slight numbness at his lower le lip.

Multiple 4um-thick formalin-xed paran-embedded

sections of the specimen were stained with hematoxylin and

eosin. Microscopic examination (Fig 3A, 3B) revealed that

it mainly composed of a brocellular proliferating tissue

containing numerous strands, and cord of odontogenic

epithelium. Numerous multinucleated giant cells were

seen to be dispersed into the connective tissue with a

whorled pattern. Small areas of the connective tissue

containing odontogenic epithelium and multinucleated

giant cells were separated together. Areas of myxoid

stroma and hemorrhage were also observed. Immuno-

histochemical stains of cytokeratin 19 was used to conrm

the structure that mimicked odontogenic epithelium.

e immuno-histochemical was positive on cytoplasm

of odontogenic epithelial cells (Fig 3C). Morphology was

compatible with HCOF-CGCG. At six-months follow up,

no evidence of recurrence occurred, both clinically and

radiographically (Fig 1C). e numbness was completely

resolved.

DISCUSSION

e HCOF-CGCG is a rare combination and was

rst reported by Allen in 1992.

e previous reports and

the international conferences so far found the number

of cases has increased gradually and is reckoned to be

46 cases including this case. e available information

of all cases are presented in Table 1. e lesion occurs

mostly in second and third decade of life with a female

predilection and is more frequent in a tooth bearing

posterior mandibular area (86%) alongside the buccal

expansion. e tooth displacement is also found in

some cases.

16-18

A few cases report the relation of the

lesion with an orthodontic, root canal treated tooth,

embedded tooth, and cherubism.

15,19,20

ere are only

2 cases reporting pain whilst the remaining have been

asymptomatic.

9,17

Radiographically, 60% presented as a well-dened

unilocular radiolucency and the rest of the lesions presented

with multilocular radiolucency. e ill-dened border

characteristic can be found in some cases.

e extensive

Fig 2. e specimen of the lesion that was completely enucleated.

Fig 3A. e histopathology of the combined lesion was composed

of areas of giant cell granuloma and odontogenic broma. e areas

of sole giant cell granuloma (red arrow) and odontogenic broma

(blue arrow) areas were demonstrate at periphery of the lesion.

Fig 3B. e histopathology of the areas of giant cell granuloma

presented scattering multinucleated giant cells are in the inamed

brous tissue.

Fig 3C. e histopathology a cytokeratin 19 staining was presented

in cytoplasm of odontogenic epithelial rests.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

429

Case Author/Year Age/Sex Location Clinical features Radiographic Treatment Recurrence

No.     ndings

1-3 Allen, 1992

3F; 66/14/30 Mandible premolar/molar area Root canal treated tooth involvement, Multi/uniloclar Curettage Recurrence

orthodontic treatment, expansion

4.-6. Fowler, 1993

Reported 3 cases but no clinical data available

7-14 Odell, 1997

5F;5/20/21/22/39/43/50, Maxilla, posterior mandible Buccal expansion, history of extraction, Unilocular Curettage Recurrence

M;11 rapid growth, buccal perforation, tooth mobility

15. Taylor, 1999

17/F Mandible/ canine/premolar Buccal expansion, tooth displacement, Unilocular Curettage None

area thining cortex

16. Kruse-Losler, 22/F Mandible/ from right to left Lingual and inferior expansion Multilocular, Surgical None

2006

molar area scalloped border excision

17-23 Pontes, 2008

Reported 7 cases with 5 males and 2 females Recurrence

at mandible 3 cases

24-30 Tosios, 2008

6M; 18/50/73/15/59/25, Mandible/ premolar/molar Cherubism, perforation Multi/uniloclar, Surgical None

F;20 area premolar apex excision

31. Younis, 2008

57/F Mandible/ premolar/molar area Buccal expansion Unilocular Curettage None

32. Marina, 2008

24/F Mandible/ from right to left Mandibular swelling progressively for N/A Curettage and None

molar area 8 years, tooth displacement reconstruction

33. Bologna-Molina, 14/M Mandible/ premolar/molar area Buccal expansion Multilocular Resection and None

2011

reconstruction

34. Castillo, 2011

14/M Mandible/ molar area Expansion, tenderness, tooth displacement Unilocular Curettage None

35-36 Eversole, 2011

42/F, 27/F Body of mandible, Body of N/A N/A N/A N/A

mandible and ramus

37-38 Mosqueda- 14/M, 14/M Mandible/ molar area, Both buccal and lingual expansion, Unilocular, Surgical None

Taylor, 2011

Mandible/ premolar/molar area tooth displacement, tenderness Multilocular excision

on palpation, 6 months evolution

39. Damm, 2013

75/F Mandible/ anterior N/A Unilocular Curettage None

40. Eliot, 2015

22/F Mandible/ canine/ Swelling at mandible and Multilocular Curettage None

premolar/molar area submandibular area

41. Schultz, 2017

12/F Mandible/ anterior Asymptomatic Multilocular Enucleation None

and selective

extraction

42. Leite, 2017

28,29

42/F Mandible/ molar area N/A N/A Surgical excision None

43-46 Upadhyaya, 2M; 10/62, F;63 Mandible anterior, posterior Buccal and lingual expansion, Unilocular Curettage None

2018

embedded tooth

46. Current 27/M Mandible premolar/ Asymptomatic, buccal expansion, Unilocular Curettage None

case report molar area progression less than 3 years, retained root

TABLE 1. Information of previously reported cases.

N/A is dened as no available data.

Noted that cases from Fowler and Hassan provided insucient information but the cases are still accumulated the total number of the cases.

Vijintanawan et al.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

430

Case Report

SMJ

lesions have been reported in a few cases which included a

sinus involvement and crossing the midline.

Taylor et al.,

reported that the lesion was distinctive from surrounding

bone and had no encapsulation.

Also, some authors

reported that the lesion caused a bony cortex thinning

and perforation.

9,19,21

Most of COF and CGCG, were

treated by curettage. One case was treated by resection

and reconstruction to prevent a mandibular fracture.

e recurrent lesions reported 17% which were treated

by curettage. Similar to the nature of COF and CGCG,

recurrences have been observed with slightly higher

rate in the aggressive behavior, ill-dened margin and

multilocular lesions.

1,10,11

Odell et al., stated that the

lesion tends to behave like the CGCG much more than

the COF due to its prevalence and behavior.

Histo-pathologically, several cases of the HCOF-

CGCG were found to be contained as densely collagenous

to bromyxoid stroma with several nests of inactive

odontogenic epithelium. e areas have been revealed

to merge independently with a variably plump to narrow

spindle-shaped brocellular connective tissue containing

multinucleated giant cells.

18,21,23

Similarly, in this reported

case, features of HCOF-CGCG like components and

the odontogenic epithelial nests in the mature brous

connective tissue were seen histologically.

e pathogenesis of hybrid lesion is still unclear.

ere are 3 theories mentioning the origin.

1. e rst theory was proposed by Allen et al, who

rst reported the hybrid lesion. e theory suggested

“collision tumor” which is defined by synchronous

occurrence of COF and CGCG. is theory seems seldom

to happen and is opposed by recurrent lesion. Almost

all the recurrencea presented with combination of COF

and CGCG. e combination seems to be a specic

characteristic of it.

2. e second theory is “primary CGCG induced

secondary COF”. is theory stated the cells present within

CGCG produce factors like growth factors, chemokines,

and cytokines which induce the growth of odontogenic

cell line and can result in development of COF.

3. e third one is opposite to the second theory.

e primary tumor is COF and the trauma or other

stimulus induces a giant cell reaction.

Now, it cannot certainly conclude the origin of

this hybrid lesion. The evidences of lesion vicinity,

histopathology, and recurrence are a conjecture of the

pathogenesis.

According to this current case, the lesion occurred

in an early adult male which is dierent from the other

studies. e clinical features were similar to the other

reports which are buccal expansion and asymptomatic.

e radiographic nding was well-dened with corticated

border which corresponded to the characteristics during

removal. e lesion showed no inltration pattern and

was detached from the surrounding bone. It caused the

buccal bone thinning which is similar other reported

cases. In addition, the lesion progressed within 3 years

since the past radiograph conrmed the presence of the

lesion. In surgical point of view, this intra-bony lesion

was easy to remove by enucleation and curettage.

Our opinion suggested a possibility that this lesion

could be a “collision tumor”. e growth of collision

tumor arose from the synchronous occurrence of COF

in the same area as a CGCG.

9,21

is theory correlated

with our case due to the presence of intrabony mixed-

up area of these two lesions. is case displayed three

dierent areas including: odontogenic broma, giant

cell granuloma and two merged lesions. e sole areas of

COF and CGCG were present only at the border of the

specimen. e central area of the specimen demonstrated

mixed histopathologic features of odontogenic broma

and giant cell granuloma-like lesion. In conclusion,

the HCOF-CGCG was considered as a rare oral lesion.

e nature of the lesion was mostly non-aggressive

behavior. e treatment by enucleation and curettage

seemed to be promising. e histopathological feature

was characterized by the presence of the bro-cellular

proliferating tissue containing numerous strands, cord of

odontogenic epithelium, and numerous multinucleated

giant cells. e presence of COF and CGCG could be

merged or independent.

ACKNOWLEDGMENTS

e authors would like to thank the sta and dental

assistants including colleagues and co-workers in the

Department of Oral and Maxillofacial Surgery, Faculty

of Dentistry, Mahidol University.

Funding: is research did not receive any nancial

support

Conict of interest: All authors declared that there is

no conict of interest in this research

Ethics approval: Not required

REFERENCES

1. Correa Pontes FS, Lacerda de Souza L, Paula de Paula L, de

Melo Galvão Neto E, Silva Gonçalves PF, Rebelo Pontes

HA.Central odontogenic broma: An updated systematic

review of cases reported in the literature with emphasis on

recurrence inuencing factors. J Craniomaxillofac Surg 2018;

46:1753-7.

2. El-Naggar AK, Chan JKC, Takata T, Grandis JR, Slootweg PJ.

e fourth edition of the head and neck World Health Organization

blue book: editors’ perspectives. Hum Pathol. 2017;66:10-12.

Volume 71, No.5: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

431

3. Neville BW, Damm DD, Allen CM, Chi AC. Oral and maxillofacial

pathology. 4

ed. Elsevier: St. Louis; 2016.

4. Wright JM, Vered M. Update from the 4

edition of the

World Health Organization classication of head and neck

tumours: odontogenic and maxillofacial bone tumors. Head

Neck Pathol 2017;11:68-77.

5. Barnes L, Eveson JW, Reichart P, Sidransky D, editors: Pathology

and genetics of head and neck tumors. Philipsen HP, Reichart

PA, Sciubba JJ, Van der Waal I, series editors. Odontogenic

tumors World Health Organization classication of tumors.

Lyon, France: IARC Press, 2005. p. 315.

6. Kim J, Ellis GL. Dental follicular tissue: misinterpretation as

odontogenic tumors. J Oral Maxillofac Surg 1993;51:762-7.

7. Ikeshima A, Utsunomiya T. Case report of intra-osseous

broma: a study on odontogenic and desmoplastic bromas

with a review of the literature. J Oral Sci 2005;47:149-57.

8. Eversole LR. Odontogenic broma, including amyloid and

ossifying variants. Head Neck Pathol 2011;5:335-43.

9. Mosqueda Taylor A, Bermúdez Flores V, Díaz Franco MA.

Combined central odontogenic broma and giant cell granuloma-

like lesion of the mandible: report of a case and review of the

literature. J Oral Maxillofac Surg 1999;57:1258-62.

10. de Lange J, van den Akker HP, van den Berg H. Central giant

cell granuloma of the jaw: a review of the literature with

emphasis on therapy options. Oral Surg Oral Med Oral Pathol

Oral Radiol Endod 2007;104:603-15.

11. Kruse-Losler B, Diallo R, Gaertner C, Mischke KL, Joos U,

Kleinheinz J. Central giant cell granuloma of the jaws: a clinical,

radiologic, and histopathologic study of 26 cases. Oral Surg

Oral Med Oral Pathol Oral Radiol Endod 2006;101:346-54.

12. Sciubba JJ FJ, Kahn LB, editors. Atlas of tumor pathology:

tumors and cysts of the jaw. Washington, DC: Armed Forces

Institute of Pathology: 2001 p.161-6.

13. Regezi JA. Odontogenic cysts, odontogenic tumors, broosseous,

and giant cell lesions of the jaws. Mod Pathol 2002;15:331-41.

14. Wolvius EB, de Lange J, Smeets EE, van der Wal KG, van

den Akker HP. Noonan-like/multiple giant cell lesion syndrome:

report of a case and review of the literature. J Oral Maxillofac

Surg 2006;64:1289-92.

15. Allen CM, Hammond HL, Stimson PG. Central odontogenic

broma, WHO type: A report of three cases with an unusual

associated giant cell reaction. Oral Surg Oral Med Oral Pathol

Oral Radiol 1992;73:62-6.

16. de Lima MDe D, de Aquino Xavier FC, Vanti LA, de Lima

PS, de Sousa SC. Hybrid central giant cell granuloma and central

odontogenic broma-like lesion of the mandible. Otolaryngol

Head Neck Surg 2008;139:867-8.

17. Cortés-Castillo G, Liceaga-Reyes R, Mosqueda-Taylor A. Lesión

mandibular inusual de broma odontogénico central combinado

con granuloma central de células gigantes mandibular. Rev

Odont Mex 2011;15:126-31.

18. Mosqueda-Taylor A, Martínez-Mata G, Carlos-Bregni R,

Agustin-Vargas P, Toral-Rizo V, Cano-Valdéz AM, et al. Central

odontogenic broma: new ndings and report of a multicentric

collaborative study. Oral Surg Oral Med Oral Pathol Oral

Radiol Endod 2011;112:349-58.

19. Tosios KI, Gopalakrishnan R, Koutlas IG. So-called hybrid

central odontogenic broma/central giant cell lesion of the

jaws. A report on seven additional cases, including an example

in a patient with cherubism, and hypotheses on the pathogenesis.

Head Neck Pathol 2008;2:333-8.

20. Upadhyaya JD, Cohen DM, Islam MN, Bhattacharyya I. Hybrid

central odontogenic broma with giant cell granuloma like

lesion: a report of three additional cases and review of the

literature. Head Neck Pathol 2018;12:166-74.

21. Odell E, Lombardi T, Barrett AW, Morgan P, Speight PM. Hybrid

central giant cell granuloma and central odontogenic broma‐

like lesions of the jaws. Histopathology 1997;30:165-71.

22. Bologna Molina R, Pacheco Ruiz L, Mosqueda Taylor A,

Huesca Ramirez HG, Ponce Lonato JA, González-González

R. Central Odontogenic Fibroma combined with Central

Giant Cell lesion of the mandible. Immunohistochemical

prole. J Clin Exp Dent 2011;3:e348-51.

23. Younis RH, Scheper MA, Lindquist CC, Levy BL. Hybrid central

odontogenic broma with giant cell granuloma-like component:

case report and review of literature. Head Neck Pathol 2008;2:

222-6.

24. Fowler C, Tomich C, Brannon R, Houston G. Central odontogenic

broma: clinicopathologic features of 24 cases and review of

the literature. Oral Surg Oral Med Oral Pathol Oral Radiol

Endod 1993;76:587.

25. Damm DD. Interradicular radiolucency. Hybrid giant cell

granuloma and odontogenic broma. Gen Dent 2013;61:77-78.

26. Eliot C, Kessler HP. Clinical Pathologic Conference Case 1:

A Multilocular Radiolucency in the Posterior Mandible. Oral

Surg Oral Med Oral Pathol Oral Radiol 2015;119:e289-92.

27. Schultz K, Rosebush M. Clinical pathologic conference case 3.

Annual Meeting of the American Academy of Oral and

Maxillofacial Pathology. Newport, Rhode Island; 2017.

28. Leite T, Mendes E, Abrahao A, Andrade B, Canedo N, Agostini

M, etal. Central odontogenic broma: report of three new

cases from Brazil. Annual Meeting of the American Academy

of Oral & Maxillofacial Pathology. Newport, Rhode Island;

2017.

29. Leite T, Mendes E, Abrahao A, Andrade B, Canedo N, Agostini

M, et al. Central odontogenic broma: report of three new

cases from Brazil. Oral Surgery, Oral Medicine, Oral Pathology

and Oral Radiology 2017;124(3):e229-e30.

Vijintanawan et al.