Volume 71, Number 2, March-April 2019
Siriraj Medical Journal
SMJ
ISSN 2629-995X
ORIGINAL ARTICLE
95 A Comparison of the Bacterial Culture Results of Maxillary
Sinus Mucosa and Pus Collections for Chronic Maxillary
Rhinosinusitis
Anan Bedavanija, et al.
102 Impact Assessment of Smell and Taste Disorders on
Quality of Life in Thais Using the SF-36 Health Survey
(Thai version)
Bannapuch Pinkaew, et al.
110 Risk Scoring System for the Prediction of Postpartum
Blood Loss over 300 mL at Chiang Rai Regional Hospital
Natakorn I-Tuporn, et al.
117 The Evaluation of Posterior Cingulate Gyrus by Diffusion
Tensor Imaging in Alzheimer’s Disease Patients Compared
with Normal Control Subjects
Chanon Ngamsombat, et al.
123 Comparison of Academic Outcomes of Epidemiology and
Biostatistics Course between Traditional Lecture and
Flipped Classroom Blended Approach in 2
nd
Year Medical
Students: A Retrospective Cohort Study
On-anong Binsomprasong, et al.
127 Re – evaluation of Rinne Test with Aluminum Alloy
Tuning Fork 256 Hz and 512 Hz
Patcharaporn Phonpornpaiboon, et al.
131 Hearing and Balance Survey in Thai Elders
Samut Chongvisal, et al.
143 Scintigraphic Detection of Splenosis and Accessory Spleen
by Using Tc-99m Labelled Denatured Red Blood Cells
Apichaya Claimon, et al.
150 Does Acute Kidney Injury Condition Affect Revised BAUX
Score in Predicting Mortality in Major Burn Patients?
Kusuma Chinaroonchai, et al.
158 An Association of Cryptococcus Neoformans/C. gattii
Genotype and HIV Status in Asia: A Systematic Review
Popchai Ngamskulrungroj, et al.
165 Clinical Application of Toothbrush Technique for
Specimen Collection of Tinea Capitis
l
Lalita Matthapan, et al.
REVIEW ARTICLE
168 Gummy Smile: A Review of Etiology, Anifestations, and
Treatment
Basel Mahardawi, 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 Bannapuch Pinkaew, 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 Bualo, 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 Oalmologia 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 Oce 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.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
123
Original Article
SMJ
On-anong Binsomprasong, M.Ed.*, Lokachet Tanasugarn, M.D.**, Pongsakorn Buraphat, M.D.***, Mayuree
Homsanit, M.D., Ph.D.**
*Education Department, **Department of Preventive and Social Medicine, ***Department of Pharmacology, Faculty of Medicine Siriraj Hospital,
Mahidol University, Bangkok 10700, ailand.
Comparison of Academic Outcomes of Epidemiology
and Biostatistics Course between Traditional
Lecture and Flipped Classroom Blended approach
in 2
nd
Year Medical Students: A Retrospective
Cohort Study
Corresponding author: Mayuree Homsanit
E-mail: mayuree.hom@mahidol.ac.th
Received 1 September 2017 Revised 16 February 2018 Accepted 23 May 2018
ORCID ID: http://orcid.org/0000-0002-9838-5636
http://dx.doi.org/10.33192/Smj.2019.19
ABSTRACT
Objective: To compare academic outcomes between integrated ipped classroom (FC) and solely traditional lecture
(TL) teaching approach among 2
nd
year medical students in Epidemiology and Biostatistics course.
Methods: General information and academic achievements including grade point average, scores of the basic
statistics course and scores of Epidemiology and Biostatistics course were extracted from the database of the Education
Department. e records were excluded if they belonged to repeated or transferred students. e students in 2014
and 2015 academic year represented the TL group and the FC group respectively. e main outcome was the
examination score from the two groups which was converted to a comparable scale out of a hundred.
Results: Six-hundred twenty-eight records were included in the analysis. Based on independent t-test, the TL group
showed higher mean score as compared to the FC group (p-value < 0.01). Further comparison using ANCOVA
adjusted for age, gender, hometown location, high school location, grade point average and score of a basic statistics
course showed a higher score in the TL group (p-value < 0.01).
Conclusion: TL yielded higher examination score than FC from this retrospective study.
Keywords: Flipped classroom; traditional lecture; epidemiology and biostatistics (Siriraj Med J 2019;71: 123-126)
INTRODUCTION
According to a competency-based framework,
paradigm shi in medical education transforms traditional
lecture (TL) toward other methods to create lifelong
learners.
1,2
Depending on co-operation of both students
and teachers, active learning curriculum demonstrated
long-lasting knowledge by developing intrapersonal skills
to solve future problems.
3,4
e ipped classroom (FC) is an educational method
that reverses a student role from a passive learner to an
active learner. Certain self-studied resources such as
recommended textbooks, articles or even video records
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
124
are arranged for the students prior to each class. During
the class, activities were designed for application or
integration of the given assignments. Teachers are not
lecturers but they are coaches or supporters.
5
us,
engagement of both students and teachers are highly
required to achieve the excellent outcome.
6,7
e Faculty of Medicine Siriraj Hospital, Mahidol
University, Bangkok, ailand (“the faculty”) provides the
Epidemiology and Biostatistics course for the second-year
medical students. Before 2014 academic year, this course
mainly consisted of one-way TL in which instructors
were required to deliver the lecture to a large group of
audiences (about 300 students). However, “Outcome-
Based Curriculum” was implemented in 2014. e new
curriculum emphasized the importance of active learning
methods and encouraged to eliminate all previous passive
teachings. us, reduction in TL and integration of FC
were implemented to all courses.
Several studies demonstrated that FC could improve
knowledge as compared before and aer the class
8
, or
compared with TL.
9
Other evidences
10-12
demonstrated
efficacy and advantage of this educational method,
although one systematic review focused on medical
education suggested that there is no strong evidence
for the eectiveness of FC.
13
Inconsistency in learning
outcome of FC may come from dierence in participants’
background, learning culture or heterogeneity of education
tools.
As far as we know, there is no evidence in ailand
comparing learning outcome between TC and FC in
Epidemiology and Biostatistics. e objective of this study
was to evaluate changes in Epidemiology and Biostatistics
learning outcomes aer implementing FC in our institute.
MATERIALS AND METHODS
is study was conducted as a retrospective cohort
design. e study was approved by Siriraj Institutional
Review Board (SIRB), Faculty of Medicine Siriraj Hospital,
Mahidol University (Si 536/2558).
Data retrieval and management
General information and academic outcome of the
medical students in 2014 and 2015 academic year were
acquired from the Education Department, Faculty of
Medicine Siriraj Hospital, Mahidol University, ailand.
e background information of each student included
gender, age, hometown location, high school location,
grade point average (GPA), scores of a basic biostatistics
course in pre-medical year and score of Epidemiology and
Biostatistics course. Any student with repeated grade or
transferred credit was excluded from the analysis. We
used data of all eligible students (students enrolled in
2014-2015 academic year) for analysis.
Traditional lecture course (TL) vs. Flipped Classroom
(FC) course
e 2014 cohort (total students = 319) received TL
method. e TL course consisted of thirteen hours of one-
way lecture and six hours of workshops. e evaluation
consisted solely of examination score. e FC method
was implemented in the 2015 group (total students =
313). e course provided eleven hours of interactive
lecture and four hours of workshop. e FC course also
included one additional online module on revisiting basic
statistics which was about three hours. Additionally,
two formative evaluations were incorporated in the FC
course which were not scored, but were required to be
eligible to take the nal examination. Both courses were
designed based on the same set of learning outcome
from the Medical Competency Assessment Criteria for
National License 2012 of ailand.
14
Also, the same set
of instructors taught in both courses on each topic. Since
the FC group score was categorized as a multiple-choice
question (MCQ), attitude, essay, and coursework, we only
compared MCQ and essay score with the total score of
the TL group (100% of MCQ). us, the score of the FC
group (50% of MCQ and 50% of short answer question)
was converted to a scale of a hundred to be comparable
with the TL score. Test analysis of both courses revealed
that their MCQ test property fell into the same category
with very good average diculty index and good average
discrimination index.
Statistical analysis
All analyses were performed with PASW Statistics
for Windows version 18 (SPSS Inc., Chicago, Ill, USA).
Descriptive statistics were presented as mean ± SD and n
(%). Independent t-test was used to compare the dierence
between mean of scores. ANCOVA was used to further
compare the mean dierence with adjustment for age,
gender, hometown, high school, grade point average
and a score of basic statistics course as covariates.
Power analysis
Because we recruited data of all medical students
who met eligibility criteria for this study, no sample size
was calculated. We performed power analysis using two-
sided test for mean dierence. Based on total students
enrolled in the TL and FC group (319 vs. 313 students),
we estimated that we could retrieve data without missing
at 95% of total enrolled students. Using pooled average
and standard deviation (SD) of examination score from
Binsomprasong et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
125
Original Article
SMJ
a previous study
15
, power of the study was estimated to
be > 90% at a α = 0.05 (dierence of pooled means =
4.5, pooled SD of the control group = 5.94, pooled SD
of the experimental group = 6.13).
RESULTS
Six-hundred and thirty-two records were eligible
for the study which comprised of 319 records in the
traditional lecture group and 313 records in the ipped
classroom group. In the traditional lecture group, three
records were excluded: two records belonged to repeated
grade students and one record was a transferred student.
For the ipped classroom group, one record belonged
to a transfer student and was excluded. erefore, a
total of 628 records were included in the analyses. e
demographic data of both groups is presented in Table 1.
e independent t-test revealed that the TL group
had a higher score than the FC group (p-value < 0.01).
Subsequently, ANCOVA with adjustment for covariates
showed the same outcome (p-value < 0.01). e details
of the comparison are shown in Table 2.
DISCUSSION
is study illustrates the dierence in academic
outcome between TL and FC method in Epidemiology and
Biostatistics courses among ai 2
nd
year undergraduate
medical students. e main nding from this study
suggests that TL yields better academic achievement
compared to FC in this subject.
It is clear that the FC group had statistically lower
examination score even though all the teaching topics
were the same and the test analysis illustrated similar
test property. Additionally, the ANCOVA with adjusted
covariates presented a consistent result and the power
analysis guaranteed a sucient power. Other studies
aiming at exploring the eect of implementing FC on
the same or related subjects have illustrated the similar
academic outcome between the two groups.
16,17
Hence,
TABLE 1. Characteristic of TL and FC groups.
Characteristic Traditional Lecture group Flipped Classroom group
(N = 315) (N = 313)
Gender: female 153 (48.6) 145 (46.3)
Age (year) 19.63 ± 0.91 19.50 ± 0.58
Hometown located in Bangkok 152 (48.3) 146 (46.6)
High school located in Bangkok 182 (57.8) 188 (60.1)
Grade point average during 1
st
year 3.30 ± 0.39 3.53 ± 0.31
Score of basic statistics course 60.24 ± 12.72 75.46 ± 10.58
*Data is presented in n (%) for categorical data and mean ± SD for continuous data
TABLE 2. Comparison of academic outcome between TL and FC group.
Score comparison TL group
1
FC group
1
P-value 95%CondenceInterval(CI)
Examination score
2
77.97 ± 9.44 69.82 ± 10.83 < 0.001 6.56, 9.74
Examination score with 77.97 ± 9.44 69.82 ± 10.83 < 0.001 11.16, 14.15
adjustment for covariates
3
1
Data is presented in mean ± SD
2
p-value and 95% CI from independent t-test
3
p-value and 95% CI from ANCOVA with gender, age, hometown location, high school location, GPA and score of basic statistic course as
covariates
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
126
our results support the evidence that the integration of
FC into the curriculum did not show better academic
outcome, but worse.
Nevertheless, a number of studies have showed a
signicant increase in students’ satisfaction in the FC
group.
10,13,18
Furthermore, implementation of FC results
in the better development of several practical so skills.
19
us, it is clear that integration of FC should not be aimed
only at better academic outcome of the students, but also
focusing on maturation process of the learning ability of
the students to eventually acquire adult learning which
is the ultimate outcome for learners in this era. us, it
is highly suggested that evaluation of these skills should
be included in the FC curriculum as well.
is study did have some limitations. Firstly, the
active learning policy was implemented in every course
for the 2015 cohort. us, the result of this study could be
aected by the overt workload from other courses as well.
Secondly, other factors that could aect the examination
score such as examination preparation technique or
additional peer tutoring within each cohort were not
explored due to the nature of a retrospective study.
Nevertheless, this study did retrieve all related objective
data, controlled numerous potential confounders and
presented consistent ndings across statistical analyses
which were comparable with other published studies.
CONCLUSION
TL method shows better academic outcome as
compared to FC method in the subject of Epidemiology
and Biostatsitics from the historical records of ai 2
nd
year medical students.
ACKNOWLEDGMENTS
We would like to thank the Education Department
for the provided data, all the instructors and coordinators
of the course and Ms. Saengnapha Tarpick for their
support.
Conict-of-Interest Notication Page: is project
was supported by Medical Education Fund. e authors
declare that they have no other conicting interest. MH
was the main instructor and OB was the coordinator of
both courses, although all data was obtained through
Department of Education of the Institute.
REFERENCES
1. Schumacher DJ, Englander R, Carraccio C. Developing the
master learner: applying learning theory to the learner, the
teacher, and the learning environment. Acad Med 2013;88:
1635-45.
2. Collins J. Education techniques for lifelong learning: principles of
adult learning. Radiographics: a review publication of the
Radiological Society of North America, Inc 2004;24:1483-9.
3. Hurtubise L, Roman B. Competency-based curricular design
to encourage signicant learning. Curr Probl Pediatr Adolesc
Health Care 2014;44:164-9.
4. Inra JA, Pelletier S, Kumar NL, Barnes EL, Shields HM. An
active learning curriculum improves fellows’ knowledge and
faculty teaching skills. Adv Med Educ Pract 2017;8:359-64.
5. Moett J. Twelve tips for “ipping” the classroom. Med Teach
2015;37:331-6.
6. Sharma N, Lau CS, Doherty I, Harbutt D. How we ipped the
medical classroom. Med Teach 2015;37:327-30.
7. Ramnanan CJ, Pound LD. Advances in medical education
and practice: student perceptions of the ipped classroom.
Adv Med Educ Pract 2017;8:63-73.
8. Tainter CR, Wong NL, Cudemus-Deseda GA, Bittner EA.
e “Flipped Classroom” Model for Teaching in the Intensive
Care Unit. J Intensive Care Med 2017;32:187-96.
9. Rui Z, Lian-rui X, Rong-zheng Y, Jing Z, Xue-hong W, Chuan Z.
Friend or Foe? Flipped Classroom for Undergraduate
Electrocardiogram Learning: a Randomized Controlled Study.
BMC Med Educ 2017;17.
10. Betihavas V, Bridgman H, Kornhaber R, Cross M. e evidence
for ‘ipping out’: A systematic review of the ipped classroom
in nursing education. Nurse Educ Today 2016;38:15-21.
11. Moraros J, Islam A, Yu S, Banow R, Schindelka B. Flipping for
success: evaluating the eectiveness of a novel teaching approach
in a graduate level setting. BMC Med Educ 2015;15:27.
12. Mortensen CJ, Nicholson AM. e ipped classroom stimulates
greater learning and is a modern 21
st
century approach to
teaching today’s undergraduates. J Anim Sci 2015;93:3722-31.
13. Chen F, Lui AM, Martinelli SM. A systematic review of the
eectiveness of ipped classrooms in medical education. Med
Educ 2017;51:585-97.
14. Shin YI, Yang CY, Joo MC, Lee SG, Kim JH, Lee MS. Patterns
of using complementary and alternative medicine by stroke
patients at two university hospitals in Korea. Evid Based
Complement Alternat Med 2008;5:231-5.
15. Hu R, Gao H, Ye Y, Ni Z, Jiang N, Jiang X. Eectiveness of
ipped classrooms in Chinese baccalaureate nursing education:
A meta-analysis of randomized controlled trials. Int J Nurs
Stud 2017;79:94-103.
16. Evans KH,ompson AC,O’Brien C,Bryant M,Basaviah
P,Prober C,et al. An innovative blended preclinical curriculum
in clinical epidemiology and biostatistics: Impact on student
satisfaction and performance. Acad Med 2016;91:696-700.
17. Galway LP, Corbett KK, Takaro TK, Tairyan K, Frank E. A
novel integration of online and ipped classroom instructional
models in public health higher education. BMC Med Educ
2014;14.
18. Graham M. Blended learning: A way to engage undergraduate
students in learning epidemiology and biostatistics? Australian
Epidemiologist 2016;23:16-9.
19. Lin Y,Zhu Y,Chen C,Wang W,Chen T,Li T,et al. Facing the
challenges in ophthalmology clerkship teaching: Is ipped
classroom the answer? PLoS One2017;12:e0174829.
Binsomprasong et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
127
Original Article
SMJ
Patcharaporn Phonpornpaiboon, M.D., Kanokrat Suvarnsit, M.D., Sarun Prakairungthong, M.D.
Department of Otorhinolaryngology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.
Re – evaluation of Rinne Test with Aluminum Alloy
Tuning Fork 256 Hz and 512 Hz
Coresponding author: Sarun Prakairungthong
E-mail: tiktono99@gmail.com, tiksarun99@yahoo.com
Received 24 July 2018 Revised 17 January 2019 Accepted 20 January 2019
OCID ID:http://orcid.org/0000-0002-2098-5199
http://dx.doi.org/10.33192/Smj.2019.20
ABSTRACT
Objective: To study the sensitivity, specicity and accuracy of the 256 Hz and 512 Hz aluminum tuning fork in
the detection of conductive hearing loss by quick Rinne test.
Methods: e patients with hearing problems recruited from out-patient unit were tested with the 256 Hz and 512
Hz aluminum quick Rinne test. e audiometry was performed on the same day and the results were compared.
Results: During the study period, 246 ears with conductive hearing loss and 246 ears with non-conductive hearing
loss were recruited. e 256 Hz Rinne test had higher sensitivity than the 512 Hz Rinne test (93.83% and 71.95%
respectively). e specicity and accuracy of the 512 Hz Rinne test was markedly greater than the 256 Hz Rinne
test (91.18 % vs. 26.7% specicity and 83.57% vs. 61.85% accuracy). e 512 Hz Rinne test had sensitivity over 80%
when the air-bone gap was equal to or greater than 20 dB. e sensitivity was even higher (more than 90%) if the
air-bone gap was equal to or greater than 30 dB.
Conclusion: is study demonstrated that the 512 Hz Rinne test had better accuracy than the 256 Hz Rinne test
for the diagnosis of conductive hearing loss. e 512 Hz Rinne test can detect the air-bone gap of 30 dB or greater
which indicates the surgical role with the sensitivity more than 90%. We recommend that 512 Hz Rinne test should
be used as a screening tool for the detection of conductive hearing loss.
Keywords: Rinne; tuning fork; accuracy; conductive hearing loss; air-bone gap (Siriraj Med J 2019;71: 127-130)
INTRODUCTION
e tuning fork test for hearing evaluation has
long been widely used. Various techniques for tuning fork
usability were developed; for instance, Weber test, Bing
occlusion test, Schwaback test and Rinne test. However,
the Rinne test is primarily used as one of the bedside
tools in the evaluation of patients with hearing problem.
Despite the easily accessibility of the audiometry which
is considered as a gold standard for hearing diagnosis,
tuning fork 256 Hz and 512 Hz were still recommended
for general practitioners and otoneurologists for several
circumstances; for example, screening for conductive
hearing loss, pre-and post- operative evaluation of middle
ear surgery
1
and detection of the air-bone gap. Even
though, several researches have tried to evaluate the
value of tuning fork since 1980s
2
, the results had wide
variation. One study
3
suggested that a 512 Hz tuning fork
should be used for screening patients with conductive
hearing loss. However, another study advised to use 256
Hz tuning fork instead.
3
Also, the tuning fork material is
another issue to consider for the various results. Many
studies didn’t mention tuning fork material.
2,3,4
Also, one
study used uncommon material such as metal tuning
fork which is rarely used in ordinary practice.
5
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
128
Our study aimed to re-assess the accuracy, specicity,
and sensitivity of the 256 Hz and 512 Hz aluminum
tuning fork test by performing quick Rinne test as the
primary objective. e secondary objectives were the
following;
1. To determine how to interpret equivocal Rinne test
2. To determine the results of tuning fork test in comparison
with the air-bone gap.
e ndings from this study will guide us for the
way to evaluate of hearing problem patients in out-patient
clinic and primary care unit.
MATERIALS AND METHODS
is cross-sectional study was approved by the Siriraj
Institutional Ethical Committee (Si 587/2015). Subjects
were recruited from the out-patient otolaryngology
clinic during May 2015 to July 2016. Routine history
taking and otological examination were performed in
individual patients. e patients with the age of 18 years
and older suspicious for abnormal hearing were recruited.
6
e exclusion criteria was patients with learning or
psychological problems. e procedure was started with
the quick Rinne test performed on both ears by a group of
well-trained doctors in the tightly closed room (estimate
background noise of 53.2 dBA SPL). e examiners were
blinded from both patients’ diagnosis and otological
examination. e Rinne test was performed using 256
Hz and 512 Hz aluminum tuning fork by the loudness
comparison method (quick Rinne test).
7
e vibrating tuning fork was placed rmly behind
the pinna (BC position) for few seconds then moved to the
front of the ear canal approximately 2 to 2.5 centimeters
from the tragus with parallel position to the skull (AC
position). e examinee would be asked to compare the
loudness of each position. Which site was louder?
2
e
results were recorded in 3 categories: negative Rinne
test when the examinee heard louder when placing
tuning fork on the mastoid bone than when holding in
front of the ear (BC>AC), positive Rinne test when the
examinee heard louder while holding the fork in front
of the ear than placing on the mastoid bone (AC>BC),
and equivocal Rinne test when the examinee could not
tell the dierence between the two positions (AC=BC).
We classied the examinee into 2 groups: conductive
hearing loss group and nonconductive hearing loss
group. e calculated number for each group was 246
ears which were collected in order to avoid the bias. e
profound sensorineural hearing loss ears were excluded
from the study.
RESULTS
ere were 296 cases recruited in this study. Most
of them are women (61.5%). Age ranged between 18 to
87 years with the mean age of 54.4 years (S.D. 15.22).
Demographic data are shown in Table 1.
Data recorded when performing the test with 256
Hz tuning fork was mostly negative Rinne test on both
conductive hearing loss (92.6%) and nonconductive
hearing loss (65.85%) groups. When using 512 Hz tuning
fork, the negative Rinne test was recorded in 71.95% of
conductive hearing loss subjects. Whereas, the positive
Rinne test was reported in 86.5% of nonconductive
hearing loss subjects.
e equivocal result was recorded in similar amount
of 5% in both conductive and nonconductive hearing
loss group either performing with 256 Hz or 512 Hz
tuning fork. Interestingly this equivocal result lead to
nonconductive situation in 256 Hz tuning fork while
this leads to no clue in 512 Hz tuning fork. All results
from all tests in this study are shown in Table 2.
The data was calculated with the exclusion of
equivocal results for sensitivity, specicity, accuracy,
positive predictive value and negative predictive value
of 256 Hz and 512 Hz quick Rinne test. erefore, the
number of the tested ears decreased to the nal of 464
ears in 256Hz tuning fork group and 465 ears in 512Hz
tuning fork group.
As the result, the 256 Hz Rinne tuning fork test has
very high sensitivity (93.83%), low specicity (26.7%),
and moderate accuracy (61.85%). However, all the data
in 512 Hz tuning fork group was higher than 256 Hz
tuning fork group except for the sensitivity. (Fig 1)
TABLE 1. Demographic data.
Data Number (%)
Sex
Male 114 (38.5)
Female 182 (61.5)
Education
Uneducated 2 (0.7)
Grade 1-6 75 (25.3)
Grade 7-12 56 (18.9)
Diploma 24 (8.1)
Undergraduate 139 (47%)
Number of used ear
Unilateral 100 (33.8%)
Bilateral 196 (66.2%)
Phonpornpaiboon et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
129
Original Article
SMJ
TABLE 2. Results of 256 Hz and 512 Hz Rinne tuning fork test.
Conductive hearing loss Nonconductive hearing loss
256 Hz (246 ears) 512 Hz (246 ears) 256 Hz (246 ears) 512 Hz (246 ears)
Number (%) Number (%) Number (%) Number (%)
Rinne negative 228 (92.6) 177 (71.95) 162 (65.85) 19 (7.72)
Rinne positive 15 (6.09) 56 (22.76) 59 (23.98) 213 (86.5)
Rinne equivocal 3 (1.21) 13 (5.28) 25 (10.16) 14 (5.69)
Fig 1. Sensitivity, specicity, positive predictive value, negative predictive value, and accuracy of 256 Hz compare with 512 Hz Rinne tuning
test. (*positive predictive value, **negative predictive value)
In this study authors used the average air-bone gap
according to Committee on Hearing and Equilibrium
guidelines for the evaluation of results of treatment of
conductive hearing loss.
9
From the results, dierent
frequencies tuning forks had dierent power to determine
conductive hearing loss. e 256 Hz tuning fork had
very high power (92.7%) to detect conductive hearing
loss even with only 10 dB air-bone gap. Whilst, 512Hz
tuning fork would have a high power (91.9%) to detect
conductive problem if the air-bone gap is greater than
30 dB air-bone gap. (Fig 2)
DISCUSSION
Despite the low specicity of 256 Hz tuning fork,
the sensitivity of 256 Hz Rinne tuning fork test is greater
than 512 Hz tuning fork test (Fig 1). is yield the similar
result as previous studies of Burkey et al
2
and Chole et al.
3
Burkey et al
2
suggested that the sensitivity of
equivocal Rinne test in conductive hearing loss can be
improved by applying 512 Hz tuning fork.
Chole et al
3
and Browning et al
4
reported that
whether they included equivocal results of tuning fork test
or not, the sensitivity and specicity were not dierent.
We found that the equivocal Rinne test comprised of 5%
of subjects and our data showed that the results did not
correlate with either positive nor negative Rinne test.
3,4
For this reason, we did not include the equivocal Rinne
test in the calculation for sensitivity and specicity.
Rinne’s sensitivity of detecting conductive hearing
loss in each range of air-bone gap is essential for clinical
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
130
Fig 2. Sensitivity of air-bone gap detection of 256 Hz and 512 Hz Rinne tuning fork.
appliance. Crowley et al
8
studied between the Rinne tuning
fork test 256, 512, 1024, and 2048 Hz in 153 ears and
air-bone gap detection. He concluded that the air-bone
gap equal to 20 dB or over was necessary for Rinne test
to distinguish a conductive hearing loss. Burkey et al
1
studied the 512 Hz Rinne test of conductive hearing loss
ears, and he concluded that Rinne’s sensitivity for the
detection of conduction loss equal to 20 dB or greater
was 94.9%. Our study showed that 512 Hz tuning fork
can detect 20 dB conductive problem or greater with the
sensitivity of 81%. Moreover, the sensitivity reached to
92.3% if the conductive problem was 30 dB or greater.
erefore, we can conclude that the 512 Hz tuning fork is
appropriate for pre-operative and post-operative evaluation
in conductive hearing loss patients or unquestionable
otosclerotic patients.
CONCLUSION
is study showed two important results. Firstly,
the study showed that the 512 Hz Rinne test has more
accuracy than 256 Hz Rinne test and can be a workhorse
of screening conductive hearing loss in clinical practice.
Secondly, the 512 Hz tuning fork can detect air-bone gap
of 30 dB or greater which indicates surgical role with
the sensitivity of 92.3%.
REFERENCES
1. Shea PF, Ge X, Shea JJ, Jr. Stapedectomy for far-advanced
otosclerosis. Am J Otol 1999;20:425-9.
2. Burkey JM, Lippy WH, Schuring AG, Rizer FM. Clinical utility
of the 512-Hz Rinne tuning fork test. Am J Otol 1998;19:59-62.
3. Chole RA, Cook GB. e Rinne test for conductive deafness. A
critical reappraisal. Arch Otolaryngol Head Neck Surg 1988;
114:399-403.
4. Browning GG, Swan IR. Sensitivity and specicity of Rinne
tuning fork test. BMJ 1988;297:1381-2.
5. Stankiewicz JA, Mowry HJ. Clinical accuracy of tuning fork
tests. Laryngoscope 1979;89:1956-63.
6. MacKechnie CA, Greenberg JJ, Gerkin RC, McCall AA, Hirsch
BE, Durrant JD, et al. Rinne revisited: steel versus aluminum
tuning forks. Otolaryngol Head Neck Surg 2013;149:907-13
7. Golabek W, Stephens SD. Some tuning fork tests revisited.
Clin Otolaryngol Allied Sci 1979;4:421-30.
8. Crowley H, Kaufman RS. e Rinne tuning fork test. Arch
Otolaryngol 1966;84:406-8
9. Committee on Hearing and Equilibrium guidelines for the
evaluation of results of treatment of conductive hearing loss.
American Academy of Otolaryngology-Head and Neck Surgery
Foundation, Inc. Otolaryngol Head Neck Surg 1995;113:186-7.
Phonpornpaiboon et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
131
Original Article
SMJ
Samut Chongvisal, M.D., FCOT.*, Walop Supavanich, M.D., FCOT.**, Kanthong ongyai, M.D., FCOT., MA.*,
Sarun Prakairungthong, M.D., FCOT.*, Suvajana Atipas, M.D., FCOT.*, Siriporn Limviriyakul, M.D., FCOT., M.Sc.*
*Department of Otorhinolaryngology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, **Department of Otorhinolaryngology,
Banphaeo Hospital, Samutsakorn 74120, ailand.
Hearing and Balance Survey in Thai Elders
Coresponding author: Siriporn Limviriyakul
E-mail: siriporn.lim@mahidol.ac.th
Received 21 September 2017 Revised 25 May 2018 Accepted 30 June 2018
ORCID ID: http://orcid.org/0000-0001-5872-8625
http://dx.doi.org/10.33192/Smj.2019.21
ABSTRACT
Objective: To survey hearing and balance problems of a sampling group of ai elders in the central region of ailand.
Methods: A population study was people aged 60 and over who agreed to join the survey at the 14 elderly
social associations. e questionnaires were used as a self-report for hearing and balance problems. Aer ears
examination by ENT specialists, the participants underwent air-conduction audiometry for hearing screening.
Results: 828 participants 259 (31.3%) males and 569 (68.7%) females, age ranged from 60 to 97 (median=68,
mean ± SD = 69.65 ± 6.89 years), had completed screening program for hearing and balance problems using both
questionnaires and examinations. ere was a statistically signicant correlation between self-report hearing loss and
severe or profound hearing loss in both ears (P <0.05). All age groups reported dizziness and vertigo but the elderly
over 75 years reported more falls with a statistical signicance (P=0.021). e elderly who did not report dizziness
or vertigo had a1.1-1.6 times higher fall histories than those who reported (P <0.001). e participants who had
hypertension and dyslipidemia reported more falls with statistical signicances (P=0.025 and 0.036 respectively).
Conclusion: Our hearing and balance questionnaires may be used as a tool for screening at a primary care for the
elderly > 70 years old especially in those with underlying diseases. Hearing and balance screening will provide early
detection and primary intervention to minimize the negative impact in the elderly’s daily life.
Keywords: Balance; hearing loss; falls; elderly; ear pathology; aging; screening (Siriraj Med J 2019;71: 131-142)
INTRODUCTION
Denition of older or elderly persons varies depending
on denition among countries or organizations. Gorman
M.
1
dened ‘old’ as ‘the aging process is biological reality
which has its own dynamic, largely beyond human control.
However, it is also subject to the constructions by which
each society makes sense of old age.’ In the developed
country, age of retirement is about 60 to 65 years old,
nevertheless, the United Nations (UN) agreed age over
60 years to be the older population.
2
Age of retirement
in ailand depends on organization in which private
employees average about 55 and government ocers are
about 60 years old.
3
ai elderly are dened according
to Elderly Act, Article 3 and the Department of Older
Persons, which was 60 years old and over. In 2015, UN
reported 12% of the global population aged 60 and
over was expected to rise to be a quarter of the global
population by 2050.
4
Asia contains about 56% of the
global elderly population.
5
Help Age international global
network reported elderly population in ailand in
2012 was about 13.7% of population and was expected
to reach 31.8% in 2050.
6
Hearing loss in elderly was called ‘presbycusis’ (or
age-related hearing loss), which literally means ‘hearing in
the older age’.
7
According to WHO classication, hearing
impairment was classied into normal (0-25 dB), mild
(26-40 dB), moderate (41-60 dB), severe (61-80dB) and
profound (>80 dB) hearing loss.
8
Hearing disability in
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
132
ailand was dened as pure tone average at 0.5,1,2 KHz ≥
40 dB on the better ear. Overall prevalence of presbycusis
in the United states, was dened as hearing loss > 25 dB
which was about 23% at age range between 65-75 years
and 40% of people over 75 years old.
9
Otological center,
Bangkok Unit (1991)
10
reported hearing loss 40.4% of
ai elderly at age more than 60 years. Bunnag C. (2002)
11
reported 52% hearing loss in ai elders. Hearing loss
can aect elderly daily life especially communication
and socialization.
Dizziness and vertigo are the common complaints
in the elderly. Falls are common causes of injuries among
elderly and led to many serious consequences in older
people. Elderly over 65 years of age have risk around
one-third of older people living in the community of
falling at least once per year.
12
One in ten of ai elderly
experienced falls accidentally both inside and outside
their home.
5
Falls prevention by early detection and
intervention will decrease serious injuries.
e Otological Center, Bangkok Unit (OCBU) was
the rst International Federation of Otorhino-laryngology
Society and International Society of Audiology (IFOS/ISA)
recognized center for prevention of hearing impairment
and deafness. e mission of OCBU is ‘better hearing
and balance for all’. As mentioned above, elderly is a
major healthcare problems, so screening of hearing
and balance problems in terms of ENT specialist are
primary community preventive intervention for hearing
loss and falls. Aer retirement, the elderly in ailand
usually join the elderly social association nearby their
houses. ese gatherings create proper activities for
their own sake, group and community as well as the
health promotion to both update proper knowledge and
activities such as group exercise at the associations, so we
surveyed hearing and balance problems to nd out their
hearing and balance problems that could be treated and
rehabilitated to prevent disability that might happen.
Objective: to survey hearing and balance problems of
sampling group of elderly in the central region of ailand.
Outcome measurement
Prevalence of ear pathology, hearing and balance
problems in older population in the central region of
ailand.
Setting
Fourteen elderly social associations in Samutsakorn
province as a sampling group for central region of ailand,
which is around 100 kilometers far from Siriraj Hospital.
MATERIALS AND METHODS
e study was approved by the Institutional Review
Board and Ethical Committee of Faculty of Medicine
Siriraj Hospital (Si 463/2014).
Population study
e population study were people age 60 and over
who agreed to join the survey at elderly social associations
in Samutsakorn province. e entire sampling population
was conducted at 14 elderly social associations: Tung
Kralok, Kanchana, Rong Khae, Wat Rang Nin Pradit,
Yok Krabud, Lak Sam, Am Phang, Don Sakae, Lak Song,
Nong Song Hong, Prathumthong Pattana, Khong Tan,
Kaset Pattana, and Jet Rew.
Exclusion criteria
Elderly who did not agree to join survey.
Methods
e representatives of the elderly social society
announced the ‘screening days for hearing and balance”.
e study was conducted during January to December,
2014. All volunteers had to nish questionnaires as a
self-report for hearing and balance problems designed
by OCBU for elderly survey in 2002.
11
ere were 16
questions, which were occupation, underlying diseases,
hearing, ear and balance problems. Aer ear examination
by ENT specialists, they underwent air-conduction
audiometry for hearing screening. e specialists lled
the ear examination and audiological testing forms
(Appendix).
Trained audiometricians performed air conduction
hearing thresholds measuring both ears in a silent room
which ambient noise less than 50 dBA
13
with a portable
audiometer and headphone.
Aer ear, hearing and balance screening, participants
with ear, hearing or balance problems received the referral
forms to access the direct channel to the ENT clinics in
the hospital aliations.
Statistical analysis
We used IBM® SPSS Version 20 to analyze data.
Descriptive statistic was used for characteristics of patients,
pathological ear ndings, hearing and balance problems.
Self-report hearing, degree of hearing loss, hearing disability
and falls were analyzed using Pearson Chi square test.
Continuous data more than 2 groups was analyzed by
Krusal-Wallis test. P < 0.05 was considered as statistical
signicance.
Chongvisal et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
133
Original Article
SMJ
Fig 1. Underlying diseases and hearing disability
RESULTS
ere were 1,560 out of 5,888 members (26.5%) of
14 elderly social associations who participated in service
survey. e hearing and balance screening questions as
well as examination were completed in 828 members.
Demographic information showed 259 males (31.3%)
and 569 females (68.7%); with age range 60 - 97 (median:
68, mean ± SD:69.65 ± 6.89) years old. Past occupation
was divided into not working and working in noisy
area at least 20 years. Elderly who did not work in noisy
areas were 9 (1.08%) as oce workers, 19 (2.29%) as
house workers and 800 (96.9%) as others (e.g. freelances,
businessmen, farmers) and those who worked in noisy
areas were 9 noisy area workers (1%), and 1 soldier
(0.1%). Two hundred and eighty (33.8 %) participants
had underlying chronic diseases, which were 118 (14.3%)
hypertension (HT), 314 (37.9%) HT and other comorbid
diseases (diabetes mellitus; DM, dyslipidemia; DLP,
chronic renal failure; CRF), 19 (2.3%) DM, 72 (8.7%)
DLP, and 5 (0.6%) other diseases.
Questionnaires about ear and hearing problems
demonstrated that elderly had three common complaints
which were ear itching, hearing loss and tinnitus; 449
(54.2%), (278 (33.6%) and 158 (19.1%), respectively. ey
had reported balance problems, which were vertigo 299
(36.1%), dizziness 89 (10.7%) and fall 59 (7.1%), Table 1.
Pathological ndings were 133 (16.1%) earwax
impaction, 19 (2.3%) otitis externa, 18 (2.2%) ear drum
perforation, 12 (1.4%) ear drum bulging, 8 (1%) foreign
body in the ear, 4 (0.5%) auricular malformations (Table 1).
Elderly had 159 (19.2%) unilateral and 442 (53.4%)
bilateral hearing loss. is study classied elderly people
into three age ranges for hearing problems, which were
60-70, 71-80 and >80 years. PTA of better hearing were
(Mean ± SD) 31.44 ±11.6, 36.52 ± 12.91 and 41.83 ±
11.37 dBHL, respectively, Table 2. ere were statistically
signicant dierences in each age group as age increased
(P < 0.001). Hearing disability found 293 (58.5%), 208
(41.5%), and 652 (15.1%) in 60 -70, 71- 80 and > 80
years, respectively. ere were statistical signicances
in PTA and hearing disability in each age group (P <
0.001) (Table 2). Hypertension (N-92, 34.1%) showed a
statistical signicance (P-0.007, OR-1.953 (1.192, 3.197))
in hearing disability group, but not in DM, DLP and
HT with other comorbidities, Fig 1. Self-report hearing
problems showed statistical signicance to severe and
profound hearing loss in both ears (P < 0.05) (Table 3).
Elderly reported balance problems 131 (15.8%), 128
(15.5%), 91 (10.9%) and 74 (8.9%) in age group of 60-
65, 66-70, 71-75 and > 75 years, respectively. ere was
no statistical signicance in each group except falls in
age group of > 75 years (P- 0.021) (Table 4). Fiy-nine
fallers did not have history of vertigo nor dizziness with
statistical signicance (P< 0.001). Fallers had underlying
diseases, which were 5 (8.5%) DM, 6 (27.3%) HT,
26 (61.9%) HT with comorbidity and 34 (57.6%) DLP
(Table 5). ere were statistical signicances in underlying
diseases between fallers and non-fallers in DM and DLP
(P- 0.025 and 0.036).
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
134
TABLE 1. Questionnaires on hearing and balance problems.
Questions Number (%)
(N=828)
Family History of hearing loss 90 (10.9%)
Ever exposure to noise 453 (54.7%)
Duration of exposure (years)
Often got cold 211 (25.4%)
Ear and hearing problems
Ear itching 449 (54.2%)
Self-report Hearing loss 278 (33.6%)
Duration of hearing loss (years) (Mean±SD) 0.36 ± 2.86
Tinnitus 158 (19.1%)
Otalgia 92 (11.1%)
Otorrhea 35 (4.2%)
Ear trauma 20 (2.4%)
Ear surgery 1 (0.1%)
Balance problems
Vertigo 299 (36.1%)
Dizziness 89 (10.7%)
Fall 59 (7.1%)
Ear pathology
Earwax impaction 133 (16.1%)
Otitis externa 19 (2.3%)
Ear drum perforation 18 (2.2%)
Foreign body 8 (1%)
Ear drum bulging 12 (1.4%)
Auricle malformation 4 (0.5%)
DISCUSSION
is study recruited participants from age of 60
years who met the criteria. ere were 828 out of 1560
(53.08%) elderly who participated the hearing and balance
screening. From demographic data women attended
screening test about 2 times than men (female: male-
569:259), which meant that female elderly are more likely
to be member of elderly social association and took care
of their health more than men. Report of ai elderly
showed women had life expectancy longer than men
5,14
(mean age; female-79 and male-72 years), which were
elderly sex ratio of 80 males for every 100 females.
e most common ear and hearing problems from
questionnaires was ear itching (54.2%), which might be
caused by local or systemic diseases. e most common
pathological otologic nding was earwax impaction (133
out of 828; 16.1%) which may associate with hearing
loss, tinnitus, itching, otalgia, discharge or dermatologic
diseases.
15
Our report showed earwax impaction about
16.1%, which was the same as previous reports (19 to
65% of patients over 65 years old).
11,15
is study classied participants into 3 groups
(60-70, 70-80, > 80 years) to clarify hearing problems in
each decade of life. It can be seen that PTA increased as
Chongvisal et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
135
Original Article
SMJ
TABLE 2. Pure tone average and hearing disability in each age group.
Age Right (dBHL) Left (dBHL) Hearing disability
OR
(years) Mean SD Mean SD (Better ear ≥ 40 dB)
60 - 70 31.44 11.6 32.16 12.33 293 1
(58.5%)
71 - 80 36.52 12.91 37.03 14.15 208 2.18
(41.5%) (1.57, 3.03)
P < 0.001*
> 80 41.83 11.37 45.04 16.45 52 6
(15.1%) (2.53, 14.24)
P < 0.001*
P value < 0.001* < 0.001*
Overall 33.2 12.33 34.1 13.71
Min 0 10
Max 85 100
Median 32.25 32.5
Mean (SD) 33.85 34.67
(12.45) (13.74)
Median ambient noise (dBA) 48.8 dB, *statistical signicance P < 0.05, Kruskal-Wallis test, Chi- square test
TABLE 3. Self-report and degree of hearing loss.
Degree of hearing loss Self-report hearing loss Crude OR P-value
(dB) No Yes (95%CI)
Right ear
Normal 153 73 1
(27.8%) (26.3%)
Mild 276 118 0.896 0.542
(26-40) (50.2%) (42.4%) (0.63,1.28)
Moderate 111 71 1.341 0.159
(41-60) (20.2%) (25.5%) (0.89,2.02)
Severe 0 3 2.73 0.002*
(61-80) (0%) (1.1%) (1.14,6.5)
Profound 0 3 0.32 0.013*
(>80) (0%) (1%) (0.27,0.39)
Left ear
Normal 146 66 1
(26.5%) (23.7%)
Mild 270 123 1.00 0.967
(26-40) (49.1%) (44.2%) (0.70,1.45)
Moderate 113 65 1.27 0.262
(41-50) (20.5%) (23.4%) (0.84,1.94)
Severe 20 20 2.21 0.021*
(61-80) (3.6%) (7.2%) (1.12,4.39)
Profound 1 4 8.85 0.021*
(>80) (0.2%) (1.4%) (0.97,80.7)
*Statistical signicance P<0 .05, Pearson Chi-square test, Fisher’s exact for cell <5
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
136
TABLE 4. Age and balance problems.
Age (years) Balance problems
Dizziness Vertigo Falls Total
(424 out of 828)
60 - 65 29 88 14 131
(22.1%) (67.2%) (10.7%) (15.8%)
OR-1 OR-1 OR-1
66-70 17 91 20 128
(13.3%) (71.1%) (15.6%) (15.5%)
OR-1.856 OR-0.832 OR-1.044
(0.963, 3.578) (0.496,1.411) (0.993, 1,097)
P - 0.18 P – 0.496 P – 0.083
71-75 25 60 6 91
(27.5%) (65.9%) (6.6%) (10.9%)
OR - 0.751 OR – 1.057 OR-1.359
(0.405,1.393) (0.611, 1.863) (0.512, 3.610)
P - 0.362 P – 0.487 P- 0.573
> 75 18 37 19 74
(24.3%) (50%) (25.7%) (8.9%)
OR – 1.248 OR – 1.262 R – 0.438
(0.647, 2.408) (0.729, 2.184) (0.213, 0.898)
P – 0.509 P – 0.405 P-0.021*
*Statistical signicance P <0.05, Pearson Chi-square test
age increased with statistical signicances in this study.
Elderly in age group of 71-80 and > 80 years had hearing
disabilities about 2 (OR - 2.18, P < 0.001) and 6 (OR 6,
P <0.001) times more or less than those in age group
of 60 -70 years, Table 2. e most common underlying
disease in this study was HT, which was related to hearing
disability about 1.9 times (OR- 1.192, 3.197) more or less
than others. Common severity of hearing loss in study
population was mild (~40-50%) and moderate (~25%)
by the audiometric evaluation, Table 3. Nevertheless,
old people reported their hearing loss when the severity
of hearing loss became severe or profound more or less
than those who had no hearing loss, Table 3 (Crude OR
~2-8, P<0.05 both ears). Elderly in ailand seem likely
to refuse hearing aids when they had hearing problems.
ey believed it was the stigmata of the handicapped and
dependent elderly, so their guardians would bring them
when they had diculties in communication with them.
Although ai government have oered free hearing
aids to whom register for the handicapped certication,
some get it, but do not use it.
Balance disorders: dizziness and vertigo; were found
to be common complaints in all age groups, but more
common in elderly.
16
More than one third of adults over
65 years old had reported falls secondary to dizziness
and vertigo each year according to the Center of Disease
Control and Prevention.
17
ailand has become an aging
society, so ai government has planned to extend the
retired age from 60 to 65 years old. is study tried to
clarify the age-related balance problems which would
disturb their daily activities and work life. From the
reason above, this study classied 5 years for each age
group. Although there were no statistical dierences
in report of dizziness and vertigo in all age groups, the
elderly age group of > 75 years old had reported falls
more than other age groups with a statistical signicance
(P- 0.021). An international consensus statement dened
a fall as “an unexpected event in which the participant
Chongvisal et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
137
Original Article
SMJ
TABLE 5. History of vertigo or dizziness correlated with falls.
History Number (%) Crude OR P-value
Fallers Non-fallers (95% CI)
Vertigo
Yes 0 299
(0%) (77.1%)
No 59 89 1.663 <0.001*
(38.9%) (22.9%) (1.458,1.896)
Dizziness
Yes 0 89
(0%) (22.9%)
No 59 299 1.197 <0.001*
(100%) (77.1%) (1.143,1.254)
Underlying diseases
Diabetes 5 158 0.358 0.025*
(8.5%) (20.5%) (0.141, 0.910)
Dyslipidemia 34 335 1.762 0.036*
(57.6%) (43.6%) (1.031,0.010)
Hypertension 6 113 0.876 0.788
(27.3%) (30%) (0.334,2.297)
Hypertension with 26 288 1.49 0.223
co-morbidity (DM, DLP, CRF) (61.9%) (52.2%) (0.782,2.838)
Abbreviations: DM = diabetes mellitus, DLP = dyslipidemia, CRF = chronic renal failure
*Statistical signicance P <0.05, Pearson Chi-square test
comes to rest on the ground, oor or lower level”.
18
ose
who reported dizziness and vertigo in this study were
not the fallers comparing to those who did not report
with a statistical signicance (OR – 1.2-1.6, P < 0.001).
is meant participants who did not have symptoms of
dizziness nor vertigo, might not concern about falling
problems. Prior studies found a marked reduction of hair
cell counts within the cristae of semicircular canals as
early as 50-60 years of age and a signicant reduction of
vestibular ganglion cells as early as 60 years of age.
17
us,
the deterioration of peripheral vestibular functions and
central compensation can no longer be compaensated
at some critical points.
19
e rate of falls in the well-
dened general older population is about 10.4-18.7%
in ailand.
20
Prasansuk S, et al., and Downton J.,
16,21
reported percentage of falling in the elderly each year
about 28-35%, 35% and 32-42% in 65+, 70+ and 75+
years. Our study elderly reported falls 10.7%, 15.6%, 6.6%
and 25.7% at age of 60-65, 66-70, 71-75 and >75 years,
respectively. e percentage of our study was about half
of the previous studies, because of the less number of
study population. Nevertheless, the peak incidence of
falls in our study was over 75 years (25.7%), which was
nearly the same as previous reports.
21
Dizziness, vertigo and falls may be secondary
to comorbid conditions other than aging. Comorbid
conditions (Intrinsic factors) that we considered in
this study were underlying diseases, which showed
statistical signicances in DM and DLP between fallers
and non-fallers. Nicholson M, et al., and Walley M, et
al.,
22,23
showed the decits of vestibulo-ocular reex and
proprioceptive loss in diabetic patients which caused the
high percentage of fall in these patients. Dyslipidemia
could lead to atherosclerosis. Prior study reported the
small vascular diseases related to brain changes impacted
on gait and speed in healthy elderly.
16,24
e extrinsic
factors was environmental area of their dwellings that
was not a part of our study. Many interventions have
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
138
been proposed to prevent falls which are exercise which
was found to be a benet of performance of head and
neck exercise as Prasansuk S, et al., reported
16
, medication
(both supplement and treatment), surgery for correction
(e.g. cataract extraction), uid/nutrition supplement,
psychological support, environment/assistive technology
and knowledge/ educational interventions.
18
However,
there are many factors that cause balance problems
individually. Old ai people usually spend their retired
life at home with their families, who take care of them.
us, family may be the one who notice their problems
as well as providing environmental area for primary
prevention of falls.
CONCLUSION
Hearing and balance problems unusually cause
negative impact eects in elderly daily’s life. Hearing
rehabilitations will resolve their communications, social
and psychological problems. Prevention of falls include
individual and environmental adjustments, and decreasing
the complications of chronic diseases in older people.
Our hearing and balance questionnaires may be used
as a tool for screening at a primary care for an elderly
> 70 year especially in those with underlying diseases.
e primary care and elderly social associations should
be alerted by giving them education about hearing and
balance problems. is study supports the information
to the health policy decision for elderly in central region,
and all of ailand which is growing to become an aging
society in the near future.
ACKNOWLEDGMENTS
is study was conducted as part of strengthening
networking of the Otological Center: Bangkok Unit. We
thanks all of our residents who graduated in, 2014.
Conicts of interest: e authors have no conicts of
interest and did not receive any funding from proted
funders.
REFERENCES
1. Gorman M. e ageing and development report: poverty,
independence and the world’s older people. In: Randel J, editor.
Earthscan Publication Ltd.; 1999. p. 3-21.
2. World Health Organization (WHO). Denition of an older or
elderly person 2016. Available from: http://www.who.int/
healthinfo/survey/ageingdefnolder/en/.
3. Kanchanachitra M, Jarassit S, Kanachanachitra C. To Retire or
not to Retire, Whose choice Is It? The 8
th
International
Conference of Population and Social Research Royal River
Hotel: Mahidol University; 2012. p. 341-57.
4. United Nations. Department of Economic and Social Aairs,
Population Division. World Population Prospects: e 2015
Revision, Volume II: Demographic Proles (ST/ESA/SER.
A/380)2015.
5. Prasartkul P. Situation of the ai Elderly 2015. Nakhon pathom:
Institute for Population and Social Research, Mahidol University;
2015.
6. Knodel, John, Chayavan N. The Changing Well-being of
ai Elderly: An Update from the 2011 Survey of Older Person
in ailand. HelpAge International;2013.
7. Roth TN. Aging of the auditory system. Handb Clin Neurol.
2015;129:357-73.
8. World Health Organization (WHO). Grading of hearing
impairment 2017. Available from: http://www.who.int/pbd/
deafness/hearing_impairment_grades/en/.
9. Seidman MD, Ahmad N, Bai U. Molecular mechanisms of
age-related hearing loss. Ageing Res Rev 2002;1:331-43.
10. Prasansuk S. Report on prevalence of hearing disability and ear
diseases in ailand Bangkok Unit: Otological Center, Department
of Otorhinolaryngology, Mahidol University; 1991.
11. Bunnag C, Prasansuk S, Nakorn AN, Jareoncharsri P, Atipas S,
Angsuwarangsee T, et al. Ear diseases and hearing in the ai
elderly population. part II. A one year follow-up study. J Med
Assoc ai 2002;85:532-9.
12. Gillespie LD, Robertson MC, Gillespie WJ, Sherrington C,
Gates S, Clemson LM, et al. Interventions for preventing falls
in older people living in the community. Cochrane Database
Syst Rev 2012:Cd007146.
13. Frank T. ANSI update: maximum permissible ambient noise
levels for audiometric test rooms. Am J Audiol 2000;9:3-8.
14. WHO. THAILAND http://www.who.int/countries/tha/en/
[25 May 2018].
15. Roland PS, Smith TL, Schwartz SR, Rosenfeld RM, Ballachanda
B, Earll JM, et al. Clinical practice guideline: cerumen impaction.
Otolaryngol Head Neck Surg 2008;139:S1-S21.
16. Prasansuk S, Siriyananda C, Nakorn AN, Atipas S, Chongvisal
S. Balance disorders in the elderly and the benet of balance
exercise. J Med Assoc ai 2004;87:1225-33.
17. Zalewski CK. Aging of the Human Vestibular System. Semin
Hear 2015;36:175-96.
18. Hopewell S, Adedire O, Copsey BJ, Sherrington C, Clemson LM,
Close JCT, et al. Multifactorial and multiple component interventions
for preventing falls in older people living in the community.
Cochrane Database Syst Rev 2016:CD012221.
19. Baloh RW, Jacobson KM, Socotch TM. e eect of aging on
visual-vestibuloocular responses. Exp Brain Res 1993;95:509-16.
20. Romli MH, Tan MP, Mackenzie L, Lovarini M, Suttanon P,
Clemson L. Falls amongst older people in Southeast Asia: a
scoping review. Public Health 2017;145:96-112.
21. Downton J. Falls. In: Luxon L, editor. Textbook od Audiological
Medicine; Clincal aspect of hearing and balance. United Kingdom:
Martin Dunitz; 2003. p. 831-40.
22. Nicholson M, King J, Smith PF, Darlington CL. Vestibulo-ocular,
optokinetic and postural function in diabetes mellitus. Neuroreport
2002;13:153-7.
23. Walley M, Anderson E, Pippen MW, Maitland G. Dizziness and
Loss of Balance in Individuals With Diabetes: Relative Contribution
of Vestibular Versus Somatosensory Dysfunction. ClinDiabetes
2014;32:76-7.
24. Pinter D, Ritchie SJ, Doubal F, Gattringer T, Morris Z, Bastin
ME, et al. Impact of small vessel disease in the brain on gait
and balance. Sci Rep2017;7:41637.
Chongvisal et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
139
Original Article
SMJ
Appendix
Hearing and Balance Questionnaires
Otological Center: Bangkok Unit, Faculty of Medicine Siriraj Hospital, Mahidol University
Record NO……………………….. Sex………………
Date……………………………….
__________________________________________________________________________________________
1. Occupations: work at least 20 years
Oce worker…….years Noisy area worker e.g. Factory ……years
Policeman……..yeas Solider……..years
House worker……..years Others………………………
2. Do you have hearing loss?
2.1 Right No Yes………years
2.2 Le No Yes………. years
2.3 Hearing loss in family history
No Yes
3. Have you ever expose to loud noise?
No Yes
4. Noise in ears (tinnitus) Right No Yes Le No Yes
5. Otalgia Right No Yes Le No Yes
6. Ear discharge Right No Yes Le No Yes
7. Previous ear surgery Right No Yes Le No Yes
8. Ear itching Right No Yes Le No Yes
9. Ear trauma e.g. severe head trauma, beating or slapping at ears
Right No Yes Le No Yes
10. Ototoxic drugs No Yes …………………….
11. Oen get a cold No Yes
12. Have you ever had spinning sensation? (Vertigo)
No Yes
13. Have you ever had dizziness?
No Yes
14. Have you ever fallen? (Faller/Nonfaller)
No Yes
15. How oen you feel dizziness or spinning sensation?
No Sometime O and on Usually
16. Do you have these following diseases?
1. Diabetes No Yes
2. High blood pressure (Hypertension No Yes
3. High blood lipid prole (Dyslipidemia) No Yes
4. Anemia No Yes
5. Kidney diseases No Yes
__________________________________________________________________________________________
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
140
Ear Examination form
Record no………………. Date……………………..
Examiner……………………
____________________________________________________________________________________________
Otoscopic examination
Right ear Le ear
Auricle Malformation No Yes No Yes
Otitis externa No Yes No Yes
Ear Wax No Yes No Yes
Foreign body No Yes No Yes
Otorrhea No Yes No Yes
Fungi No Yes No Yes
Eardrum perforation No Yes No Yes
Red and Bulging No Yes No Yes
Retraction>moderate No Yes No Yes
Eusion No Yes No Yes
Other: specify…………………………
Normal Findings
Chongvisal et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
141
Original Article
SMJ
Audiological testing
Record no………………. Date…………………….
Audiometric no____________ Ambient____________dB _______
__________________________________________________________________________________________
Air conduction threshold
Right ear AC/(BC) Le ear
_________ 500 Hz ________
_________ 1000 Hz ________
_________ 2000 Hz ________
_________ 400 0Hz ________
Tester _________________
Tympanometry
Right ear Le ear
Compliance
Negative middle ear pressure
Flat
Stapedial reex Positive
Remark __________________________________________________
Treatment:
1. No treatment
2. Action needed
2.1 Medication
2.2 Suction
2.3 Hearing aids
2.4 Removal earwax/foreign body/______
2.5 Surgery Referral Urgent Non-Urgent
2.6 Other c Specify____________________________________________________________________
Follow up:_________________________________________________________________________________
Refer to:______________________________________________________________________________________
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
142
Referred form
Otological Center: Bangkok Unit, Faculty of Medicine Siriraj Hospital, Mahidol University
And Ear Disease Foundation (ailand)
Date ________________
Name of patient _______________________________________________
Result Right Le
1. External ear _______ _______
2. Middle ear _______ _______
3. Ear drum _______ _______
4. Hearing Hz
_______ 500 _______
_______ 1000 _______
_______ 2000 _______
_______ 4000 _______
Diagnosis_______________________________
Treatment provided
1.______________________________________
2.______________________________________
Referred for
_______________________________________________________________________
Doctor__________________
Chongvisal et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
143
Original Article
SMJ
Apichaya Claimon, M.D. *, Napaporn Tojinda, M.Sc. *, Ninmanee Taweewatanasopon, B.Sc. *, Pornphit Boonkhon,
B.Sc.*, Boontham Amornkitticharoen, M.Sc. *, Sirilak Wiriyaakradecha, M.Sc.**, Shanigarn iravit, M.D.*
*Department of Radiology, Faculty of Medicine Siriraj Hospital, **Department of Clinical Microscopy, Faculty of Medical Technology, Mahidol
University, Bangkok 10700, ailand.
Scintigraphic Detection of Splenosis and
Accessory Spleen by Using Tc-99m Labelled
Denatured Red Blood Cells
Corresponding author: Shanigarn iravit
E-mail: thiravit.mahidol@gmail.com
Received 19 October 2017 Revised 6 February 2018 Accepted 9 February 2018
ORCID ID: http://orcid.org/ 0000-0002-4347-8683
http://dx.doi.org/10.33192/Smj.2019.22
ABSTRACT
Objective: To evaluate the usefulness of Tc-99m-labelled denatured red blood cell scintigraphy (DRBCS) in the
detection of splenosis and accessory spleen.
Methods: is retrospective study reviewed 7 patients who were suspected of splenosis and accessory spleen during
2007-2014. Six patients had undergone total splenectomy before the DRBCS study. Multi-planar, SPECT and
SPECT/CT images of DRBCS were reviewed by visual analysis. Image ndings were correlated with other imaging
modalities and clinical presentations at follow-up.
Results: DRBCSs were positive in 6 patients; 5 splenosis and 1 accessory spleen. A single lesion was detected in
each patient. Locations were in splenic bed (5 patients) and in pelvic cavity (1 patient). Mean lesion size was 3.7 +
2.4 cm. SPECT or SPECT/CT imaging could eliminate false-negative results in 2 patients. Lesion uptake intensity
in 4 patients was higher than in the liver. Two patients had subsequently undergone resection of the lesions and
pathological examinations conrmed the diagnosis of splenosis and accessory spleen. Sensitivity of DRBCS was
comparable with other imaging modalities. However, DRBCS was more specic for the splenic tissue.
Conclusion: DRBCS is useful for the detection of splenosis and the accessory spleen. Detection sensitivity of
DRBCS is better than ultrasonography(US). DRBCS is more specic to the splenic tissue, compared with US and
CT. Furthermore, SPECT/CT imaging increases sensitivity, lesion localization and characterization. Clinical impact
of DRBCS is apparent including pre-surgical localization, elimination of additional follow-up imaging and invasive
procedure.
Keywords: Accessory spleen; denatured red blood cell; immune thrombocytopenia; splenosis; spleen scintigraphy
(Siriraj Med J 2019;71: 143-149)
INTRODUCTION
e spleen is the largest lymphoid organ which plays
an important role in regulating immune cells and mediators,
removing circulating microorganisms through phagocytosis
and ltering damaged cells and foreign material.
1
Splenosis
is the acquired condition from autotransplantation of
splenic tissue that usually occurs aer splenic trauma
or splenectomy. In contrast, the accessory spleen is a
congenital anomaly of the spleen.
2
ese conditions can
present a diagnostic challenge and can be particularly
dicult to distinguish by using conventional imaging
modalities such as ultrasonography (US), computed
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
144
tomography (CT) and magnetic resonance imaging (MRI).
Improved characterization may be obtained by using
superparamagnetic iron oxide as a contrast-enhancing
agent in MRI.
3
In contrast, spleen-specic imaging using
Tc-99m labeled denatured red blood cells scintigraphy
(DRBCS) is a more simple technique that can denitely
distinguish splenic tissue from other benign/malignant
non-splenic origin according to intrinsic physiologic
uptake of DRBC by functioning splenic tissue.
DRBCS was introduced as a spleen-specic tracer
since 1967 and has been shown to be of value in the
detection of both native spleen and accessory spleen.
4
Clinical indications for DRBCS include congenital/
functional asplenia, accessory spleen, space-occupying
lesion in the spleen, refractory immune thrombocytopenia
(ITP) aer splenectomy, and thoracic/ abdominopelvic
mass suspected for the accessory/ectopic spleen.
The objective of this study was to evaluate the
usefulness of DRBCS in the detection of splenosis and
accessory spleen by 1) reporting the DRBCS ndings
and compare with the other imaging modalities and 2)
investigating the clinical impact of DRBCS in the patients
with suspected splenosis and accessory spleen.
MATERIALS AND METHODS
This retrospective study was approved by The
Institutional Review Board (Si 085/2015). We searched
our internal clinical database for the time period between
2007 and 2014 and identied 8 patients who had undergone
DRBCS from our Nuclear Medicine Unit. e inclusion
criteria were the patients who had undergone DRBCS
for the evaluation of splenosis or accessory spleen. We
excluded 1 patient who had undergone DRBCS for the
evaluation of the space-occupying lesion in the native
spleen. Our nal study cohort comprised of 7 patients.
DRBCS protocol
Preparation of Tc-99m DRBC
1. Reconstitute stannous kit with normal saline and
intravenously inject to a patient.
2. Aer 20-30 min, draw blood of 10 ml into a heparinized
syringe.
3. Using a lead glass shielding, add 20 mCi of Na
99m
TcO
4
to the heparinized whole blood.
4. Aer 5 min, incubate the tube in a constant-temperature
water bath at 50 ± 1
o
C for 20 minutes with periodical
shaking.
5. Centrifuge, wash and reconstitute with NSS.
6. Check the percentage of DRBC and re-inject to the
patient. Denaturation of RBC was technically successful in
all patients. In every patient, no complications occurred
aer the re-injection. e preparation was tested in one
of our research team as a healthy volunteer and the
images shown a normal distribution (Fig 1).
DRBCS imaging protocol
Scintigraphy began 30 minutes aer the injection
of DRBC, using a gamma camera equipped with a high-
resolution, low-energy collimator. Static planar images
of the abdomen were obtained in anterior, posterior,
both anterior and posterior oblique (LAO, LPO, RAO,
RPO), and both lateral views in all patients with the same
imaging acquisition protocol. Aer planar imaging, single
photon emission computed tomography (SPECT) and
SPECT/CT imaging was obtained over the abdomen
in the selected patients. One patient (patient No.6) was
obtained planar imaging only. Two and 4 patients were
performed SPECT and SPECT/CT imaging sequentially
aer the planar scans.
Image analysis
DRBCS images were retrospectively analyzed on a
printed lm or electronic archiving system by a nuclear
medicine physician who was unaware of other clinical
data. e following ndings were determined by visual
analysis: (1) the presence or absence of abnormal tracer
uptake suggesting splenosis or accessory spleen; (2) the
intensity of uptake (compared with the liver uptake);
and (3) the location of the abnormal uptake lesion(s).
Lesion size was measured on the SPECT/CT images. In
patients without SPECT/CT imaging, lesion size was
evaluated by other imaging modalities. Other imaging
data (the US in 4 patients and contrast-enhanced CT in
1 patient) were reviewed by a radiologist. e DRBCS
Fig 1. DRBCS of a normal volunteer. e uptake in spleen is
homogeneously intense. Liver uptake is faint. No marrow uptake is
observed.
Abbreviations: Ant: Anterior; Post: Posterior; LAO: Le anterior
oblique; Lt LAT: Le lateral; LPO: Le posterior oblique: RAO: Right
anterior oblique: Rt LAT: Right lateral; RPO: Right posterior oblique.
Claimon et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
145
Original Article
SMJ
ndings were compared with other imaging modalities
and pathological ndings as a gold standard.
Statistical analysis
e analyses were performed by using statistical
soware (SPSS version 18.0; SPSS Inc., Chicago, IL,
USA). All continuous (quantitative) data, including the
patient’s age and lesion size, were summarized.
RESULTS
Four patients were male and 3 were female. e
median age was 31 years (range 5 – 70 years). ree
patients were children; age ranged from 5-8 years. Six
patients had undergone total splenectomy prior to the
DRBCS; 1 of these had undergone 2 sessions of splenectomy
according to refractory ITP (patient No. 4). None of the
study population had a history of abdominal trauma.
e clinical indications for performing DRBCS were
to localize residual splenic tissue causing refractory ITP
post-splenectomy (5 patients), to dierentiate between
splenosis and metastatic cancer (1 patient) and to evaluate
pre-operatively for the biopsy conrmed accessory spleen
(1 patient). Time intervals between splenectomy and
DRBCS were within 1 year in 3 patients, 3 years in 1
patient and 33 years in 2 patients.
Of the 5 patients with refractory ITP, 1 patient had
secondary ITP due to HCV infection (patient No. 3), 1
patient had acute ITP with a subdural hematoma and
had undergone emergency splenectomy (patient No. 1)
and the rest were chronic ITP patients. All 5 patients
were classied as ‘no response’ to treatment based on the
criteria of the International Working Group (a platelet
count <30 x 10
9
/L or less than 2-fold increase in platelet
count from baseline or the presence of bleeding).
5
Patient characteristics, clinical information, image
acquisitions/ ndings, and pathological results, are listed
in Table 1.
DRBCS image quality was excellent in 5 scans.
A patient (No. 3) showed an abnormal homogeneous
increased tracer uptake in the liver and bone marrow.
Specically, the uptake in the residual spleen was less
than in the liver. In this case, planar imaging gave a false
negative result, but SPECT/CT imaging could demonstrate
the residual spleen (Fig 2). Another patient (No. 2) had
uptake in splenosis less than in the liver, but no marrow
uptake.
e scan was negative in a patient with refractory
ITP aer an emergency splenectomy for acute ITP with
subdural hematoma (Fig 3). e rest 6 scans were positive
for residual spleen or accessory spleen. A single positive
lesion was detected in all patients. Median lesion size was
2.8 cm (range 1.4-7.4 cm). Location of the all residual
spleens was in the le upper quadrant of the abdomen
(Fig 4). e accessory spleen was in the pelvic cavity
along with the lobulated-shaped native spleen in the
splenic bed. In this patient, the accessory spleen was
subsequently removed and pathologically conrmed
splenic tissue (Fig 5).
SPECT and SPECT/CT imaging were subsequently
obtained aer the routine multi-planar views in 2 and
4 patients, respectively. SPECT/CT could identify the
residual spleen in a patient with negative multi-planar
scan (patient No. 3). is patient had crescent-shaped
residual spleen in the location that almost attached to the
le liver lobe, which was misinterpreted as the elongated
le liver lobes in the planar images. However, the lesion
was clearly demonstrated in the 3-D imaging by SPECT/
CT scanning.
Two out of 7 patients had undergone resection
of the residual/accessory spleen following the positive
DRBCS (patient No. 4 and 7). e pathological results
conrmed splenic tissue in both cases. However, there
was no rising of the platelet level aer the surgery in
patient No. 4 and she needed long-term treatment with
corticosteroid. Patient No. 7 required resection of the
accessory pelvic spleen due to compression eect to the
bladder. His urinary incontinence subsided aer the
surgery. DRBCS prevented futile investigation/ surgery
in patient No. 5, who was conrmed to have a residual
spleen, not a focus of metastatic colon cancer.
Removal of the residual spleen was not undertaken
in 3 patients with refractory ITP despite the positive
DRBCS results (patient No. 2, 3 and 6). One of them
(patient No. 6) had a spontaneous rising of the platelets to
the normal range without any specic treatment. Patient
No. 3 responded to corticosteroid and Eltrombopag.
Nonetheless, patient No. 2 as well as patient No. 1 (whose
DRBCS result was negative), had no response to the
medical treatment.
Comparison of DRBCS and other imaging modalities
results was summarized in Table 2. e detection sensitivity
of DRBCS is better than US. DRBCS is more specic to
spleen tissue, compared with US and CT.
DISCUSSION
Splenosis is the acquired benign condition. It is
estimated to occur in 26%-67% of the patients aer
traumatic or iatrogenic splenic parenchymal rupture.
2
e
mechanism is the autotransplantation of the spillage cells
to the nearby body cavity or hematogenously spreading
of the cells and implanting to the remote sites.
6
e most
common location of splenosis is in the le upper quadrant
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
146
TABLE 1. Baseline patient characteristics and main results.
No. Sex/ Age Splenectomized Indication Planar SPECT or Intensity
2
Lesion Lesion size Location Pathology
(years) SPECT/CT
1
number (cm)
3
1 M/31 + Refractory ITP - - 0 0 NA NA NA
2 F/5 + Refractory ITP + + 1 1 2 Splenic bed NA
3 F/54 + Refractory ITP - + 1 1 4.7 Splenic bed NA
4 F/44 + RefractoryITP + + 3 1 1.4 Splenicbed Conrmed
5 F/70 + DDx. Metastasis vs. + + 3 1 2.8 Splenic bed NA
splenosis
6 M/8 + Refractory ITP + NA 3 1 NA Splenic bed NA
7 M/7 - Pelvicmass + + 3 1 7.4 Pelvic Conrmed
Abbreviations: M: Male; F: Female; ITP: Immune thrombocytopenia; DDx: Dierential diagnosis; SPECT: Single photon emission computed tomography; CT: Computed tomography; NA: Not available.
1
SPECT only in patient No. 1 and 2; SPECT/CT in patient no. 3-5 and 7
2
Intensity of uptake, compared to the liver: 0 = no uptake, 1 = less than the liver, 2 = equal to the liver, 3 = greater than the liver
3
Maximal diameter; measured on SPECT/CT images. Patient No. 2 was measured on ultrasonography (performed SPECT without CT)
Claimon et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
147
Original Article
SMJ
Fig 2. Abnormal distribution and false-negative planar scan. A 54-year old female with HCV-related refractory ITP status post splenectomy
33 years ago (Patient No. 3). DRBCS planar images (a) show abnormally intense uptake in the liver and diused marrow uptake. Planar
images are negative. DRBCS SPECT/CT (b) images reveal a faint uptake corresponding with a curvilinear so tissue locates beneath the
posterior aspect of the le hemidiaphragm, measured about 1.5 x 4.7 x 1.6 cm. that suggestive of splenosis.
Abbreviation: LPO: Le posterior oblique.
Fig 3. Negative study. A 31-year old man with acute ITP with subdural
hematoma status post emergency splenectomy (Patient No. 1).
DRBCS was sent due to no platelet response aer splenectomy.
DRBCS images show no evidence of residual splenic tissue. He has
been treated with high-dose prednisolone, but no platelet response.
Abbreviation: LPO: Le posterior oblique.
Fig 4. Splenosis. A 44-year old woman with refractory ITP, status post 2 sessions of splenectomy
(33 and 38 years ago), with persistently low platelets (Patient No. 4). DRBCS (a Planar and b
SPECT/CT images) shows a focal increased uptake in a 1.4 cm. so tissue mass near the upper
pole of the le kidney that suggestive of splenosis. e lesion was resected and pathological
result conrmed splenic tissue.
Fig 5. Accessory spleen in the pelvic cavity. A 7-year old boy presented with urinary incontinence (Patient No. 7). e CT shows the small
spleen (a) and a lobulated homogeneous enhancing so tissue mass in the pelvic cavity (b). e dierential diagnosis includes lymphoma,
germ cell tumor and other mesenchymal tumors. e pelvic mass was biopsied and revealed splenic tissue. DRBCS (c) was sent for pre-
operative evaluation. ere is increased uptake in the pelvic mass, size about 7.4 x 2.6 cm. (arrow) as well as in the lobulated shaped spleen
in the le upper quadrant (dot arrow). He had undergone removal of the accessory spleen and the pathological result conrmed the diagnosis.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
148
TABLE 2. Comparison of DRBCS and other imaging modalities.
No. DRBCS Other imaging modalities
4 Positive US Negative
5 Positive US Negative
7 Positive US Positive with concordant result
8 Positive US Positive with inconclusive nature of the lesion
6 Positive CECT Positive with inconclusive nature of the lesion
Abbreviations: US: ultrasonography; CECT: contrast enhanced computed tomography.
of the abdomen. Usually, it is asymptomatic, incidentally
detected and requires no treatment, because it may have
some immunologic value and microorganism ltering
function.
3
However, in patients with ITP, splenosis leads
to failure of the platelet response aer splenectomy, owing
to the presence of functioning splenic tissue. Detection
and removal of the residual functioning splenic tissue
are essential for the treatment in this patient group.
DRBCS is the recommended imaging modality for the
detection of splenosis.
6
e uptake mechanism is spleen-
specic, based on the sequestration of the DRBC by the
pulp region of splenic tissue
7
. e sensitivity of DRBCS
is higher than that of US, CT
6,8
, Tc-99m labeled sulfur
colloid scintigraphy
6,9,8
, and laparoscopy.
10
Moreover,
DRBCS is non-invasive, technically simple and able to
obtain whole-body imaging within a single injection.
Despite the superiority of DRBCS, it is not frequently
reported and studied. In our institution, only 8 patients
and 1 healthy volunteer performed DRBCS since we
established this test in 2007.
DRBCSs are able to detect both splenosis and accessory
spleen. In the literature, splenosis is usually observed
in multiple
11
, but we detected a solitary lesion in every
case. Our ndings conrmed that the most common
location of splenosis is in the le upper quadrant. We
found a patient with pelvic accessory spleen, which is
a very rare condition. In literature, accessory spleens
were found in splenic hilum in 75% of cases, and less
in the gastrosplenic ligament, splenorenal ligament and
pancreatic tail. e development of an accessory spleen
presumably involves multifocal embryonic origins. In
most reports, spleens detected in pelvic regions are from
a dierent origin, the wandering spleens, which are
the abnormal position of the spleen caused by splenic
ligament laxity.
12
The sensitivity of DRBCS was confirmed to be
higher than other modalities such as US, CT, and Tc-99m
sulfur colloid scan, especially for the small lesion and in
the unusual location.
9,8,13,4
e dierential diagnosis for
intraabdominal so tissue nodule seen on CT images
including peritoneal metastasis, lymphadenopathy,
lymphoma, pancreatic mass, endometriosis, uterine
mass, teratoma and splenosis.
6,14
DRBCS was more specic
to splenic tissue than CT. In our patients, DRBCS ruled
out the metastatic nodule and teratoma, so futile biopsy
or surgery could be avoided.
In the current study, DRBCSs were positive in 6
(85.7%) patients. Of these, 1 was negative on the planar
scan but was clearly positive on the SPECT/CT scan. e
planar scan was negative due to the superimposition of
splenic tissue by the liver tail. e problem was eliminated
by SPECT/CT which provided attenuation correction,
three-dimensional view, and higher contrast resolution.
Moreover, lesion characterization and anatomical localization
using CT technique are important for the patients who
planned for pre-surgery. e superior sensitivity of
SPECT or SPECT/CT imaging was in concordance with
the previous study.
15
However, another study showed
no contribution of SPECT to the planar imaging.
16
In this study, only 1 out of 4 patients with refractory
ITP and positive DRBCS results had undergone removal
of the splenic tissue. However, that patient still suered
from persistently low platelets aer the surgery. On the
other hand, another patient had spontaneously increased
platelet level to the normal range without any surgical or
medical treatment. Because DRBCS is spleen-specic, the
false-positive result is not possible and have never been
reported in the literatures.
16
Post-surgical failure could be
described by several factors such as rapid regeneration
of the residual splenic tissue, the presence of IgG against
platelet and other sites of platelet sequestration.
17
ere were limitations to this study. Firstly, this was
retrospective which could have many uncontrolled factors.
Secondly, it contained a small sample size. irdly, we
Claimon et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
149
Original Article
SMJ
used dierent gamma cameras with dierent scanning
parameters. And lastly, pathological conrmation was
not done in every case.
CONCLUSION
DRBCS is useful for the detection of splenosis
and the accessory spleen. e detection sensitivity of
DRBCS is better than US. DRBCS is more specic to
spleen tissue, compared with US and CT. Furthermore,
SPECT/CT imaging increases sensitivity, localization
and lesion characterization. Clinical impact of DRBCS is
apparent including pre-surgical localization, elimination
of additional follow-up imaging and invasive procedure.
ACKNOWLEDGMENTS
e research authors received a Chalermprakiat
grant from the Faculty of Medicine, Siriraj Hospital,
Mahidol University.
Conict of Interest: e research authors (Apichaya
Claimon and Shanigarn iravit) received a Chalermprakiat
grant from the Faculty of Medicine, Siriraj Hospital,
Mahidol University. All other authors (Napaporn Tojinda,
Ninmanee Taweewatanasopon, Pornphit Boonkhon,
Boontham Amornkitticharoen, and Sirilak Wiriyaakradecha)
declare that they have no conict of interest.
Ethical Approval: All procedures performed in studies
involving human participants were in accordance with
the ethical standards of the institutional and/or national
research committee and with the 1964 Helsinki declaration
and its later amendments or comparable ethical standards.
REFERENCES
1. de Porto AP, Lammers AJ, Bennink RJ, ten Berge IJ, Speelman
P, Hoekstra JB. Assessment of splenic function. Eur J Clin
Microbiol Infect Dis2010;29:1465-73.
2. Wu C, Zhang B, Chen L, Zhang B, Chen X. Solitary perihepatic
splenosis mimicking liver lesion: a case report and literature
review. Medicine 2015;94:e586.
3. Fan B, Shepherd JA, Levine MA, Steinberg D, Wacker W,
Barden HS et al. National Health and Nutrition Examination
Survey whole-body dual-energy X-ray absorptiometry reference
data for GE Lunar systems. J Clin Densitom2014;17:344-77.
4. Ehrlich CP, Papanicolaou N, Treves S, Hurwitz RA, Richards
P. Splenic scintigraphy using Tc-99m-labeled heat-denatured
red blood cells in pediatric patients: concise communication.
J Nucl Med1982;23:209-13.
5. Neunert C, Lim W, Crowther M, Cohen A, Solberg L, Jr.,
Crowther MA. e American Society of Hematology 2011
evidence-based practice guideline for immune thrombocytopenia.
Blood 2011;117:4190-207.
6. Lake ST, Johnson PT, Kawamoto S, Hruban RH, Fishman EK. CT
of splenosis: patterns and pitfalls. AJR Am J Roentgenol2012;199:
W686-93.
7. Owunwanne A, Halkar R, Al-Rasheed A, Abubacker KC, Abdel-
Dayem H. Radionuclide imaging of the spleen with heat denatured
technetium-99m RBC when the splenic reticuloendothelial
system seems impaired. J Nucl Med1988;29:320-3.
8. Hagan I, Hopkins R, Lyburn I. Superior demonstration of splenosis
by heat-denatured Tc-99m red blood cell scintigraphy compared
with Tc-99m sulfur colloid scintigraphy. Clin Nucl Med2006;31:
463-6.
9. Massey MD, Stevens JS. Residual spleen found on denatured red
blood cell scan following negative colloid scans. J Nucl Med
1991;32:2286-7.
10. Gigot JF, Jamar F, Ferrant A, van Beers BE, Lengele B, Pauwels
S, et al. Inadequate detection of accessory spleens and splenosis
with laparoscopic splenectomy. A shortcoming of the laparoscopic
approach in hematologic diseases. Surg Endosc1998;12:101-6.
11. Muñoz-Bellvis L, Quiñones-Sampedro JE, Iglesias M, Blanco
O, Fernandez LMG, Velasco CE et al. Contribution of the single
photon emission computed tomography with 99mTc red blood
cells in splenosis. Intern J Res Med Sci 2017;4:2464-7.
12. Varga I, Babala J, Kachlik D. Anatomic variations of the
spleen: current state of terminology, classification, and
embryological background. Surg Radiol Anat.2018;40:21-29.
13. Hagman TF, Winer-Muram HT, Meyer CA, Jennings SG.
Intrathoracic splenosis: superiority of technetium Tc 99m
heat-damaged RBC imaging. Chest 2001;120:2097-8.
14. Mortelé KJ, Mortelé B, Silverman SG. CT Features of the
Accessory Spleen. AJR Am J Roentgenol2004;183:1653-7.
15. Horger M, Eschmann S, Lengerke C, Claussen C, Pfannenberg C,
Bares R. Improved detection of splenosis in patients with
haematological disorders: the role of combined transmission-
emission tomography. Eur J Nucl Med Mol Imaging2003;30:
316-9.
16. Ekmekci S, Diz-Kucukkaya R, Turkmen C, Adalet I. Selective
Spleen Scintigraphy in the Evaluation of Accessory Spleen/
Splenosis in Splenectomized/Nonsplenectomized Patients
and the Contribution of SPECT Imaging. MolImagingRadionucl
er2015;24:1-7.
17. Supe A, Parikh M, Prabhu R, Kantharia C, Farah J. Post-
splenectomy response in adult patients with immune
thrombocytopenic purpura. Asian J Transfus Sci2009;3:6-9.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
150
Kusuma Chinaroonchai, M.D., Pornprom Muangman, M.D., Jatuporn Sirikun, M.D.
Department of Surgery, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.
Does Acute Kidney Injury Condition Affect Revised
BAUX Score in Predicting Mortality in Major Burn
Patients?
Corresponding author: Jatuporn Sirikun
E-mail: j_sirikun@hotmail.com
Received 21 August 2018 Revised 13 November 2018 Accepted 4 February 2019
ORCID ID: http://orcid.org/0000-0001-5514-6355
http://dx.doi.org/10.33192/Smj.2019.23
ABSTRACT
Objective: is study aimed to assess the accuracy of the revised BAUX score for predicting mortality among the
major burn patients with acute kidney injury (AKI) compared with non-AKI group. e epidemiologic information
and risk factors of AKI in major burn patients were also the point of interest.
Methods: is study was a retrospective cohort study. e medical records of 144 major burn patients admitted at
the burns unit of Siriraj Hospital from 2010-2016 were reviewed and important data were retrieved.
Results: Age, hypertension, diabetes mellitus, severity of the burn injuries, and inhalation injuries were the factors
related to AKI in major burn patients. e mortality rate due to AKI in burn patients was high (44.4%). e accuracy
of the revised BAUX score in predicting the mortality among the major burn patients from our series was only
fair (66.7%).
Conclusion: AKI aected on mortality of the major burn patients. Until the better predictor comes up, the revised
BAUX score should be considered as a predictor of mortality in these patients.
Keywords: Acute kidney injury; acute renal failure; major burns; mortality rate; revised BAUX score (Siriraj Med J
2019;71: 150-157)
INTRODUCTION
In 2016, of the 447 burn patients treated at Siriraj
Hospital, 62 were admitted to the burns unit. e length
of stay per case averaged 23 days, costing about US$11,000
per admission. In addition, many hundreds of thousands
of ais in the wider population receive burns every year.
is data suggests that millions of dollars are being spent
by the government every year to pay for medical treatment
related to burns. Better burn treatment can save money,
but the most important objective for improving care is
to enhance the quality of life of burn victims, which is
priceless.
A systemic response aer a burn injury occurs aer
at least 15% of the total body surface area is involved.
1,2
Proinammatory cytokines are secreted and stimulate
the body into a hypermetabolic phase in preparation
for stress. Copious intravascular fluid leakage, the
signature phenomenal response of endothelial cells to
the proinammatory cytokines, is one of the crucial
phases that can result in edema formation in non-burned
tissues. is can progress to burn shock if a burn patient
receives inadequate uid resuscitation treatment. An
acute kidney injury in a burn victim can happen from
many pathways, such as a poor resuscitation process
or a hyper-response of the body to the kidney itself due
to the secretion of pro-inammatory cytokines. is
condition can result in an increased mortality rate for
burn victims, especially among major burn patients.
Acute kidney injury was defined by the Acute
Dialysis Quality Initiative Group, which was founded
Chinaroonchai et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
151
Original Article
SMJ
by nephrologists and intensivists. It was classied into 3
stages: risk (R), injury (I), and failure (F). e consequences
of this condition are divided into 2 groups: sustained loss
of kidney function (L), and end-stage kidney disease (E).
e 3 stages and 2 groups combine to form the acronym,
RIFLE. e RIFLE criteria
3
are used with critically-ill
patients to form an early diagnosis of acute kidney injury
conditions in order to decrease the mortality rate of this
group of patients.
4,5
However, the criteria are limited
by the availability of the baseline creatinine value of a
patient for use as part of the basic data in the criteria.
Consequently, the Acute Kidney Injury Network criteria
(AKIN criteria) were developed by the Acute Kidney
Injury Network (AKIN) to provide a more accurate tool
than the RIFLE criteria for the diagnosis of acute kidney
injuries. e AKIN criteria classies acute kidney injuries
into 3 stages, based on the method of treatment.
6,7
It has been reported that the mortality rate of
critically-ill patients with acute kidney injuries varies
from 10%–100%, depending on the type of illness.
8
Another study reported a 34.9% (mean: 41.9%) mortality
rate for burn patients with acute kidney injuries, which
is 5–6 times higher than the mortality rate of non-AKI
burn patients.
9
Even when a sub-group of AKI patients
in that study received renal replacement therapy (RRT)
for the treatment of renal failure, their mortality rate was
still very high at 80%.
9
e revised BAUX score has been reported as the
most accurate score for predicting mortality due to burns.
is score’s factors are age, percentage of total body
surface area (%TBSA) burn, and inhalation injuries.
10
A
revised BAUX score of ≥ 100 is equivalent to a prediction
of a 100% mortality rate.
Revised BAUX score = Age + %TBSA + 17x
(x as inhalation injury: 1 = yes, 0 = no)
To make it more convenient to calculate the score,
it was developed into nomogram form.
11
Nevertheless,
an acute kidney injury condition is still not one of the
factors used to calculate the score.
MATERIALS AND METHODS
is studied was conducted with the approval of the
Ethics Committee of Siriraj Hospital (Si 690/2016). e
subjects were ≥ 20% total body surface area burn injury
patients who had been admitted to the burns unit of Siriraj
Hospital from 2010–2016. ose patients aged < 18 years
or with a history of diagnosed chronic kidney disease
were excluded, as were patients who had incomplete
medical records. Of the 190 patients screened for the
study, 46 were excluded due to having incomplete data,
leaving 144 burn patients to be analyzed. e RIFLE and
AKIN criterias were used to dene the AKI group and to
classify the severity of their burns. e qualitative data
comprised of sex, underlying diseases, type of burn injury,
with-inhalation injury, and any episode of hypotension
in the rst 24 hours post-injury (hypotension was dened
as a systolic blood pressure of < 90 mmHg). ese were
presented as mean ± SD or median (IQR: percentile 25,
percentile 75) and frequency (%). e continuous data
comprised of age, length of hospital stay (days), %TBSA
burn, length of ventilator use (days), and the volume of
intravenous uids for resuscitation in the rst 24 hours
post-injury and the second 24 hours post-injury; these
were presented as frequency (%).
e program Statistical Package for Social Science
(SPSS) version 18.0 was used to analyze the data. e
independent t-test was used to analyze continuous data,
while the Mann–Whitney test, Chi-square test and Fisher’s
exact test were used for the categorical data. A p-value
less than 0.05 was considered to be statistically signicant.
Factors associated with acute kidney injury were
presented as odds ratio (95% condence interval). e
p-value corresponds to the logistic regression analysis.
e association between the revised BAUX score and
the AKI-with-mortality rate were analyzed by logistic
regression analysis and presented as odds ratio (OR)
with a 95% condence interval (95% CI). e accuracy
of the revised BAUX score in predicting mortality in the
AKI group of burn patients was calculated from AUC.
RESULTS
e demographic data are in Table 1. e majority
of the burn patients were male (75.7%), and the mean
age of all burn victims was 44.09 ± 18.58 years. Essential
hypertension was the most common underlying disease,
followed by diabetes mellitus. e average size of the
burn injuries was 41.75%TBSA, and most injuries had
been caused by ame burn. e AKI group was older
than the non-AKI group (49.97 ± 20.2 vs. 39.52 ± 15.88
years, respectively, p = 0.001). e average injury of the
AKI group was more severe than that of the non-AKI
group (%TBSA 55% vs. 35%, respectively, p < 0.001,
with an incidence of > 60%TBSA 25 vs. 9, respectively,
p < 0.001). Inhalation injuries were mostly found in
the AKI group (27 vs. 19, respectively, p = 0.019). Most
patients in the AKI group required ventilator support
(80.9% vs. 30.9%, respectively, p < 0.001), and the length
of ventilator use was longer for the AKI than the non-
AKI group (19 vs. 7 days, respectively, p = 0.001). e
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
152
TABLE 1. Demographic data.
Factors
AKI group Non-AKI group Total
P-value
(n=63) (n=81) (n=144)
Mean age (years) 49.97 ± 20.2 39.52 ± 15.88 44.09 ± 18.58 0.001*
Sex
(Male:female) 48(76.2%):15(23.8%) 61(75.3%):20(24.7%) 109(75.7%):35(24.3%)
Underlying diseases
– HT 15(23.8%) 6(7.4%) 21(14.6%) 0.008*
– DM 6(9.5%) 1(1.2%) 7(4.9%) 0.043*
– DLP 5(7.9%) 2(2.5%) 7(4.9%) 0.24
– Coronary artery disease 2(3.2%) 1(1.2%) 3(2.1%) 0.581
– Gout 0(0%) 1(1.2%) 1(0.7%) 1
– Cirrhosis 1(1.6%) 0(0%) 1(0.7%) 0.438
– Asthma 0(0%) 3(3.7%) 3(2.1%) 0.257
– COPD 1(1.6%) 0(0%) 1(0.7%) 0.438
– Other eg. CVA, 10(15.9%) 11(13.6%) 21(14.6%) 0.813
Parkinson disease, etc.
Type of burn injury
– Flame burn 51(81%) 53(65.4%) 104(72.2%) 0.042*
– Scald burn 6(9.5%) 10(12.3%) 16(11.1%) 0.79
– Electrical burn 5(7.9%) 18(22.2%) 23(16%) 0.023*
– Other eg. contact hot objects. 1(1.6%) 0(0%) 1(0.7%) 0.438
%TBSA (IQR) 55(38.5,73) 35(25,46) 41.75(30,60) <0.001*
Depth burn wound
– 2
nd
degree (IQR) 49(30,63) 35(25,50) 38(28,58) 0.015*
– 3
rd
degree (IQR) 36.5(15,43.5) 15(10,37.5) 28(10,40) 0.225
Inhalation injury 27(42.9%) 19(23.5%) 46(31.9%) 0.019*
Length of hospital 36(17,60) 31(20,54) 34.5(18.5,56.5) 0.622
stay, day (LOS) (IQR)
Ventilator used 51(80.9%) 25(30.9%) 76(52.8%) <0.001*
Total ventilator days 19 7 13 <0.001*
(IQR) (10,38.5) (5,11) (7,26)
1
st
24 h uid (mL) 12,967 9,656 11,679
(IQR) (9,800,19,000) (6,630,13,982) (7,180,17,236)
0.001*
1
st
24 h uid (mL/kg/%TBSA) 4.05 4.21 4.07
0.413
(IQR) (2.93,5.29) (3.07,5.45) (2.98,5.33)
2
nd
24 h uid (mL) 8,640 5,950 7,378.5
<0.001*
(IQR) (7,100,10,513) (4,800,7860) (5,360,8,985)
2
nd
24 h uid (mL/kg/%TBSA) 2.44 2.48 2.47
0.417
(IQR) (1.73,3.08) (1.79,3.29) (1.75,3.2)
Hx of Hypotension in 1
st
24 h 16(25.4%) 4(4.9%) 20(13.9%) 0.001*
(SBP < 90 mmHg)
* P-value < 0.05 Abbreviations: HT = essential hypertension; DM = diabetes mellitus; DLP = dyslipidemia; COPD = chronic obstructive
pulmonary disease; CVA = cerebrovascular disease; IQR = interquartile range (25
th
percentile, 75
th
percentile); h = hours; %TBSA = % total
body surface area burn; SBP = systolic blood pressure.
Chinaroonchai et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
153
Original Article
SMJ
AKI group required more uid resuscitation in the rst
and second 24 hours post-injury, and also developed
hypotension more oen than the non-AKI group in the
rst 24 hours post-injury.
e APACHE II scores calculated for both groups
are in Table 2. e APACHE II score for the AKI group
was signicantly higher than that for the non-AKI group
(16.86 ± 8.16 vs. 6.49 ± 4.39, p < 0.001). Parameters for
the two groups (such as GCS, MAP, RR, FiO
2
, arterial
pH, Hct, serum HCO3 and serum creatinine) diered
signicantly.
A logistic regression was performed to calculate
the odds ratio of the factors (Table 3). e odds ratio
of patients needing to use a ventilator was very high at
11.23. e DM and HT odds ratios were 8.42 and 3.91,
respectively. e odds ratio for a history of hypotension
in the rst 24 hours post injury was 6.55. Extensive
injury > 60%TBSA showed an odds ratio of 5.26, while
inhalation injury had an odds ratio of 2.45.
63 patients were diagnosed with AKI (Table 4).
e highest incidence of AKI (29) occurred with RIFLE-
Risk and AKIN stage 1; all of the patients in those two
categories were successfully treated using conservative
treatment with medication. A total of 13 other patients
received renal replacement therapy; 12 of those patients
were RIFLE-Failure, while the 13th was RIFLE-Loss; and
all 13 were in AKIN stage 3.
e incidences of mortality at 28 days and at 6
months were signicantly higher for the AKI group,
with p <0.001 (Table 5). Revised BAUX scores were
calculated for each group. e score for the AKI group
(115.59 ± 27.73) was higher than that for the non-AKI
group (82.56 ± 24.41), with p < 0.001 (a revised BAUX
score ≥ 100 predicts 100% mortality rate
10,12
). e revised
BAUX scores were also signicantly higher for the dead
patients in both groups (Table 6).
Both the revised BAUX scores and AKI in Table 7
are also related to the mortality rate from multivariate
analysis, with OR 1.04, p < 0.001 and OR 8.69, p = 0.002,
respectively.
DISCUSSION
In the past, AKI in burns was considered to be
caused by hypovolemia resulting from inadequate
volume resuscitation in the rst 24 hours, post injury.
13,14
Historically, Parkland’s formula was used to determine
TABLE 2. APACHE II scores.
Factors
AKI group Non-AKI group Total
P-value
(n=63) (n=81) (n=144)
APACHE II score 16.86 ± 8.16 6.49 ± 4.39 11.03 ± 8.14 <0.001*
– GCS 10.87 ± 2.72 13.89 ± 2.09 12.57 ± 2.81 <0.001*
– BT (*C) 37.92 ± 0.84 37.89 ± 0.64 37.9 ± 0.74 0.833
– MAP (mmHg) 86.35 ± 11.39 90.8 ± 10.5 88.85 ± 11.08 0.016*
– HR (beat/min) 131.43 ± 126.17 107.64 ± 16.28 118.05 ± 84.79 0.095
– RR (per min) 21.84 ± 4.86 20.17 ± 2.08 20.9 ± 3.66 0.013*
– FiO
2
0.5 ± 0.2 0.42 ± 0.09 0.45 ± 0.15 0.004*
– PaO
2
(mmHg) 134.07 ± 80.28 116.43 ± 52.36 124.15 ± 66.37 0.134
– Arterial pH 7.32 ± 0.11 7.38 ± 0.05 7.34 ± 0.1 0.048*
– Serum HCO
3
(mmol/L) 19.87 ± 5.01 23.9 ± 3.75 22.14 ± 4.78 <0.001*
– Na (mmol/L) 134.54 ± 25.03 137.85 ± 4.14 136.4 ± 16.85 0.303
– K (mmol/L) 4.21 ± 0.76 4.09 ± 0.59 4.14 ± 0.67 0.28
– Serum Cr (mg/dL) 2.01 ± 1.58 0.77 ± 0.25 1.31 ± 1.23 <0.001*
– Hct (%) 35.93 ± 11.19 40.71 ± 8.6 38.62 ± 10.06 0.006*
– WBC (cells/mm
3
) (IQR) 14,110 10,920 11,645
0.232
(5,960, 21,860) (6,800, 16,010) (6,560, 18,025)
Data is presented as mean ± SD, *p-value < 0.05. Abbreviations: APACHE II = Acute Physiology and Chronic Health Evaluation version
II; GCS = Glasgow Coma Score; BT = body temperature; MAP = mean arterial pressure; HR = heart rate, RR = respiratory rate, IQR =
interquartile range (25
th
percentile, 75
th
percentile)
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
154
TABLE 3. Associated factors for AKI.
Factor
Univariate Crude OR
P-value Factor
Univariate Crude OR
P-value
(95%CI) (95%CI)
Age 1.03 (1.01, 1.05) 0.001* APACHE II score 1.29 (1.18, 1.41) <0.001*
HT 3.91 (1.31, 13.05) 0.008* GCS 0.64 (0.55, 0.74) <0.001*
DM 8.42 (0.97, 392.01) 0.043* BT 1.05 (0.67, 1.65) 0.832
Electrical Injury 0.3 (0.08, 0.92) 0.023* MAP 0.96 (0.93, 0.99) 0.019*
Total %TBSA injury 1.04 (1.02, 1.06) <0.001* HR 1.03 (1.01, 1.06) 0.004*
– <20%TBSA 1 (0, 1) 0.438 RR 1.15 (1.03, 1.28) 0.011*
– 20%–40%TBSA 0.24 (0.11, 0.5) <0.001* FiO
2
56.26 0.006*
(3.18, 995.95)
– 41%–60%TBSA 1.33 (0.6, 2.93) 0.462 PaO
2
1 (1, 1.01) 0.123
– >60%TBSA 5.26 (2.09, 14) <0.001* Arterial pH 0 (0, 1.07) 0.052
Type of burn injury Serum HCO
3
0.81 (0.74, 0.88) <0.001*
– 2
nd
degree 1.02 (1.01, 1.04) 0.01* Na 0.99 (0.96, 1.01) 0.294
– 3
rd
degree 1.03 (0.99, 1.08) 0.134 K 1.32 (0.8, 2.17) 0.28
Inhalation Injury 2.45 (1.13, 5.35) 0.019* Cr 195.62 <0.001*
(26, 1471.65)
Ventilator used 11.23 (4.71, 27.65) <0.001* Hct 0.95 (0.92, 0.99) 0.005*
Event of hypotension in 6.55 (1.93, 28.2) 0.001* WBC 1 (1, 1) 0.083
1
st
24 h (SBP<90 mmHg)
*P-value < 0.05 Abbreviations: HT = essential hypertension; DM = diabetes mellitus; %TBSA = % total body surface area burn; h = hours;
SBP = systolic blood pressure; GCS = Glasgow Coma Score; BT = body temperature; MAP = mean arterial pressure; HR = heart rate;
RR = respiratory rate; Cr = serum creatinine; Hct = hematocrit; WBC = white blood cell
TABLE 4. Type of AKI (total AKI group = 63 patients).
RIFLE Incidence AKIN Incidence Dead in 28 days Dead in 6 months
Risk 29 (46%) Stage 1 29 (46%) 3 (16.7%) 6 (21.4%)
Injury 21 (33.3%) Stage 2 21 (33.3%) 8 (44.4%) 12 (42.9%)
Failure 12 (19%) Stage 3 13 (20.6%) 7 (38.9%) 10 (35.7%)
Loss 1 (1.59%)
Total 18 28
ESKD 0
Abbreviations: RIFLE = Risk-Injury-Failure-Loss of Kidney Function-End-stage Kidney Disease; AKIN =Acute Kidney Injury Network;
ESKD = end stage kidney disease
Chinaroonchai et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
155
Original Article
SMJ
Total AKI group Non-AKI group P-value
(n=144) (n=63)
97.01 ± 30.61 115.59 ± 27.73 82.56 ± 24.41 <0.001*
Dead Survive P-value Dead Survive P-value
(n=28) (n=35) (n=3) (n=78)
Revised BAUX score 127.23 ± 27.72 106.27 ± 24.31 0.002* 125.33 ± 44.06 80.91 ± 22.23 0.036*
the uid resuscitation volume required daily. However,
this standard formula was oen found to calculate more
uid than patients needed in practice. e current trend
is to use the same formula to estimate the uid volume,
but then to adjust the uid rate every 2 hours, based on
the patient’s clinical response. is technique better
balances the uid needs of the patient, thereby ensuring
that there is adequate uid resuscitation. e trend to
keep a dryer status has been proven to decrease the
incidence of acute complications such as compartment
syndrome, both in the extremities and abdomen.
15,16
Late complications such as infection, which is the
biggest killer for late death burns, have been reported
to relate to the hyper-resuscitation volume that had
been given.
17
Even in the case of patients who received
adequate uid resuscitation, AKI can still happen from
other pathways. For example, the direct-injury eect of
pro-inammatory cytokines acts on the stress response
pathway, which stimulates the Renin-angiotensin-system
(RAS); the RAS hormones then enhance the kidney’s
vascular tone, which boosts its glomerular ltration rate.
Moreover, sometimes cytokines themselves can create a
direct injury to the kidneys and the myocardial muscle;
this worsens the perfusion to the kidney, which results
in additional and indirect kidney damage.
13
e stress response to burn injuries in the elderly
is not the same as that for younger people.
18
Due to the
elderly’s poor body reserves resulting from the aging
process, they are prone to develop more complications
aer burn injuries, such as AKI, as found in the study
by Wu et al.,
19
and our study. Essential hypertension
and diabetes mellitus are diseases that aect systemic
vascular quality. e poor quality vessels aect end-organ
perfusion of the heart and kidney. Low renal perfusion,
TABLE 5. Mortality rates for both groups.
Factors
AKI group Non-AKI group Total
P-value
(n=63) (n=81) (n=144)
Mortality at 28 days 18 (28.6%) 2 (2.5%) 20 (13.9%) <0.001*
post injury
Mortality at 6 months 28 (44.4%) 3 (3.7%) 31 (21.5%) <0.001*
post injury
*P-value < 0.05
TABLE 6. e revised BAUX scores for each group.
TABLE 7. e relationship of the revised BAUX scores and AKI with the mortality rate.
Data presented as mean ± SD, *p-value < 0.05
Univariate analysis Multivariate analysis
Crude OR P-value Adjusted OR P-value
(95%CI) (95%CI)
Revised BAUX 1.06 (1.03, 1.08) <0.001* 1.04 (1.02, 1.06) <0.001*
AKI 20.8 (5.93, 73.01) <0.001* 8.69 (2.23, 33.84) 0.002*
*P-value < 0.05, CI = condence interval
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
156
especially in a stressed condition, creates further damage
to kidneys, which results in the acute kidney injury
condition.
Most patients in this series received their injuries
from ame burns, thus sharing a common etiology with
burn victims in other countries all around the world.
However, in the present study, the AKI condition was
found to be more common for the electrical-injury
group than had been reported in other studies. is may
be explained by the fact that this kind of injury causes
more severe damage to the tissue than a ame injury,
and it can develop the rhabdomyolysis condition, which
burdens the kidneys and makes them prone to further
injury. Pro-inammatory cytokines play a major role
in the systemic response of the body aer major burn
injuries. AKI occurs more frequently when patients
have larger areas of burn injury. e severity of the
body’s systemic response to a burn injury is related to
the inammatory cytokine level, which varies with how
large the injured area is.
20
Second-degree burns play a
signicant role in the development of AKI complications.
is may be explained by the level of cytokines that are
stimulated and secreted. Most of these are triggered by
second-degree burns of the skin
21
, which create more
hemodynamic changes from uid shis than from third
degree wounds, in which all cells are dead.
22,23
In this study, the presence of inhalation injuries,
ventilator use, episodes of hypotension in the rst 24
hours, and arterial pH were related to AKI, which was
also found in the study by Schneider et al.
24
e AKI
group had a longer average hospital stay and more
complications, consistent with other studies.
9,13,14,19,23
Moreover, the mortality in the AKI group in our study
was 44.4%, which is in the range of 4%–90% from other
studies.
13,14,23
e average uid resuscitation volume in the AKI
group was larger than that for the non-AKI group,
although aer calculating the volume per kg per %TBSA,
the dierence proved to be non-signicant. Currently,
there is no valid, xed formula available for calculating
resuscitation uids for all types of burn patients. e best
thing that we can do is to closely monitor and titrate the
uid rate based on a patient’s hemodynamic status to
reach the optimum and adequate resuscitation endpoint.
During the rst 24 hours, the resuscitation-uid rate
for both groups was about 4 mL/kg/%TBSA, calculated
using Parkland’s formula; the need for uid decreased
by a half during the second 24-hour period.
e data in Table 2 shows that APACHE II is a
good scoring system for evaluating the severity of a burn
injury with a coexistent AKI condition. e AKI group
showed signicantly higher APACHE II scores than
the non-AKI group (p <0.001). As for the parameters
of APACHE II, it was found that GCS, MAP, RR, FiO2,
arterial pH, serum HCO
3
, Hct and serum creatinine
diered signicantly between the AKI and non-AKI
groups.
Most of the AKI patients were RIFLE-R and
AKIN stage I, which can be treated conservatively with
medication. A total of 13 patients in RIFLE-F, RIFLE-L
and AKIN stage 3 received renal replacement therapy
(RRT) treatment. Even though eective RRT treatment
was initiated in these groups, the mortality rates were
still high (38% for RIFLE-I and AKIN stage 2, 58.3%
for RIFLE-L, and 53.8% for AKIN stage 3). Detection at
an earlier stage may help to decrease patient morbidity
from the necessary intervention treatment (RRT) and
may lessen mortality.
From a multivariate analysis of the revised BAUX
score and AKI with the mortality rate, we found that
both factors relate with mortality rate with a statistically
signicant adjusted OR. is means that both factors
heavily aect the mortality rate. Consequently, the AKI
factor should be considered in order to improve the
accuracy of the calculation formula used to predict burn
mortality rates.
Using the revised BAUX score cut-o at 100 for
prediction of mortality in the AKI group (Osler T et al.
dened a score ≥ 100 as a prediction of a 100% mortality
rate
10
) as shown in Fig 1, the sensitivity is 83.9%, with
61.9% specicity. e positive predictive value is 37.7%,
and the negative predictive value is 93.3%. e accuracy
is 66.7%.
At 66.7%, the accuracy of the revised BAUX score
to predict the mortality rate in our study was quite low
compared with other studies.
24,25
e accuracy of the
calculated score may have been impacted by the fact
that nearly half of our study population (43.7%) were
in the AKI group.
is study had added another supporting evidence
that AKI is one of the important factors in predicting the
outcome of the burn patients. e total number of the
patients and the fact that nearly half of the population
has developed AKI might be the main factor which
using BAUX score in predicting mortality rate was not
as accurate as shown the previous studies.
AKI is one of the signicant conditions that increase
the mortality rate of major burn patients. e future
calculation scoring system should recognize it as an
important factor aecting mortality. e revised BAUX
score is not accurate enough to use to predict mortality
of major burn patients with AKI.
Chinaroonchai et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
157
Original Article
SMJ
ACKNOWLEDGMENTS
is research project was supported by the Faculty
of Medicine Siriraj Hospital, Mahidol University (Grant
Number: [IO] R016031018). We thank Mr.Tanut
Sornmanapong for facilitating the research coordinative
process.
Conicts of interest: It is hereby declared that there are
no conicts of interest.
REFERENCES
1. Kao CC, Gamer WL. Acute burns. Plast Reconstr Surg 2000;
105:2482-92.
2. Kramer GC. Pathophysiology of burn shock and burn edema.
In: Herndon DN. Total burn care. 4
th
ed. Edinburgh; New
York: Saunders Elsevier; 2012.p.103-13.
3. Bellow R, Ronco C, Kellum JA, Mehta RL, Palevsky P, Acute
Dialysis Quality Initiative workgroup. Acute renal failure –
denition, outcome measures, animal models, International
Consensus Conference of the Acute Dialysis Quality Initiative
(ADQI) Group. Crit Care 2004;8:R204-12.
4. Abosaif NY, Tolba YA, Heap M, Russell J, El Nahas AM. e
outcome of acute renal failure in the intensive care unit according
to RIFLE: Medel application, sensitivity, and predictability.
Am J Kidney Dis 2005;46:1038-48.
5. Uchino S, Bellow R, Goldsmith D, Bates S, Ronco C. An
assessment of the RIFLE criteria for acute renal failure in
hospitalized patients. Brit Care Med 2006;34:1913-17.
6. Molitoris BA, Levin A, Warnock DG, Joannidis M, Mehta RL,
Kellum JA, et al. Acute Kidney Injury Network Working Group.
Improving outcomes of acute kidney injury: report of an
initiative. Nat Clin Pract Nephrol 2007;3:439-42.
7. Mehta RL, Vellum Ja, Shah SV, Molitoris BA, Ronco C, Warnock
DG, et al. Acute Kidney Injury Network. Acute Kidney Injury
Network: report of an initiative to improve outcomes in acute
kidney injury. Crit Care 2007;11:R31.
8. Hoste EA, Schurgers M. Epidermiology of acute kidney injury:
How big is the problem? Crit Care Med.2008;36:S146-51.
9. Brusselaers N, Monstrey S, Colpaert K, Decruyenaere J, Blot SI,
Hoste EA. Outcome of acute kidney injury in severe burns: a
systematic review and meta-analysis. Intensive Care Med
2010;36:915-25.
10. Osler T, Glance LG, Hosmer DW. Simplied estimates of the
probability of death aer burn injuries: extending and updating
the baux score. J Trauma 2010;68:690-7.
11. Williams DJ, Walker JD. A nomogram for calculation of the
Revised Baux Score. Burns 2015;41:85-90.
12. Woods JF, Quinlan CS, Shelley OP. Predicting Mortality in
Severe Burns – What Is the Score?: Evaluation and Comparison
of 4 Mortality Prediction Scores in an Irish Population. Plast
Reconstr Surg Glob Open 2016;4:e606.
13. Emara SS, Alzaylai AA. Renal failure in burn patients: a review.
Ann Burns Fire Disasters 2013;26:12-15.
14. Ibrahim AE, Sarhane KA, Fagan SP, Goverman J. Renal dysfunction
in burns: a review. Ann Burns Fire Disasters 2013;26:16-25.
15. Cartotto R, Zhou A. Fluid creep: the pendulum hasn’t swung
back yet! J Burn Care Res 2010;31:551-8.
16. Strang SG, Van Lieshout EM, Breederveld RS, Van Waes OJ.
A systematic review on intra-abdominal pressure in severely
burned patients. Burns 2014;40:9-16.
17. Klein MB, Hayden D, Elson C, Nathens AB, Gamelli RL,
Gibran NS, et al. e association between uid administration
and outcome following major burn: a multicenter study. Ann
Surg 2007;245:622-8.
18. Jeschke MG, Patsouris D, Stanojcic M, Abdullahi A, Rehou S,
Pinto R, et al. Pathophysiologic Response to Burns in the
Elderly. EBioMedicine 2015;2:1536-48.
19. Wu G, Xiao Y, Wang C, Hong X, Sun Y, Ma B, et al. Risk
Factors for Acute Kidney Injury in Patients With Burn Injury: A
Meta-Analysis and Systematic Review. J Burn Care Res 2017;
38:271-82.
20. Sood RF, Gibran NS, Arnoldo BD, Gamelli RL, Herndon DN,
Tompkins RG, et al. Early leukocyte gene expression associated
with age, burn size, and inhalation injury in severely burned
adults. J Trauma Acute Care Surg 2016;80:250-7.
21. Sakallioglu AE, Basaran O, Karakayali H, Ozdemir BH, Yucel
M, Arat Z, et al. Interactions of systemic immune response
and local wound healing in dierent burn depths: an experimental
study on rats. J Burn Care Res 2006;27:357-66.
22. Farina JA, Jr., Rosique MJ, Rosique RG. Curbing inammation
in burn patients. Int J Inam 2013;2013:715645.
23. Schneider DF, Dobrowolsky A, Shakir IA, Sinacore JM, Mosier
MJ, Gamelli RL. Predicting acute kidney injury among burn
patients in the 21
st
century: a classication and regression tree
analysis. J Burn Care Res 2012;33:242-51.
24. Pantet O, Faouzi M, Brusselaers N, Vernay A, Berger MM.
Comparison of mortality prediction models and validation
of SAPS II in critically ill burns patients. Ann Burns Fire
Disasters 2016;29:123-9.
25. Salehi SH, As’adi K, Abbaszadeh-Kasbi A, Isfeedvajani MS,
Khodaei N. Comparison of six outcome prediction models in
an adult burn population in a developing country. Ann Burns
Fire Disasters 2017;30:13-17.
Fig 1. AUC of revised BAUX score at cut o value of 100.
Sensitivity 83.9%
Specicity 61.9%
PPV 37.7%
NPV 93.3%
Accuracy 66.7%
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
158
Popchai Ngamskulrungroj, M.D.,Ph.D.*, Patcharee Kammarnjassadakul, M.D.,Ph.D.**, Kantarawee Khayhan,
M.D. ,Ph.D.***, Poramate Pitak-Arnnop, M.D.D.D.S.,Ph.D.****, Pathompong Ungprasert, M.D.*****, Sujiraphong
Pharkjaksu, M.D.,M.Sc.*
*Department of Microbiology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, **Department of Clinical Microbiology, Faculty
of Medical Technology, Huachiew Chalermprakiet University, Samut Prakan 10270, ***Department of Microbiology and Parasitology, Faculty of Medical
Sciences, University of Phayao, Phayao 56000, ailand, ****Department of Oral and Maxillofacial Surgery, University Hospital of Marburg UKGM,
Faculty of Medicine, Philips University, Marburg, Germany, *****DepartmentofResearchandDevelopment,FacultyofMedicineSirirajHospital,Mahidol
University, Bangkok 10700, ailand.
An Association of Cryptococcus neoformans/
C. gattii Genotype and HIV Status in Asia:
A Systematic Review
Corresponding author: Popchai Ngamskulrungroj
E-mail: popchai.nga@mahidol.ac.th
Received 21 August 2018 Revised 12 November 2018 Accepted 4 February 2019
ORCID ID: http://orcid.org/0000-0001-5162-5498
http://dx.doi.org/10.33192/Smj.2019.24
ABSTRACT
Objective: It has been known that VNI molecular type of Cryptococcus neoformans/C. gattii is strongly associated
with HIV patients. However, this paradigm has recently been challenged because of the high prevalence of VNI
molecular type among non-HIV patients with cryptococcosis in East Asia. e purpose of this study was to answer
the question: “Among cryptococcosis in Asia, is there an association between the genotype and the patient’s HIV
status?”
Methods: Using a systematic review and meta-analysis study design, we included all relevant published data, which
were any type of study designs, mainly studied clinical Cryptococcus neoformans/C. gattii strains isolated in Asia
and had available molecular typing data. e primary study variables were Cryptococcus neoformans molecular type
(VNI/non-VNI or ST5/non-ST5) and the HIV status of the patients at the time of diagnosis. We used a random-
eects meta-analysis model to estimate the prevalence of HIV infection.
Results: Sixteen retrospective descriptive studies during 2005 – 2018 (1,584 isolates) were included. Most of the
cryptococcosis cases in East Asian countries were in non-HIV patients (72.4-81.8%), which diered from non-
East Asian countries (2.6-28.3% associated with non-HIV patients). In East Asia, the HIV prevalence among VNI
and ST5 infected patients ranged from 7.5% - 46.7% with the pooled prevalence of 19.8% (95% CI, 12.2% - 30.4%)
and 5.3% - 52.4% with the pooled prevalence of 19.9% (95% CI, 6.9% - 45.3%), respectively. In non-East Asia, the
HIV prevalence among VNI and ST5 infected patients ranged from 48.3% - 98.8% with the pooled prevalence of
81.9% (95% CI, 73.3% - 88.2%) and 52.3% - 88.0% with the pooled prevalence of 74.9% (95% CI, 40.7% - 92.8%),
respectively. Statistical heterogeneity was high in both analyses with the I
2
of 79-89% in all analyses.
Conclusion: Our results conrmed the low prevalence of HIV prevalence among VNI and ST5 strains in East
Asian countries. e emergence of high virulence genotype causing disease in non-HIV patient is highly unlikely,
because the VNI and ST5 were associated with HIV patients in other Asian countries. It can be hypothesized that
the low HIV prevalence among VNI and ST5 strains in East Asian is due to the high susceptibility to cryptococcosis
of people living in this region. is requires further investigation.
Keywords: Cryptococcosis; VNI; Cryptococcus neoformans; HIV; Asia (Siriraj Med J 2019;71: 158-164)
Ngamskulrungroj et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
159
Original Article
SMJ
INTRODUCTION
Cryptococcus neoformans and C. gattii play an
important role in today’s medical practice and have received
intensive attention from researchers/clinicians because
they cause not only meningitis but also pneumonia in
human worldwide.
1
e two species are dierent in many
aspects. is includes environmental niche, molecular
epidemiology, pathogenesis and clinical manifestations
once they cause the disease, the so-called “cryptococcosis”.
1
It is known that C. neoformans is associated mainly
with immunosuppression, especially among HIV-infected
patients, while C. gattii generally causes cryptococcosis in
immunocompetent patients.
2
However, some reports have
demonstrated contradictory results. Recent studies from
China and South Korea showed that most of cryptococcosis
patients caused by C. neoformans were HIV-negative
patients without other immunocompromised conditions.
3,4
Especially in China, cryptococcosis frequently occurred
in immunocompetent patients.
4,5
C. neoformans is composed of 4 major molecular types,
VNI, VNII, VNIII, and VNIV and C. gattii is composed
of 4 major molecular types, VGI, VGII, VGIII and VGIV.
6
e VNI-molecular type of C. neoformans is the most
common cryptococcal molecular type and typically
associated with HIV patients.
7
However, VNI in China
was associated with non-HIV patients.
4
Further molecular
typing among the VNI strains by the Multi Locus Sequence
Typing (MLST)
8
revealed the non-HIV-associated VNI
strains was almost exclusively belonged to the sequence
type 5 (ST5) genotype.
4
is phenomenon has also been
found in Korea
3
and Japan
9
erefore, the association
of this VNI-ST5 genotype and non-HIV status may be
possible. However, although ais are a close sibling of
Chinese, there was no association between VNI-ST5 and
non-HIV status.
10
e East Asian race is highly susceptible
to VNI-ST5 or is a high virulent genotype capable of
causing infection in immunocompetent individuals.
A search of the published data revealed no evidence
of any randomized trials or systematic reviews addressing
the association of the VNI-ST5 and HIV status in Asian
patients with cryptococcosis. e purpose of the present
report was to complete a systematic review and meta-
analysis to answer the following research question:
“Among cryptococcosis in Asia, is there an association
between the genotype and the patient’s HIV status?”
We hypothesized that such association is limited only
to patients in East Asia – China and neighboring. Our
specic aims were to perform a systematic review of the
published data and identify studies for analysis and to
execute a meta-analysis to determine whether the VNI-
ST5 causes diseases mainly in non-HIV patients across
all Asian countries.
MATERIALS AND METHODS
Study design
To address our research question, we designed and
implemented a systematic literature review and meta-analysis.
Samples
Our review of the published data was designed and
conducted in concordance with the PRISMA (Preferred
Reporting Items for Systematic Reviews and Meta-
Analyses) guidelines.
11
ree authors (P. N., P. K. and
K. K.) searched the published studies from January
1999 (The year that cryptococcal molecular typing
system was rstly established
12
) to May 2018, using the
National Library of Medicine (PubMed, available from
http://www.pubmed.gov), the Web of Science (available
from https://webonowledge.com), and the Cochrane
database (available from http://www.cochranelibrary.
com/) using specic medical subject headings and key
words, including (Cryptococ*) AND (molecular typ* OR
molecular epidemiology). Because each Cryptococcus species
comprises 4 major molecular types: C. neoformans, VNI
(AFLP1), VNII (AFLP1A of AFLP1B), VNIII (AFLP3)
and VNIV (AFLP2); and C. gattii, VGI (AFLP4), VGII
(AFLP6), VGIII (AFLP5) and VGIV (AFLP7) (REF),
we also used the specic terms for all major molecular
types (VNI OR VNII OR VNIII OR VNIV OR VGI OR
VGII OR VGIII OR VGIV OR AFLP1 OR AFLP1A OR
AFLP1B OR AFLP2 OR AFLP3 OR AFLP4 OR AFLP6
OR AFLP5 OR AFLP7) AND (Cryptococ*) to identied
all publications investigating cryptococcal molecular
typing. Citations in the accessed articles were further
analyzed for additional published data.
In addition, the available post-meeting online
materials of relevant meetings including American
Society for Microbiology (ASM) 2010-2015, 48
th
-55
th
Interscience Conference on Antimicrobial Agents and
Chemotherapy (ICAAC), ASM microbe 2016-2017,
International Society for Human & Animal Mycology
(ISHAM) 2000-2015 and International Conference on
Cryptococcus and Cryptococcosis (ICCC) 4
th
-10
th
were
searched for additional eligible studies.
Study inclusion and exclusion criteria
Studies were included if they were mainly studied
Asian isolates, provided data on statistics, patient’s
HIV status and major molecular type data. Searching
methodology was mentioned above. We excluded studies
that, selectively included only one type of immune status
of patients or genotype of the strains, were case-reports
or case-series, did not identify the correlation between
HIV status with individual isolates or contained 100%
duplicated cryptococcal isolates.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
160
Study variables
e primary predictor variable was the genotype
(i.e., major molecular type or sequence type). e main
outcome variable was the patient’s HIV status. The
primary analysis of interest was the prevalence of the
patient’s HIV positivity across study groups (i.e., VNI and
ST5 in East Asia [China, Japan, Taiwan, and Korea] and
Non-East Asia [ailand, Malaysia, Singapore, Vietnam,
Indonesia, India, Kuwait, and Qatar] countries). e
other study variables were classied as demographic,
anatomic, environmental and genetic. e demographic
data were total sample size, number of subjects in both
genotype groups, and mean patient’s age. e anatomic
location of infection, environmental niche and genetic
proles were also recorded.
Data extraction and statistical analysis
Using a standardized data extraction form, two
authors (P. Ng. and K. K.) extracted and tabulated all
data. Discrepancies were resolved by group discussion of
the investigator authors. As standard quality assessment
of descriptive study for meta-analysis was not available,
a quality assessment for case-control studies was used
according to a previous published method.
13
A quality
scale for observational study was used. A narrative
overview is provided summarizing the data gathering from
included published data. e pooled prevalence of HIV
positivity and 95% CI for ST5 and VNI were calculated
using DerSimonian-Laird random-eect model with
double arcsine transformation.
14
e pooled prevalence
and 95% CI were calculated for each region (East Asia
and non-East Asia). Random-eect model, rather than
xed-eect, was used because of the high likelihood
of between-study heterogeneity. Cochran’s Q test and
I
2
statistic were used to determine the between-study
heterogeneity. is I
2
statistic quanties the proportion of
total variation across studies that is due to heterogeneity
rather than chance. A value of I
2
of 0% to 25% represents
insignicant heterogeneity, greater than 25% but less than
or equal to 50% represents low heterogeneity, greater than
50% but less than or equal to 75% represents moderate
heterogeneity, and greater than 75% represents high
heterogeneity.
15
e soware Comprehensive meta-
analysis (Englewood, NJ, USA) and IBM SPSS statistics
18.0 (IBM (ailand), Bangkok, ailand) were used for
statistical analysis.
Research ethics
e recommendations of the Helsinki declaration
were thoroughly maintained during this study. Since
this study did not involve human subjects or records
directly, ethical approval by an ethics committee and
consent from the authors of the articles studied were
not required.
RESULTS
Search results
e literature searches initially yielded 1,630 potentially
relevant studies; titles and abstracts were screened, and
full papers were obtained for relevant articles (Fig 1).
A total of 36 full-text articles were reviewed, of which
20 were excluded for various reasons. e remaining 16
studies were nally included in this meta-analysis and
yielded a total of 1,584 isolates.
3-5,7,10,16-26
ere was no
additional published data from citations in the accessed
articles.
Characteristics and quality of included studies
e characteristics of the 16 included studies are
shown in Table 1. All were retrospective descriptive studies.
Quality of the included studies was scored 4.5 to 8.5 out of
14.5. All included studies reported the data on HIV status
in at least some of the isolates but the availability of this
data varied considerably across the studies, ranging from
13.5% to 100% (median 80.2%). Only strains with HIV
status data were used in this current analysis. Only 7 studies
had information of ST5 genotype with HIV data (Table 1).
Prevalence of HIV and genotype across all studies of
cryptococcosis
Cryptococcosis in Asia was not common in non-
HIV patients (2.6-28.3%), except in East Asia (72.4-
81.8%) and Northwest Asia (60-70%). Interestingly,
VNI molecular type was predominant across all Asian
countries, regardless of HIV status of the patients. Among
the VNI strains, ST5 was more prevalent than non-ST5
only in East Asia. e results were summarized in Fig 2.
HIV prevalence in East Asian countries
The HIV prevalence among VNI-infected cryptococcosis
patients from the included 8 studies ranged from 7.5%
- 46.7% with the pooled prevalence of 19.8% (95% CI,
12.2% - 30.4%) (Fig 3). e HIV prevalence among ST5
infected patients from the 5 studies ranged from 5.3%
- 52.4% with the pooled prevalence of 19.9% (95% CI,
6.9% - 45.3%) (Fig 4). Statistical heterogeneity was high
in both analyses with the I
2
of 89% in both analyses.
HIV prevalence in non-East Asian countries
The HIV prevalence among VNI-infected cryptococcosis
patients from the 9 studies ranged from 48.3% - 98.8%
with the pooled prevalence of 81.9% (95% CI, 73.3% -
Ngamskulrungroj et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
161
Original Article
SMJ
Fig 1. Study selection process.
Fig 2. Summary of the collected
data; n=a total number of strains
with the major molecular type
information (*Availability of the
ST data was varied: China = 234,
Korea = 13, Japan = 20, Taiwan =
0, ailand = 195, Malaysia = 0,
Singapore = 0, Vietnam = 136,
Indonesia = 38, India = 58, Kuwait
= 10, and Qatar = 5; **Availability
of HIV data was varied: China =
349, Korea = 59, Japan = 20, Taiwan
= 293, ailand = 87, Malaysia =
62, Singapore = 62, Vietnam = 136,
Indonesia = 38, India = 172, Kuwait
= 10, and Qatar = 5).
Fig 3. Meta-analysis of HIV prevalence in VNI
infected patients. Khayhan et al. contained data of
China and Japan for East Asia, and India, Indonesia,
Kuwait, Qatar and ailand for non-East Asia.
88.2%) (Fig 3). e HIV prevalence among ST5-infected
patients from the 3 studies ranged from 52.3% - 88.0%
with the pooled prevalence of 74.9% (95% CI, 40.7%
- 92.8%) (Fig 4). Statistical heterogeneity was high in
both analyses with the I
2
of 79% in both analyses.
DISCUSSION
e study’s principal aim was to identify the particular
association between the cryptococcosis genotypes VNI/
ST5 and the non-HIV status of patients in Asia. We
hypothesized that this association would be seen only in
East Asia countries. Our specic aims were to conduct
a systematic review and meta-analysis to identify the
relationship between the VNI/ST5 strain and the patient’s
HIV status.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
162
TABLE 1. Characteristics of the included studies.
Country No. of
Study center
HIV data HIV Primary
Study
of origin isolates
Study design Years
types
coverage prevalence typing
(%) (%) method
Dou et al.
China 83
Descriptive, 2007-
2015* retrospective 2013
Multicenter 80.7 32.8 MLST
Feng et al.
China 115
Descriptive, 1976-
2008 retrospective 2007
Multicenter 67.8 3.8 M13
Chen et al.
China 129
Descriptive, 1980-
2008* retrospective 2006
Multicenter 100.0 8.6 M13
Chowdhary
India 160
Descriptive, 2002-
et al. 2011 retrospective 2009
Multicenter 85.6 84.7 M13
Jain et al.
India 57
Descriptive, before
2005 retrospective 2005
Single center 61.4 40 URA5
Choi et al.
Korea 78
Descriptive, 1990-
2010* retrospective 2008
Multicenter 79.5 22.6 M13
Tay et al.
Malaysia 96
Descriptive, 2003-
2010 retrospective 2004
Multicenter 39.6 94.7 URA5
Tay et al.
Malaysia 78
Descriptive, 1980-
2006 retrospective 2003
Multicenter 30.8 87.5 URA5
Chan et al.
Singapore 62
Descriptive, 1999-
2014 retrospective 2007
Single center 100.0 80.6 URA5
Tseng et al.
Taiwan 219
Descriptive, 1997-
2013 retrospective 2010
Multicenter 89.0 27.7 M13
Liaw et al.
Taiwan 100
Descriptive, 1994-
2010 retrospective 2004
Single center 98.0 27.6 M13
Kaocharoen
Thailand 386
Descriptive, 1993-
et al. 2013 retrospective 2005
Multicenter 13.5 88.5 M13
Hatthakaroon
Thailand 51
Descriptive, 2012-
et al. 2017* retrospective 2014
Single center 68.6 77.1 MLST
Day et al.
Vietnam 151
Descriptive, 1996-
2017* retrospective 2010
Single center 90.1 72.1 MLST
Khayhan
Multiple** 476
Descriptive, 1983-
et al. 2013* retrospective 2009
Multicenter 76.1 62.7 MLST
Chen et al.
China 86
Descriptive, 2016-
2018* retrospective 2017
Single center 87.2 46.7 MLST
*ST5/HIV data available, **China, India, Indonesia, Japan, Kuwait, Qatar, ailand
Ngamskulrungroj et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
163
Original Article
SMJ
Fig 4. Meta-analysis of HIV prevalence in ST5 infected patients. Khayhan et al. contained data of China for East Asia, and Qatar and
ailand for non-East Asia.
Cryptococcosis, especially from C. neoformans
VNI molecular type, has been found to be linked to HIV
infection since the HIV pandemic in 1980s.
1,6
Our study
conrmed this association. e pooled HIV prevalence
among non-East Asia patients with VNI infection was
>80%. However, contradictory observations from China
and other East Asian countries have been reported since
2008.
3,4,21
Our analysis included the results of those
studies and demonstrated a strong link between the VNI
molecular type and non-HIV patients in this region with
the low pooled prevalence of HIV among VNI-infected
cryptococcosis patients (only 19.8%). Although the exact
mechanisms behind this observation are not known, few
possible explanations include the highly virulent VNI
strain of C. neoformans and/or the high susceptibility
of individuals living in those areas.
e MLST has been introduced and used to study
the association between the VNI molecular type and
non-HIV patient’s status.
8
is technique provides more
genotypic details of microorganisms. With this particular
technique, ST5 was found to be highly prevalent (86.9%)
among the VNI molecular type strains in East Asia.
4,5
is
might be implied that ST5 is a highly virulent genotype
among VNI molecular type strains and specically infects
non-HIV individuals. In fact, our analysis demonstrated
the similar nding that the pooled HIV prevalence among
ST5 strains was low in East Asia (19.9%). High pooled
HIV prevalence was found in other Asia countries only
(74.9%). As one would expect that if the ST5 strains
is high virulent, it should be able to infect non-HIV
individual regardless of their countries. erefore, the
assumption about the highly virulent genotype causing
infection in non-HIV individuals should be omitted.
Another assumption is that people in East Asia
might be susceptible to cryptococcosis more than those
in other Asian region. A recent environmental study in
China revealed that ST5 prevalence in nature was low
(6.2%); thereby, one must be highly susceptible to the
causal microorganism to be infected. In contrast, ST31
was commonly found in China (66.7%)
27
, but rarely
caused any diseases (only 1.3% among cryptococcosis in
East Asian countries, data not shown). However, it is still
unclear whether this high susceptibility is due to genetic
background or other factors. Molecular epidemiological
studies focused on patient’s genetics are still warrant.
Our study suers from many limitations. First, not
all published data provided underlying diseases other than
HIV. e non-HIV patients may be immunocompromised
because of other conditions. is can be considered as the
common pitfall and limitation in evidence-based medicine
28
and may help to explain the extraordinary relationship
between the non-HIV patient and cryptococcosis in
East Asia. Second, high statistical heterogeneity in this
study may arise from many factors, such as dierence
in methodology and quality of study. Lastly, all of the
included studies were descriptive. e overall numbers of
isolates/patients in some countries were relatively small
and may not be representative to the whole nation of
the country. Taken all together, the results of this study
should be interpreted with caution.
CONCLUSION
Our results revealed the low HIV prevalence in
cryptococcosis patients with VNI and ST5 strains in East
Asia. e emergence of high virulence genotype causing
disease in non-HIV patient seems to be impossible, because
the VNI and ST5 strains were mostly associated with
HIV patients in other Asian regions. erefore, the high
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
164
susceptibility to cryptococcosis of people living in East
Asia may be caused by the low HIV prevalence among
VNI and ST5 strains in this geographic area. However,
mechanism of this possibly high susceptibility requires
further investigations.
Disclosure of potential conicts of interest
e authors indicate full freedom of manuscript
preparation and no potential conicts of interest.
REFERENCES
1. Kwon-Chung KJ, Fraser JA, Doering TL, Wang Z, Janbon G,
Idnurm A, et al. Cryptococcus neoformans and Cryptococcus
gattii, the etiologic agents of cryptococcosis. Cold Spring Harb
Perspect Med 2014;4:a019760.
2. Kwon-Chung KJ, Saijo T. Is Cryptococcus gattii a Primary
Pathogen? J Fungi (Basel) 2015;1:154-67.
3. Choi YH, Ngamskulrungroj P, Varma A, Sionov E, Hwang SM,
Carriconde F, et al. Prevalence of the VNIc genotype of
Cryptococcus neoformans in non-HIV-associated cryptococcosis
in the Republic of Korea. FEMS Yeast Res 2010;10:769-78.
4. Chen J, Varma A, Diaz MR, Litvintseva AP, Wollenberg KK,
Kwon-Chung KJ. Cryptococcus neoformans strains and infection
in apparently immunocompetent patients, China. Emerg Infect
Dis 2008;14:755-62.
5. Chen YH, Yu F, Bian ZY, Hong JM, Zhang N, Zhong QS, et
al. Multilocus Sequence Typing Reveals both Shared and Unique
Genotypes of Cryptococcus neoformans in Jiangxi Province,
China. Sci Rep 2018;8:1495.
6. Meyer W, Castaneda A, Jackson S, Huynh M, Castaneda E,
IberoAmerican Cryptococcal Study G. Molecular typing of
IberoAmerican Cryptococcus neoformans isolates. Emerg
Infect Dis 2003;9:189-95.
7. Khayhan K, Hagen F, Pan W, Simwami S, Fisher MC, Wahyuningsih
R, et al. Geographically structured populations of Cryptococcus
neoformans Variety grubii in Asia correlate with HIV status
and show a clonal population structure. PLoS One 2013;8:e72222.
8. Meyer W, Aanensen DM, Boekhout T, Cogliati M, Diaz MR,
Esposto MC, et al. Consensus multi-locus sequence typing
scheme for Cryptococcus neoformans and Cryptococcus gattii.
Med Mycol 2009;47:561-70.
9. Mihara T, Izumikawa K, Kakeya H, Ngamskulrungroj P,
Umeyama T, Takazono T, et al. Multilocus sequence typing
of Cryptococcus neoformans in non-HIV associated cryptococcosis
in Nagasaki, Japan. Med Mycol 2013;51:252-60.
10. Hatthakaroon C, Pharkjaksu S, Chongtrakool P, Suwannakarn
K, Kiratisin P, Ngamskulrungroj P. Molecular epidemiology
of cryptococcal genotype VNIc/ST5 in Siriraj Hospital, ailand.
PLoS One 2017;12:e0173744.
11. Swartz MK. e PRISMA statement: a guideline for systematic
reviews and meta-analyses. J Pediatr Health Care 2011;25:1-2.
12. Meyer W, Marszewska K, Amirmostofian M, Igreja RP,
Hardtke C, Methling K, et al. Molecular typing of global isolates
of Cryptococcus neoformans var. neoformans by polymerase
chain reaction ngerprinting and randomly amplied polymorphic
DNA-a pilot study to standardize techniques on which to base
a detailed epidemiological survey. Electrophoresis 1999;20:
1790-9.
13. La Torre G, Chiaradia G, Gianfagna F, De Laurentis A, Boccia
S, Ricciardi W. Quality assessment in meta-analysis. Ital J
Public Health 2006;3:44-50.
14. Mathes T, Kuss O. A comparison of methods for meta-analysis
of a small number of studies with binary outcomes. Res Synth
Methods 2018;9:366-81.
15. Higgins JP, ompson SG, Deeks JJ, Altman DG. Measuring
inconsistency in meta-analyses. BMJ 2003;327:557-60.
16. Chowdhary A, Randhawa HS, Sundar G, Kathuria S, Prakash A,
Khan Z, et al. In vitro antifungal susceptibility proles and
genotypes of 308 clinical and environmental isolates of Cryptococcus
neoformans var. grubii and Cryptococcus gattii serotype B
from north-western India. J Med Microbiol 2011;60:961-7.
17. Dou HT, Xu YC, Wang HZ, Li TS. Molecular epidemiology of
Cryptococcus neoformans and Cryptococcus gattii in China
between 2007 and 2013 using multilocus sequence typing and
the DiversiLab system. Eur J Clin Microbiol Infect Dis 2015;34:
753-62.
18. Feng X, Yao Z, Ren D, Liao W, Wu J. Genotype and mating
type analysis of Cryptococcus neoformans and Cryptococcus
gattii isolates from China that mainly originated from non-
HIV-infected patients. FEMS Yeast Res 2008;8:930-8.
19. Tay ST, Lim HC, Tajuddin TH, Rohani MY, Hamimah H, ong
KL. Determination of molecular types and genetic heterogeneity
of Cryptococcus neoformans and C. gattii in Malaysia. Med
Mycol 2006;44:617-22.
20. Tay ST, Rohani MY, Hoo TS, Hamimah H. Epidemiology of
cryptococcosis in Malaysia. Mycoses 2010;53:509-14.
21. Kaocharoen S, Ngamskulrungroj P, Firacative C, Trilles L,
Piyabongkarn D, Banlunara W, et al. Molecular epidemiology
reveals genetic diversity amongst isolates of the Cryptococcus
neoformans/C. gattii species complex in ailand. PLoS Negl
Trop Dis 2013;7:e2297.
22. Jain N, Wickes BL, Keller SM, Fu J, Casadevall A, Jain P, et al.
Molecular epidemiology of clinical Cryptococcus neoformans
strains from India. J Clin Microbiol 2005;43:5733-42.
23. Chan M, Lye D, Win MK, Chow A, Barkham T. Clinical
and microbiological characteristics of cryptococcosis in
Singapore: predominance of Cryptococcus neoformans compared
with Cryptococcus gattii. Int J Infect Dis 2014;26:110-5.
24. Tseng HK, Liu CP, Ho MW, Lu PL, Lo HJ, Lin YH, et al.
Microbiological, epidemiological, and clinical characteristics
and outcomes of patients with cryptococcosis in Taiwan,
1997-2010. PLoS One 2013;8:e61921.
25. Liaw SJ, Wu HC, Hsueh PR. Microbiological characteristics of
clinical isolates of Cryptococcus neoformans in Taiwan: serotypes,
mating types, molecular types, virulence factors, and antifungal
susceptibility. Clin Microbiol Infect 2010;16:696-703.
26. Day JN, Qihui S, anh LT, Trieu PH, Van AD, u NH, et al.
Comparative genomics of Cryptococcus neoformans var. grubii
associated with meningitis in HIV infected and uninfected
patients in Vietnam. PLoS Negl Trop Dis 2017;11:e0005628.
27. Dou H, Wang H, Xie S, Chen X, Xu Z, Xu Y. Molecular
characterization of Cryptococcus neoformans isolated from
the environment in Beijing, China. Med Mycol 2017;55:737-47.
28. Pitak-Arnnop P, Hemprich A, Pausch NC. Evidence-based
oral and maxillofacial surgery: some pitfalls and limitations.
J Oral Maxillofac Surg 2011;69:252-7.
Ngamskulrungroj et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
165
Original Article
SMJ
Lalita Matthapan, B.Sc.*, Sumanas Bunyaratavej, M.D.*, Charussri Leeyaphan, M.D.*, Chuda Rujitharanawong,
M.D.*, Pichaya Limphoka, M.D.*, Waranyoo Prasong, B.Sc.*, Chulaluk Komoltri, DrPh.**
*Department of Dermatology, **Department of Research Development, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.
Clinical Application of Toothbrush Technique for
Specimen Collection of Tinea Capitis
Corresponding author: Chulaluk Komoltri
E-mail: chulaluk.kom@mahidol.ac.th
Received 15 November 2017 Revised 3 August 2018 Accepted 11 September 2018
ORCID ID: http://orcid.org/0000-0001-8620-1126
http://dx.doi.org/10.33192/Smj.2019.25
ABSTRACT
Objective: To demonstrate the ecacy of toothbrush culture test. Moreover, inuential factors which aected the
success of these methods were analyzed.
Methods: To retrospectively analyze clinical presentation and investigation data of 59 Buddhist novices who diagnosed
TC. e laboratory data of the toothbrush technique was reviewed and the ecacy of this technique was investigated.
Results: e ecacy of toothbrush technique was demonstrated as 49.2% by comparing to the standard method. e
analysis proved that more than 15% area TC involvement of scalp group achieved higher positive fungal culture outcome
to be 65.5 % (p-value = 0.013). In the aspect of the clinical presentation, the eectiveness of the hairbrush technique
signicantly increased in the gray patch TC group (p-value 0.024). For the reason that the toothbrush technique
demonstrated statistically higher ecacy in a group of patients with > 15 % gray patch area involvement (p-value = 0.007).
Conclusion: e toothbrush technique was a good cooperation method that was recommended to apply in TC
patients who presented as gray-patch type and more than 15% scalp area involvement.
Keywords: Tinea capitis; ectothrix dermatophyte infection; gray-patch type; toothbrush technique; hair-plucking
technique (Siriraj Med J 2019;71: 165-167)
INTRODUCTION
Tinea capitis (TC) is a contagious condition that
commonly diagnosed in children due to immature sebum
production.
1
Outbreaks of TC usually especially occurred
in poor personal hygiene and low socioeconomic area. TC
is reported in several areas in the world. Interestingly, the
dierence of causative organisms is demonstrated in the
geographic region.
2
For example, Trichophyton tonsurans
is the most reported species in the United States.
3
e
variety of etiologic organisms in dierent countries in Asia
is demonstrated. Trichophyton violaceum is commonly
found in India,
4
whereas, Microsporum canis is the main
causative organism in ailand.
5
Moreover, the crossover
organism is detected in each outbreak. Asymptomatic
carriers are assumed as a major role in the spread of
infection.
6
e carrier is dened as an individual who
has dermatophyte-positive scalp culture without any
appearance of TC.
7
Clinical presentations of TC depend
on the degree of hair invasion and inammation that
varies from a scaly noninamed lesion to erythematous
lesions with permanent alopecia. Ectothrix remains a
dermatophyte infection that conned to hair surface. In
contrast, endothrix refers to the invasion of dermatophyte
into the hair sha and internalize into the hair cell.
Practically, clinical presentations are categorized into 4
types as follows: (1) gray-patch type: single or multiple
scaly patches with lack of inammation, (2) black dot
type: broken hair at the level of the scalp on nonscarring
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
166
alopecia (3) kerion: intense inamed plaques on scarring
alopecia and (4) favus: chronic inammation with thick
yellow crusts on scarring alopecia.
4
In the mycological
laboratory, standard investigation of TC is a microbial
culture from a specimen obtained by hair-plucking
technique. is technique was collecting infected hairs
and scraping scale by a sharp blade.
8
However, pain
usually was the major complaint of patients of the
standard technique, furthermore, the panic of the blade
was observed in children. Owing to disadvantages, non-
invasive techniques such as cotton swab and toothbrush
were developed to collect infected scalp scales.
9,10
Toothbrush technique was gently performed on the
suspected area by sterile toothbrush for scale scratching
and then pressed onto the surface of sabouraud dextrose
agar (SDA) culture medium, consequently. However,
studies about the toothbrush technique were limited in
number. Accordingly, the main purpose of the study is
to determine the ecacy of the toothbrush technique,
comparing with standard technique focusing on specimen
collection in TC patients. Duration and satisfaction on
specimen collection were also analyzed.
MATERIALS AND METHODS
Clinical and laboratory data of 59 Buddhist novices
who conrmed TC diagnosis by the standard technique
were reviewed in this retrospective study. is study was
approved by Siriraj Institutional Review Board, Faculty
of Medicine Siriraj Hospital (Si 624/2014) and consent
from the Royal ai Monastery in India. Demographic
and clinical data of novices whom toothbrush technique
was used to collect specimens concurrently with standard
method were included in the analysis. Ecacy of toothbrush
techniques was reviewed and applicable situations were
explored.
In this study, statistical analysis was achieved by
Statistical Package for the Social Sciences (SPSS, Inc.,
Chicago, IL, USA) version 22. To compare the ecacy
of the toothbrush technique with the standard technique
was analyzed by Mann-Whitney U test. Additionally,
Fisher’s exact test was applied to identify signicant
factors that inuence the ecacy of the toothbrush
technique. e reliability of all statistical signicance
was demonstrated with a p-value that was less than 0.05.
Specimen collection
All participants have performed scalp examination
to detect broken hairs and scaling. All the lesions were
cleansed with 70% alcohol. Then, both of standard
technique and toothbrush technique was applied to
collect a specimen from all area of the scalp with the
addition of a suspected area such as an area having the
scale or broken hair until obtaining an adequate amount
of specimen.
10
Aer that, the specimen was placed onto
the medium and continue the incubation process. e
evidence of growth was recorded in the patient’s medical
history. e laboratory data were enrolled in this study
to further analysis.
RESULTS
All of Fiy-nine of Buddhist novices with TC, the
mean age (SD) of patients was 11.6 (1.9) years. All cases
had positive fungal culture from standard technique
including hair-plucking technique and scraping scale by
the sharp blade to conrmed diagnosis TC. In the aspect
of epidemiology, Trichophyton violaceum was the major
isolated TC pathogen (33 cases, 55.9%). Additionally,
T. mentagrophytes (27 cases, 45.8%), Microsporum canis
(21 cases, 35.6%), and T. tonsurans (8 cases, 13.6%) were
reported respectively. Mixed dermatophyte infection
was noticeable in 34 cases (57.6%).
To compare outcomes between toothbrush technique
and standard technique, the ecacy of the toothbrush
technique to collect a specimen for culture was 49.2% (29
cases). Adjacent areas of scalp including face, neck, ear,
and shoulder areas were dened as extended scalp areas.
Extended area involvement was associated with more
ecacy of toothbrush technique. (p-value = 0.036). In
this study, median scalp area involvement in patients with
successful toothbrush techniques was 20%, compared with
patients whose toothbrush failed to collect a specimen
(11.5%). More than 15% area involvement was signicantly
higher ecacy of this technique that demonstrated by
the increased ecacy to be 65.5 % (19 cases, p-value =
0.013).
Clinical presentations were considered and described
as follows: (1) gray-patch type (51 cases, 86.4%) (2) black
dot type (2 cases, 3.4%) (3) kerion (37 cases, 62.7%).
No favus type was reported in this study. Focus on the
gray-patch type, scalp area involvement was related with
positive toothbrush technique. To demonstrate in the
gray-patch group, the median scalp area involvement of
positive toothbrush technique patients was 25% while
other types of TC were 11.5% (p-value 0.024).
Patients with >15 % gray patch area involvement
had a statistically higher probability to obtain a positive
fungal culture from the toothbrush technique (p-value
= 0.007). Ecacy of toothbrush technique through the
patient selection with gray-patch type and more than
15% of scalp area involvement was increasing from
49.2% to 66.7%.
Matthapan et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
167
Original Article
SMJ
DISCUSSION
Outbreaks of TC were still reported as global health
care problems and economic burden. TC remained a
common disease in childhood. To conrm the diagnosis
of TC, positive fungal infection in hair or scalp was
performed by hair-plucking technique and scraping
scale by a sharp blade which was the standard technique.
However, most children are not well-cooperated due to
pain and panic. Previous studies reported noninvasive
techniques including hairbrush, cytobrush and toothbrush
techniques. Toothbrush technique was performed by
sterile toothbrush and rub it in gently on the aected area.
is technique was preferred to apply in asymptomatic
carrier patients.
11
According to this study, the toothbrush
technique was also practical in patients with symptomatic
TC, however, the patient selection by clinical presentation
is considered as the important points. is study showed
that patients with gray patch and extended scalp area
involvement were most suitable for the toothbrush
technique.
e clinical presentations were categorized into 2
groups by the severity of inammation. e gray-patch and
black dot type were classied to be a non-inammatory
group of infection. Owing to the severe inammation,
the kerion and favus were labeled as an inammatory
group. Gray-patch type predominantly demonstrated
scaling with lack of inammation that was consistent with
ectothrix TC. For this reason, it was appropriate to apply
with the toothbrush technique for collecting scale. More
than that this type is the most common presentation of
TC, therefore it is practicable to use in screening and in
outbreaks period. Moreover, this technique was simple
and painless therefore it may easy to apply in children.
To compare with other endothrix types including black
dot and kerion types, the toothbrush technique results in
unfavorable outcomes due to the inability to detect the
fungal penetration in the hair sha. Owing to the fact
that the hair-plucking and scraping scale by sharp blade
technique was more recommended to be performed for
endothrix TC investigation.
Other important factor in the toothbrush technique
was scalp area involvement. e previous study mentioned
that the greater surface area involves, the higher ecacy of
hairbrush technique signicantly increases.
10
However, the
experiment to set suitable conditions for the toothbrush
technique has been never established. As stated in the
results of the toothbrush technique in this study, more
than 15% scalp area involvement of TC was suitable
for practice. Accordingly, the ecacy of the toothbrush
technique depended on patient selection in the aspect of
clinical presentation and scalp area involvement. us
the patients with gray-patch presentation and more than
15% scalp area involvement were suggested to conrm TC
diagnosis by toothbrush technique, especially in outbreaks.
Furthermore, laboratory technicians mostly agreed with
the toothbrush technique was more appreciative and
comfortable in practice than the standard technique.
Time-saving on specimen collection and good patient’s
cooperation were reasonable to be the advantages of this
technique. Furthermore, time for fungal culture process
was not dierent between the toothbrush technique and
standard technique.
In conclusion, the toothbrush technique was
recommended to practice with “gray-patch” TC patients
who presented more than 15% scalp area involvement
due to the time-saving and well-cooperated method.
ACKNOWLEDGMENTS
We would like to express our deepest appreciation to
Assistant Professor Dr.Orawan Supapueng who provided
us the advice about statistical analysis to complete this
report. Moreover, we would also like to show our gratitude
to Mr. Chanai Muanprasat for data collection assistance
during this research. With respect to the study, all authors
report no conict of interest.
REFERENCES
1. Hay RJ. Tinea capitis: Current status. Mycopathologia 2017;182:
87-93.
2. Seebacher C, Bouchara JP, Mignon B. Updates on the epidemiology
of dermatophyte infections. Mycopathologia 2008;166:335-52.
3. Price H, Taylor DR. Trichophyton tonsurans (crateriforme)
infection of the scalp. Calif Med 1952;76:283-8.
4. Bunyaratavej S, Leeyaphan C, Rujitharanawong C, Muanprasat
C, Matthapan L. Clinical and Laboratory Characteristics of a
Tinea Capitis Outbreak Among Novice Buddhist Monks.
Pediatr Dermatol 2017;34:371-3.
5. Pomeranz AJ, Sabnis SS, McGrath GJ, Esterly NB. Asymptomatic
dermatophyte carriers in the households of children with tinea
capitis. Arch Pediatr Adolesc Med 1999;153:483-6.
6. Ilkit M, Demirhindi H. Asymptomatic dermatophyte scalp
carriage: laboratory diagnosis, epidemiology and management.
Mycopathologia 2008;165:61-71.
7. Ali S, Graham TA, Forgie SE. e assessment and management
of tinea capitis in children. Pediatr Emerg Care 2007;23:662-5.
8. Williams JV, Honig PJ, McGinley KJ, Leyden JJ. Semiquantitative
study of tinea capitis and the asymptomatic carrier state in
inner-city school children. Pediatr 1995;96:265-7.
9. omas WH, John MT. Brush-Culture Method for Diagnosing
Tinea Capitis. Pediatr 1992;90:416-8.
10. Akbaba M, Ilkit M, Sutoluk Z, Ates A, Zorba H. Comparison
of hairbrush, toothbrush and cotton swab methods for diagnosing
asymptomatic dermatophyte scalp carriage. J Eur Acad Dermatol
Venereol 2008;22:356-62.
11. Ilkit M, Demirhindi H. Asymptomatic dermatophyte scalp
carriage: laboratory diagnosis, epidemiology and management.
Mycopathologia 2008;165:61-71.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
168
Basel Mahardawi, B.D.S.*, Teeranut Chaisamut, D.D.S.*, Natthamet Wongsirichat, B.Sc., D.D.S.*
,
**
*Department of Oral Maxillofacial Surgery, Faculty of Dentistry, Mahidol University, Bangkok 10400, **Walailak University International College of
Dentistry, Bangkok 10400, ailand.
Gummy Smile: A Review of Etiology,
Manifestations, and Treatment
Corresponding author: Natthamet Wongsirichat
E-mail: natthamet.won@mahidol.ac.th
Received 25 August 2018 Revised 6 December 2018 Accepted 24 December 2018
ORCID ID: http://orcid.org/0000-0003-3005-2680
http://dx.doi.org/10.33192/Smj.2019.26
ABSTRACT
Excessive gingival display, or what is known as Gummy Smile, is a clinical condition which is attracting a great
attention and focus recently. Having knowledge and background about the cause and the treatment of each case of
excessive gingival display is essential for making the accurate diagnosis and performing the correct procedure. is
article reviews the most common causes of excessive gingival display, manifestations of this condition, and some
of the important treatment options for Gummy Smile. In addition, the article also puts the light on some points
regarding the accepted measurement in several studies, and considers the prediction planning for Gummy Smile
associated with a skeletal etiology.
Keywords: Excessive gingival display; gummy smile; vertical maxillary excess (Siriraj Med J 2019;71: 168-174)
INTRODUCTION
A wonderful smile is an essential feature of beauty to
which society gives an increasing importance currently.
e concept of a nice smile basically depends on the
entity of three anatomic components: gum, teeth, and
lips.
1
In the meantime, an increasing awareness regarding
beauty and physical appearance has become a motivation
for every clinician to evaluate the important aspects of
patient’s smile and link the dynamic relationship between
the teeth, gingiva, and lips together when smiling.
2
e gingiva is a crucial aspect to be considered in
the esthetics of smiling
1
, in which the upper lip should
be elevated around 80% of its normal length, exposing
teeth and gingiva.
3
Sharma et al.,
2
dened the normal
gingival display as the gum exposure between the inferior
border of upper lip and gingival margin of anterior central
incisors when smiling. An exposure of gingiva around 0-2
mm when smiling, and 2-4 mm of the maxillary incisor
edges when the lips are at the rest state are considered as
acceptable. While more than 2 mm of gingival exposure
when a person smiles is stated by experts as an excessive
gingival display, or what is known as “Gummy Smile”,
4
which is more of a descriptive term than a diagnosed
condition, and aects a notable proportion of the population,
especially women who are aected more than men.
5
Maxillary excess both vertically and anterioposteriorly
(Bimaxillary protrusion), hyper mobility of the upper
lip due to hyper function of labial elevator muscles, and
excessive gingival display associated with altered passive
eruption are the three major causes of this condition.
6
In addition to the stated reasons, medications, dental
plaque, hereditary and incompetent lip can also result
in abnormal gingival display.
5
For adults, improved aesthetic outcomes are becoming
increasingly essential in these procedures to the point
where some patients are pursuing only the esthetic side,
Mahardawi et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
169
Review Article
SMJ
not a functional benet.
7
erefore, nding an eective
treatment for each cause of the gummy smile is a must.
Etiologies of gummy smile:
Many studies have stated the main causes of excessive
gingival display, presenting the most important factors
which may lead to having a Gummy Smile. e study
of Roe et al.,
8
described that lip length and the upper
lip mobility rate are the main contributing factors. e
previous research of Peck et al.,
9
stated that the exposure
of teeth and gingiva depends on the integrated eects
of a number of variables (increased muscle capacity,
vertical maxillary excess, greater interlabial gap at resting
position, and the amount of overjet and overbite). Pausch
et al.,
4
mentioned that abnormal gingival and maxillary
anterior teeth display may take place due to numerous
anatomic or functional factors, either hereditary or
inborn. A narrow upper lip, an irregular eruption of teeth,
excessive protuberance or vertical maxillary growth, and
hypermobility of the maxillary lip and elevator muscle
are common reasons for a Gummy Smile.
In fact, several contributing factors are aecting
individuals to have a Gummy Smile. Sometimes one of
them is presented, although in some cases more than only
one cause can be seen. Correct diagnosis of the reason
leads to a proper treatment plan. e most common and
discussed factors associated with Gummy Smile are:
Altered passive eruption:
Also known as impaired passive eruption,
1
is dened
as a condition in which the relationship between teeth,
alveolar bone in the maxilla, and the so tissues displays
an excessive gingiva. is, in turn, reveals the clinical
characteristic of Gummy Smile.
10
In other words, altered
passive eruption (APE) is characterized by excessive
gingival exposure in relation to the crowns of the maxillary
teeth.
11
In this case, the gingiva fails to migrate in the
apical direction during the eruption of teeth, thus,
it remains in a coronal position in relation with the
cementoenamel junction (CEJ), which results in having
an unacceptable gingival exposure and unfavorable small
size of the teeth when smiling.
5
Miskinyar et al.,
12
found
that the prevalence of this condition is about 12% of the
population. Altered passive eruption was rst identied
by the study of Coslet et al.,
13
and according to Rossi et al.,
10
it was classied into two types and two subtypes (Fig 1):
1. Type I : the vertical length of keratinized gingiva
is greater than normal, the mucogingival junction
(MGJ) is located in an apical position to the level of the
cementoenamel junction (CEJ), and clinical crowns
appear shorter.
2. Type II : the vertical dimension of the keratinized
gingiva is normal, although the mucogingival junction
is positioned at the level of the CEJ.
2.1 Subtype A : the measurement between the
maxillary alveolar crest and the CEJ is around 1.5 mm,
and in this case a regular attachment can be found.
2.2 Subtype B : the level of maxillary alveolar
crest is at the level of the CEJ, or occlusal to the CEJ in
some cases.
Bony maxillary excess:
This includes vertical maxillary excess and
bimaxillary protrusion.
Vertical maxillary excess:
Wolford et al.,
15
defined Maxillary vertical
hyperplasia or vertical maxillary excess as an excessive
vertical growth of the maxilla which may or may not lead
to an anterior open bite. In his study Robbins
16
explained
that to make the correct diagnosis, the face must be
divided into three equal thirds or parts for the critical
evaluation of the height of the face. Vertical maxillary
excess is noticed when the length of the lower third of
the face is more than the other two thirds, causing an
excessive gingival display.
5
In the analysis of the face, vertical maxillary excess
has the following features: longer lower third of the face,
in relation with the upper and middle thirds, a greater
display of maxillary incisors at rest, an incompetent
lip, inclination towards class II malocclusion with or
without open bite, and a noticeable Gummy Smile. e
nose is longer as well, the alar bases are small and the
zygoma appears to be generally at. e lower third of
the face is long, leading to a retrognathic shape of the
jaw.
17
Furthermore, the incisal edge of the upper anterior
teeth might be covered by the lower lip because of the
extravagant downward growth of the maxilla.
16
Peck et
al.,
9
and Mackley
18
found that a Gummy Smile is highly
related to anterior vertical maxillary excess (about 2-3
mm). Moreover, Ezquerra and Berrazueta
19
discuss
that excessive maxilla associated with the protrusion of
anterior alveolar bone consequently produce a Gummy
Fig 1. Classication of altered passive eruption (APE)
14
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
170
Smile. Whilst Wu et al.,
6
in their study also stated that
anterior maxillary height (upper incisor to the palatal
plane) was considerably more in both male (+1.03 mm)
and female (+2.13 mm) groups who have Gummy Smile.
Bimaxillary protrusion:
Bimaxillary protrusion is a frequently diagnosed
deformity in African Americans
20
and Asian populations as
well. is deformity is described regarding the protrusive
and buccally positioned maxillary and mandibular incisors,
as well as the enlarged prostration of the lips. It can be
found with an incompetent lip, excessive gingival display,
mentalis strain with an anterior open bite. Bimaxillary
protrusion refers to an abnormally protruded maxillary
and mandibular dentoalveolus. Generally, this presents
with malocclusion and dentoalveolar aring of both the
maxillary and mandibular anterior teeth, which cause
the lips to be protruded, thus, producing an additional
convexity of the facial prole. Bimaxillary protrusion is
mainly accompanied by several degrees of lip deciency
(dened as more than 4 mm of lip detachment at the rest
state). e tendency of the anteroposterior correlation is
to be a class II malocclusion with a decient mandible,
although it may also range from severe class II to class
III.
21
The etiology behind bimaxillary protrusion is
associated with various reasons between genetics and
environment, such as mouth breathing, lip biting habit
and large size of the tongue.
22
Keating et al.,
23
discovered that in Caucasians with
bimaxillary protrusion, it is likely to notice a posterior
cranial base which is shorter than normal, a prognathic
maxilla with vertical excess, mild class II skeletal
occlusion, and a prostrate prole.
Conditions causing gingival enlargement
Gingival enlargement might be the outcome of
bacterial plaque-enhanced chronic inammation and
medications.
5
Narwal et al.,
24
showed in their case report
an abnormal unilateral palatalso tissue enlargementin a
61-year-old hypertensive female using amlodipine, which
is a safe antihypertensive drug. Ritchhart et al.,
25
stated a
process behind uncontrolled cell proliferation in drug-
induced gingival overgrowth which is the suppression
of apoptotic pathways, with the following eect on the
normal gingival formation by cell turnover. Hormonal
dierences which take place in pregnancy and puberty,
in addition to the use of oral contraceptives have been
associated with gingival overgrowth.
5
Other eects such
as age, demographic and pharmacokinetic changes,
genetic predisposition, oral hygiene condition, as well as
molecular and cellular variables may impact the mechanism
of gingival enlargement.
25, 26
Orthodontic treatment
using orthodontic appliances can also participate in
the presence of gingival enlargement.
27
In the case of
leukemia, leukemic cells might inltrate to the gingiva,
28
and this will cause some manifestations to be present
such as gingival enlargement and bleeding.
29
Short upper lip
e upper lip length is dened as the length from
the base of the nose (subnasale point) to the inferior
part of the upper lip (upper lip stomion point),
8
which
is normally around 23 mm in males and 20 mm in
females.
3
If an individual with excessive gingival display
has a short upper lip as well, the esthetic compromise
will be aected.
16
e diagnosis of an excessive gingival
display can be done regarding the clinical measurement
of the upper lip length with an excessive amount of
tooth display at rest.
5
Conversely, Roe et al.,
8
found in
their study that in maximum smile, there is no notable
dierence in the gingival display between an individual
with a short or normal upper lip, and they concluded that
the association of some factors such as higher muscle
capacity, vertical maxillary excess, excessive interlabial
gap at rest, and the amount of overjet and overbite have
a greater eect on the gingival display than the upper
lip length itself. Moreover, Sarver et al.,
30
explained that
what is crucial is the correspondence between the upper
lip length, the maxilliary incisors, and the commissure
of the mouth. In other words, the lip length should be
almost equal to the commissure of the mouth.
Hypermobility of the upper lip:
In the case of normal face height, gingival levels,
lip length and length of central incisors in a patient who
has an excessive gingival display, the possible etiology
is hypermobility of the maxillary upper lip.
16
Upper lip
mobility was explained as the volume of lip movement that
exists when an individual smiles. is was measured by
deducting the incisal shown at rest from the dentogingival
display during a full smile.
Hypermobility of the upper
lip is associated with a hyper function of the lip elevator
muscles and basically leads to excessive gingival display.
19
Peck and Peck
31
reported an average lip movement
of 5.2 mm (23% decrease) from a measured lip length
of 22.3 mm during a full smile. Furthermore, in a study
evaluating spontaneous smiles, Tarantili et al.,
32
identied
a 28% decrease in the initial upper lip length, while
Roe et al.,
8
concluded the overall means of lip mobility
for females with normal lip length, females with short
lip length, males with normal lip length and males with
Mahardawi et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
171
Review Article
SMJ
short lip length were 5.8 ± 1.7 mm, 5.0 ± 1.3 mm, 6.7 ±
1.5 mm, and 5.7 ± 1.1 mm, respectively. Moreover, Sabri
et al.,
3
stated that in smiling, the upper lip is elevated by
about 80% of its original length. Bhola et al.
5
, as well as
Robbins
16
stated that the upper lip is generally elevated
around 6-8 mm from the rest position to the position
reached when a full smile takes place.
Treatment of gummy smile:
Excessive gingival display can be managed by a
variety of procedures. ese procedures include non-
surgical and surgical methods. e underlying cause of
excessive gingival display or Gummy Smile has the main
eect on the type of procedure that will be performed.
33
Non-surgical procedures may include Botulinum toxin
type A injection as well as orthodontics, While surgical
procedures might include lip repositioning or orthognathic
surgery following orthodontics.
Robbins
16
dened the healthy gingiva as having
a minimal length of approximately 3 mm from the
gingival crest to the alveolar crest. In the case of altered
passive eruption, a higher amount will exist, and the
treatment of choice is normally gingivectomy following
by aesthetic crown lengthening to attain the desired
dimension and morphology of teeth. However, when the
diagnosis shows bone levels close to the CEJ, a gingival
ap with ostectomy is performed, or what is also known
as an apically positioned ap.
34
Orthodontic eruption
or intrusion can also be done when having gingival
asymmetry on one or multiple anterior teeth
34
In the case of vertical maxillary excess, the only
treatment option to consider is orthognathic surgery.
11
is
is applied to impact the maxilla considering the amount of
gingival exposure diagnosed. Maxillary impaction allows
correction of the Gummy Smile, long face syndrome,
specic types of open bite from a skeletal origin and labial
sealing.
35
e Le Fort I osteotomy of the maxilla allows
reduction of bone between the nasal oor and apices of
maxillary teeth which permits superior repositioning
(impaction) of the maxilla.
17
When having bimaxillary protrusion, the treatment
option may be composed of rst premolar extractions
followed by the osteotomy through the extraction sites
to mobilize the anterior segment of the maxilla. e aim
is to setback the segment in addition to lesseningthe
labial aring of the incisors. Maxillary setback alone
can sometimes provide a substitute for the treatment of
anterioposterior maxillary excess without any need for
tooth extraction or segmental osteotomy in these cases:
(1) it is contraindicated to perform extractions in
an orthodontic diagnosis (no dental crowning, adequate
curve of Spee, etc.),
(2)the proclination of maxillary incisors can be
adjusted to an acceptable position with a Le Fort I osteotomy
in addition to clockwise rotation.
21
In the presence of external factors causing gingival
overgrowth, the treatment plan should be focused on the
exact cause of the enlarged gingiva. Meticulous history
taking, in addition to an excisional/incisional biopsy and/
or hematologic/histologic inspection might be performed
generally to make the correct diagnosis of the uncommon
conditions of gingival enlargement. Plaque control is an
important aspect of treatment in all the patients. Some
of these cases may resolve when the external element
that is inducing the gingival enlargement is adjusted or
suspended, For example, gingival overgrowth during
pregnancy and puberty might need the elimination of
all local irritants followed by surgical treatment for the
removal of any brotic residuals.
36
For the excessive gingival display resulted from
short upper lip and/or hypermobility of the upper
lip, a surgical procedure known as lip repositioning
surgery is preferred for a minimally invasive surgery.
It is composed of an oval mucosal excision followed
by coronally advanced ap. is procedure is done to
reduce the hyperactivity of the elevator muscles and
reform the depth of the vestibule.
37
It was rst described
by Rubinstein and Kostianovsky
38
and the aim was the
treatment of excessive gingival display associated with
hypermobility of the lip, and was then modied by Litton
and Fournier
39
to include also the treatment of Gummy
Smile caused by short upper lip, by separating the muscles
from the underlying bony structures to place the upper
lip in a coronal position. is surgical procedure was
accompanied with no complications, although there were
some incidences of relapse,
40
consequently, some attempts
were done to improve it by Miskinyar.
12
Recently, Bhola
et al.,
5
described a technique similar to the one which was
described by the article of Rubinstein and Kostianovsky
under the name of Lip Stabilization Technique (LipStaT).
e only obvious dierence between the two techniques
is that in LipStaT, a vertical incision is done posteriorly
to connect the inferior incision (at the mucogingival
junction) and the superior incision (into the vestibule).
e ratio of this incision is the height being double of
gingival exposure during a full smile. While in Rubinstein
and Kostianovsky technique, the two incisions were
approximated till they meet posteriorly.
Currently, another nonsurgical treatment option is
the injection of Botulinum toxin type A which has been
suggested for treatment of hypermobility of the upper
lip, but this may only provide temporary advantages.
41
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
172
Polo
41
reported a relapse of around 2 mm aer 24
weeks of following up. is was similar for Indra et al.,
42
by
reporting a relapse of treatment in the third month. Yet it is
still a viable option for those who do not prefer undergoing
surgeries, and need only a minimal treatment procedure.
DISCUSSION
It was stated recently that the dimension and visibility
of teeth, and upper lip position were crucial eects when
judging smile attractiveness.
43
Many studies described the
acceptable gingival exposure and the concept of Gummy
Smile (table below). Having an acceptable smile in the
eyes of other people is, in fact, a subjective view, because
the perception of a nice smile has diered from country
to country, from study to study, and even when males
or females are evaluating the smile. In other words, a
study showed that the male raters diered in respect of
gender specicity when judging the attractiveness of
the smile, while female raters rated males less gender
specic.
4
erefore, no reliable or unied measurements
were stated in the literature regarding a clear description
of having an attractive smile or when it is diagnosed as
a Gummy Smile. Although the dierent visions of an
accepted gingival exposure were evaluated in dierent
countries, they should be identical for the same ethnic
origins. Sabri et al
3
concluded that an ideal smile has
the features of an upper lip that will reach the gingival
edges when smiling, in addition to an upward or straight
curvature through the philtrum and commissures; with
maxillary incisal margins parallel to the lower lip line;
with no lateral negative space or minimal if it presents;
and a commissural line, as well as an occlusal frontal
plane parallel to the line drawn through the pupils of the
eyes; and attractive integration of gingiva and teeth. e
acceptable amount of gingival exposure when smiling
was mentioned in several studies. For instance, Kokich
et al.,
44
reported noticeably unattractive lip to gingiva
distance when the exposure is 4 mm by laypersons and
when it is 2 mm by orthodontists, while Geron and Atalia
45
accepted only 1 mm of gingival display when smiling.
Pausch et al.,
4
concluded in their study that a 0- to 2-mm
exposure of the maxillary gingiva was acceptable when
an individual smiles. In addition to that, they found that
although the Gummy Smile is unattractive, it is much
better than invisible or minimally visible maxillary teeth
when smiling. Bhola et al.,
5
explained that aected by
the increased aesthetic concerns, a gingival exposure of
more than 1 mm during smiling might be diagnosed as
excessive gingival display (EGD). Van der Geld et al.
43
concluded in their study that a lip line height that exposed
more than 4 mm of gingival display was considered as
a Gummy Smile line. Jannani et al.
46
stated that a smile
is introduced as attractive when the maxillary teeth are
entirely shown in addition to approximately 1 mm of
gingival display. However, a gingival display not more
than 2-3 mm is also contemplated as acceptable, while
a disproportionate exposure (>3 mm) is basically stated
as not pleasant. Pinto et al.,
11
mentioned three kinds of
a smile: high, medium and low, and the high smile is
considered acceptable when accompanied with exposed
gingiva of 1 to 3 mm.
Moreover, treatment of Gummy Smile was being
practiced many years ago. Several techniques and procedures
were developed to reduce the excessive gingival display
and change it into an aesthetic, acceptable appearance.
Taking the bony maxillary excess into consideration,
a review of the literature may conclude that although
the treatment of Gummy Smile caused by vertical
maxillary excess and/or bimaxillary protrusion is not a
new procedure, no studies have made a clear treatment
plan, and no one has described a step by step way to
predict the treatment according to a certain prediction
plan before performing orthognathic surgery, based
on clear measurements of moving the maxilla. us, a
thorough prediction and movement plan of the maxilla
for Gummy Smile treatment might be of a great value to
help surgeons establish an accurate treatment plan based
on scientic data, not only experience. e acceptable
gingival exposure and the diagnosis of Gummy Smile
in some previous studies is shown in Table 1.
4,5,11,43-46
CONCLUSION
Gummy Smile is a condition in which an increasing
awareness has been noticed recently. erefore, it is
attracting more attention from all the clinicians to nd
the best solutions with the least complications and
relapse, as well as the most satisfying results for every
patient. Additional unied standards of accepting and
not accepting the amount of gingival exposure might
be better to help with diagnosing and deciding whether
to perform surgery or consider the gingival exposure as
acceptable. is should be decided considering the ethnic
origin and trying to collect a wider range of opinions, not
only taking experts’ view because they will judge more
critically. is could be done by performing a greater
statistical analysis regarding this topic, including more
people to evaluate and judge.
In addition, clear prediction planning of moving
the maxilla to improve the condition of Gummy Smile,
taking into account if it is resulted from vertical maxillary
excess alone or with bimaxillary protrusion, would be of
a great benet to perform a well-planned surgery. is
Mahardawi et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
173
Review Article
SMJ
TABLE 1. Acceptable gingival exposure and the diagnosing of Gummy Smile in some studies in the literature.
Authors Year Acceptable gingival exposure Having a Gummy Smile
Kokich et al
44
1999 Less than 4 mm by laypersons 4 mm or more by laypersons 2 mm or
Less than 2 mm by orthodontists more by orthodontists
Geron and Atalia
45
2005 Untill 1 mm More than 1 mm
Van der Geld et al.
43
2011 Untill 4 mm More than 4 mm
Jannani et al.
46
2014 2-3 mm More than 3 mm
Pinto et al.
11
2015 1-3 mm More than 3 mm
Bhola et al.
5
2015 Untill 1 mm More than 1 mm
Pausch et al.
4
2017 0-2 mm More than 2 mm
surgery should be done according to certain principles and
criteria, not depending upon each surgeon’s experience
or preference. To reach this stratied surgical plan, this
literature review suggests further studies to design a
prediction formula of moving the maxilla, following
certain calculations regarding the amount of vertical
maxillary excess, bimaxillary protrusion, x-rays, and
considering opinions of experienced surgeons to get
the best aesthetic results.
ACKNOWLEDGMENTS
e authors would like to acknowledge the sta,
dental assistants, and our colleagues in the Department
of Oral and Maxillofacial Surgery, Faculty of Dentistry,
Mahidol University.
Funding: e authors received no funding
Conict of interest: e authors declared that they have
no conicts of interest.
Ethics approval: Not required
REFERENCES
1. Izraelewicz-Djebali E, Chabre C. Gummy smile: orthodontic or
surgical treatment? Dentofacial Anom Orthod 2015;18:102.
Published online: 04 September 2015.
2. Sharma A, Sharma S, Garg H, Singhal V, Mishra P. Lip
repositioning: A boon in smile enhancement. J Cutan Aesthet
Surg 2017;10:219-22.
3. Sabri R. e eight components of a balanced smile. J Clin
Orthod 2005;39:155-67.
4. Pausch NC, Katsoulis D. Gender-specific evaluation of
variation of maxillary exposure when smiling. J Craniomaxillofac
Surg 2017;45:913-20.
5. Bhola M, Fairbairn PJ, Kolhatkar S, Chu SJ, Morris T, de
Campos M. LipStaT: e Lip Stabilization Technique- Indications
and Guidelines for Case Selection and Classication of Excessive
Gingival Display. Int J Periodontics Restorative Dent 2015;35:
549-59.
6. Wu H, Lin J, Zhou L, Bai D. Classication and craniofacial
features of gummy smile in adolescents. J Craniofac Surg
2010;21:1474-9.
7. Charrier JB. Orthognathic surgery of adults and facial aesthetics.
J Dentofacial Anom Orthod 2012;15:302.
8. Roe P, Rungcharassaeng K, Kan JYK, Patel RD, Campagni WV,
Brudvik JS. e Inuence of Upper Lip Length and Lip Mobility
on Maxillary Incisal Exposure. Am J Esthet Dent 2012;2:116-
125.
9. Peck S, Peck L, Kataja M. e gingival smile line. Angle Orthod
1992;62:91-100.
10. Rossi R, Brunelli G, Piras V, Pilloni A. Altered passive eruption and
familial trait: a preliminary investigation. Int J Dent 2014;
2014:874092.
11. Pinto SCS, Higashi C, Bonafé E, Pilatti GL, Santos FA, Tonetto
MR, et al. Crown Lengthening as Treatment for Altered Passive
Eruption: Review and Case Report. World J Dent 2015;6:178-
83.
12. Miskinyar SA. A new method for correcting a gummy smile.
Plast Reconstr Surg 1983;72:397-400.
13. Coslet JG, Vanarsdall R, Weisgold A. Diagnosis and classication
of delayed passive eruption of the dentogingival junction in
the adult. Alpha Omegan 1977;7:24-8.
14. Alpiste-Illueca F. Altered passive eruption (APE): a little-known
clinical situation. Med Oral Patol Oral Cir Bucal 2011;16:
e100-4.
15. Wolford LM, Karras SC, Mehra P.Considerations for orthognathic
surgery during growth, part 2: maxillary deformities. Am J
Orthod Dentofacial Orthop 2001;119:102-5.
16. Robbins JW.Dierential diagnosis and treatment of excess
gingival display. Pract Periodontics Aesthet Dent 1999;11:265-
72.
17. Nascimento Meger M, Tiboni F, dos Santos FS, Verbicaro
T, Deliberador TM, Scariot R, et al. Surgical correction of
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
174
vertical maxillary excess associated with mandibular self-
rotation. RSBO 201714:56-61.
18. Mackley RJ. An evaluation of smiles before and aer orthodontic
treatment. Angle Orthod 1993;63:183-90.
19. Ezquerra F, Berrazueta MJ, Ruiz-Capillas A, Arregui JS. New
approach to the gummy smile. Plast Reconstr Surg 1999;104:1143-
50.
20. Scott SH, Johnston LE. e perceived impact of extraction. and
nonextraction treatments on matched samples of. African
American patients. Am J Orthod Dentofacial Orthop 1999;116:
352-8.
21. Chu YM, Bergeron L, Chen YR. Bimaxillary protrusion: an
overview of the surgical-orthodontic treatment. Semin Plast
Surg 2009;23:32-9.
22. Lamberton CM, Reichart PA, Triratananimit P. Bimaxillary
protrusion as a pathologic problem in the ai. Am J Orthod
1980;77:320-9.
23. Keating PJ. Bimaxillary protrusion in the Caucasian: a cephalometric
study of the morphological features. Br J Orthod 1985;12:193-
201.
24. Narwal A, Singh V, Bala S. Drug-induced atypical hyperplasia
enveloping salivary gland malignancy. J Indian Soc Periodontol
2017;21:409-11.
25. Ritchhart C, Joy A. Reversal of drug-induced gingival overgrowth
by UV-mediated apoptosis of gingival broblasts - an in vitro
study. Ann Anat 2018;217:7-11.
26. Nakib N, Ashra SS. Drug-induced gingival overgrowth. Dis
Mon 2011;57:225-30.
27. Trossello VK, Gianelly AA. Orthodontic treatment and periodontal
status. J Periodontol 1979;50:665-71.
28. Savona M, Talpaz M. Chronic myeloid leukemia: changing the
treatment paradigms. Oncology (Williston Park) 2006;20:
707-11.
29. Chowdhri K, Tandon S, Lamba AK, Faraz F. Leukemic gingival
enlargement: A case report and review of literature. J Oral
Maxillofac Pathol 2018;22(Suppl 1):S77-S81.
30. Sarver DM, Ackerman MB.Dynamic smile visualization and
quantication: Part 2. Smile analysis and treatment strategies.
Am J Orthod Dentofacial Orthop 2003;124:116-27.
31. McNamara JA. Jr, Brust EW, Riolo ML. So tissue evaluation
of individuals with an ideal occlusion and a well-balanced
face. In: McNamara JA Jr. (Ed.) Esthetics and the Treatment
of Facial Form. Monograph 28. Craniofacial Growth Series.
Center for Human Growth and Development, Ann Arbor;
1993.
32. Tarantili VV, Halazonetis DJ, Spyropoulos MN.e spontaneous
smile in dynamic motion. Am J Orthod Dentofacial Orthop
2005;128:8-15.
33. Abdullah WA, Khalil HS, Alhindi MM, Marzook H. Modifying
gummy smile: a minimally invasive approach. J Contemp Dent
Pract 2014;15:821-6.
34. Dolt AH 3rd, Robbins JW. Altered passive eruption: an etiology
of short clinical crowns. Quintessence Int 1997;28:363-72.
35. Steinhäuser S, Richter U, Richter F, Bill J, Rudzki-Janson I.
Prole changes following maxillary impaction and autorotation
of the mandible. J Orofac Orthop 2008;69:31-41.
36. Agrawal AA. Gingival enlargements: Dierential diagnosis
and review of literature. World J Clin Cases 2015;3:779-88.
37. Humayun N, Kolhatkar S, Souiyas J, Bhola M. Mucosal coronally
positioned ap for the management of excessive gingival
display in the presence of hypermobility of the upper lip and
vertical maxillary excess: a case report. J Periodontol 2010;81:
1858-63.
38. Rubinstein AM, Kostianovsky AS. Cosmetic surgery for the
malformation of the laugh: Original technique in Spanish.
Prensa Med Argent 1973;60:952.
39. Litton C, Fournier P. Simple surgical correction of the gummy
smile. Plast Reconstr Surg 1979;63:372-3.
40. Grover HS, Gupta A, Luthra S. Lip repositioning surgery:
A pioneering technique for perio-esthetics. Contemp Clin
Dent 2014;5:142-45.
41. Polo M. Botulinum toxin type A (Botox) for the neuromuscular
correction of excessive gingival display on smiling (gummy
smile). Am J Orthod Dentofacial Orthop 2008;133:195-203.
42. Aly LA, Hammouda NI. Botox as an adjunct to lip repositioning
for the management of excessive gingival display in the presence
of hypermobility of upper lip and vertical maxillary excess.
Dent Res J (Isfahan) 2016;13:478-83.
43. Van der Geld P, Oosterveld P, Schols J, Kuijpers-Jagtman AM.
Smile line assessment comparing quantitative measurement and
visual estimation. Am J Orthod Dentofacial Orthop 2011;139:
174-80.
44. Kokich Jr VO, Kiyak HA, Shapiro PA. Comparing the Perception
of Dentists and Lay People to Altered Dental Esthetics. J Esthetic
Dentistry 1999;11:311-24.
45. Geron S, Atalia W. Inuence of sex on the perception of oral
and smile esthetics with dierent gingival display and incisal
plane inclination. Angle Orthod 2005;75:778-84.
46. Jananni M, Sivaramakrishnan M, Libby TJ. Surgical correction
of excessive gingival display in class I vertical maxillary excess:
Mucosal strip technique. J Nat Sci Biol Med 2014;5:494-8.
Mahardawi et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
95
Original Article
SMJ
Anan Bedavanija, M.D.*, Pongsakorn Tantilipikorn, M.D, Ph.D.*, Chetta Banditsing, M.D.**, Pattarachai
Kiratisin, M.D, Ph.D.***, Chaweewan Bunnag, M.D.*
*Department of Otorhinolaryngology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, **Department of Otorhinolaryngology,
Bangkok Hospital Rayong, Rayong 21000, ***Department of Microbiology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700,
ailand.
A Comparison of the Bacterial Culture Results of
Maxillary Sinus Mucosa and Pus Collections for
Chronic Maxillary Rhinosinusitis
Corresponding author: Pongsakorn Tantilipikorn
E-mail: ptantili@yahoo.com
Received 12 December 2018 Revised 5 February 2019 Accepted 12 February 2019
ORCID ID: http://orcid.org/0000-0003-1995-4798
http://dx.doi.org/10.33192/Smj.2019.15
ABSTRACT
Objective: Although maxillary antral taps are the standard for collecting pus for bacterial culture, they sometimes
reveal no growth. Intraoperative mucosal cultures are another method to collect pathogen samples. is study
compared aerobic bacterial cultures from maxillary sinus mucosa and pus collected from chronic maxillary
rhinosinusitis patients.
Methods: A prospective study of 22 chronic maxillary rhinosinusitis patients was conducted. Antral pus and mucosa
collected during endoscopic sinus surgery were immediately sent to a microbiological laboratory. e degree of
concordance between maxillary sinus mucosa aerobic bacterial cultures and pus cultures was then analyzed.
Results: Twenty-seven specimens were obtained for the cultures. e proportions of positive mucosal and pus
cultures were 40.74% and 51.85%, respectively. e common aerobic pathogens from the two culture techniques
were Pseudomonas aeruginosa and Staphylococcus aureus. A concordance between the pus and mucosal cultures
was demonstrated by 19 out of 27 specimens (70.37%). Compared with the pus cultures, the mucosal cultures had
a specicity of 84.62% (95% CI, 54.55%-98.08%), a sensitivity of 57.14% (95% CI, 28.86%-82.34%), a predictive
value of a positive result of 80% (95% CI, 50.83%-93.93%), and a predictive value of a negative result of 64.71%
(95% CI, 48.96%-77.80%).
Conclusion: Similar pathogenic bacteria were recovered from the mucosa and pus. Given the high degree of
similarity of the bacteria found, the good concordance rate, and the high specicity and positive predictive value
of the mucosal cultures compared with the pus cultures, mucosal cultures should be a reference standard and an
option when pus is unavailable, especially with immunocompromised patients.
Keywords: Bacterial culture; concordance; chronic rhinosinusitis; maxillary sinus mucosa; pus (Siriraj Med J
2019;71: 95-101)
INTRODUCTION
Rhinosinusitis is a common disease in ailand. In
2017, the incidence among otolaryngologic outpatients at
Siriraj Hospital was about 5%. Obtaining a culture might be
useful with patients who have not responded to conventional
medical treatment, with immunocompromised patients,
and in cases of rhinosinusitis with orbital and intracranial
complications. Pus collection from the maxillary sinus
by an antral puncture or open sinus surgery is usually
used in clinical practice.
1-6
Although a maxillary antral
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
96
tap is the standard method of obtaining a sinus culture,
it is invasive and dicult to perform,
1–6
and sometimes
reveals no growth. In addition, otolaryngologists cannot
always get pus from a maxillary antral tap because of
the small amount of pus, or the absence of pus, in the
maxillary antrum. In 1979, Karma et al.
7
concluded that
an intraoperative mucosal culture is a better method to
get true pathogens than collecting pus from the maxillary
antrum; that conclusion has since been supported by the
ndings of other studies.
8-9
e objectives of the present
study were to compare aerobic bacterial cultures obtained
from the sinus mucosa with those from pus from the
antrum of chronic maxillary rhinosinusitis patients.
MATERIALS AND METHODS
A prospective study was conducted at Siriraj Hospital
between May 2010 and January 2011. All patients, which
considered for inclusion, were more than 18 years old
and diagnosed with chronic maxillary rhinosinusitis with
medical treatment failure. Before undergoing endoscopic
sinus surgery, they had to stop taking any antibiotics for
at least one week. e diagnosis of chronic rhinosinusitis
was based on the guidelines of the European Position
Paper on Rhinosinusitis and Nasal Polyps,
10
i.e., an
inammation of the nose and the paranasal sinuses,
characterized by two or more symptoms of nasal obstruction,
discolored nasal discharge or postnasal drip, frontal
pain/headache, and smell disturbance (one of which
should be nasal obstruction or discolored discharge)
for more than 12 weeks; and either a positive nding on
a nasal endoscopy (nasal polyps and/or mucopurulent
discharge from the middle meatus, and/or edema/mucosal
obstruction primarily in the middle meatus); and/or
mucosal changes within the ostiomeatal complex and/
or sinuses on preoperative computed tomography.
10
e study protocol was approved by the Institutional
Review Board Committee of the Faculty of Medicine
Siriraj Hospital, and informed consent was obtained
from every patient. During the endoscopic sinus surgery,
immediately aer opening the maxillary sinus ostium,
antral pus was collected in a sterile bottle by curved
suction; 1-2 small pieces of maxillary sinus mucosa was
also randomized collected in a sterile bottle using cutting
forceps. All specimens were sent to the microbiological
laboratory within 10 minutes and immediately incubated
on chocolate, blood, and MacConkey agar plates. Bacteria
that grew on the culture were reported as semiquantitative
identication (rare, few, moderate, and numerous growth).
e results of both cultures were compared. Samples
were considered to have a strong concordance if the
same bacterial organisms were recovered, or if there was
no growth and/or commensal ora on both the mucosal
and pus cultures. Samples were considered to have a
moderate concordance if the predominant organisms
recovered from the mucosa were also recovered from
the pus, but an additional organism was recovered in
small quantities by one of the two methods. Samples were
considered to have no concordance if the predominant
organisms recovered by the two techniques diered.
11-13
Statistical analysis
e data were presented as numbers and percentages.
e dierence in the culture rates for the maxillary sinus
mucosa and pus was calculated with McNemar’s Chi-
square test. Comparisons were made of the results of
the bacterial cultures of the maxillary sinus mucosa and
pus collections from the chronic maxillary rhinosinusitis
patients. e sensitivity, specicity, and positive and
negative predictive values of the mucosal cultures were
calculated and compared with those of the pus cultures,
as the reference standard.
RESULTS
e ages of the 22 adult patients included in this
study ranged from 26 to 83 years, with a mean of 53.36
years. ere were 11 males and 11 females. Five patients
had a bilateral maxillary sinus operation, resulting in 27
paired samples. e positive rates of the mucosal and
pus cultures for aerobic pathogenic bacteria represented
40.74% and 51.85% of samples, respectively, as shown
in Table 1. e dierence in the culture rates for the
maxillary sinus mucosa and pus samples was statistically
insignicant (p = 0.4531, 95% CI: -0.292204 to 0.084671).
e aerobic bacterial growths common to both groups
were Pseudomonas aeruginosa and Staphylococcus
aureus (Methicillin-susceptible), as shown in Table 2.
e maxillary sinus mucosal cultures were strongly
concordant with the cultures from pus from the maxillary
sinus in 19 out of the 27 aerobic samples (70.37%;
Table 1). Two of the 27 samples were considered to have a
moderate concordance (7.41%); the additional organism
was noted in small quantities in those 2 samples. e
mucosal cultures provided a specicity of 84.62% (95%
CI, 54.55%–98.08%), a sensitivity of 57.14% (95% CI,
28.86%–82.34%), a predictive value of a positive result
of 80% (95% CI, 50.83%–93.93%), a predictive value of
a negative result of 64.71% (95% CI, 48.96%–77.80%), a
positive likelihood ratio of 3.71 (95% CI, 0.96–14.37), and
a negative likelihood ratio of 0.51 (95% CI, 0.27–0.97)
compared with the pus cultures from the maxillary
sinus (Table 3).
Bedavanija et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
97
Original Article
SMJ
TABLE 1. Types of aerobic bacteria isolated from pus and mucosal cultures from the same chronic maxillary
rhinosinusitis patients.
No. Sex Age Diagnosis Pus culture Mucosal culture Concordance
1 Male 64 CRSwNP -MSSA (moderate) -MSSA (few) +
-Morganella morganii (few)
2 Male 31 CRSwNP -P. aeruginosa strain 1 (moderate) -K. pneumoniae ESBL-negative (few) -
(Rt.) -P. aeruginosa strain 2 (few)
3 Male 31 CRSwNP -P. aeruginosa (numerous) -P. aeruginosa (few) +
(Lt.) -K. pneumoniae (rare)
4 Female 62 CRSsNP -Non-fermentative GNR (few) -NG -
5 Female 43 CRSsNP -NG -NG ++
6 Female 57 CRSsNP -P. aeruginosa (moderate) -NG -
7 Male 54 CRSwNP -K. pneumoniae ESBL-negative (few) -NG -
(Rt.)
8 Male 54 CRSwNP -E. coli ESBL-negative (few) -E. coli ESBL-negative (few) ++
(Lt.)
9 Female 75 CRSsNP -MSSA (moderate) -MSSA (few) ++
10 Female 83 CRSsNP -P. aeruginosa (few) -NG -
-Commensalora(few)
11 Male 62 CRSwNP -NG -Commensalora(few) ++
(Rt.)
12 Male 62 CRSwNP -NG -NG ++
(Lt.)
13 Female 45 CRSsNP -P. aeruginosa (moderate) -P. aeruginosa (few)
-Non-fermentative GNR (moderate) -Non-fermentative GNR (few) ++
14 Female 70 CRSsNP -P. aeruginosa (numerous) -P. aeruginosa (rare) ++
15 Female 74 CRSsNP -P. aeruginosa (moderate) -P. aeruginosa (few) ++
16 Male 26 CRSwNP -NG -Commensalora(rare) ++
(Rt.)
17 Male 26 CRSwNP -NG -NG ++
(Lt.)
18 Female 58 CRSsNP -NG -NG ++
19 Male 27 CRSsNP -Enterobacter aerogenes (rare) -NG -
(Rt.)
20 Male 27 CRSsNP -NG -Enterobacter aerogenes (few) -
(Lt.)
21 Male 61 CRSsNP -NG -NG ++
22 Male 46 CRSsNP -Pasteurella multocida -Pasteurella multocida ++
-MSSA (few) -MSSA (few)
23 Male 49 CRSsNP -NG -NG ++
24 Male 32 CRSsNP -Commensalora(few) -NG ++
25 Female 80 CRSsNP -NG -S. agalactiae (few) -
-MSSA (few)
26 Male 49 CRSsNP -NG -NG ++
27 Female 26 CRSsNP -NG -NG ++
N.B.: CRSwNP = chronic rhinosinusitis with nasal polyps; CRSsNP = chronic rhinosinusitis without nasal polyps; NG = no growth;
MSSA = Methicillin-susceptible Staphylococcus aureus; GNR = Gram-negative rods; P.aeruginosa = Pseudomonas aeruginosa;
K. pneumoniae = Klebsiella pneumoniae; E. coli = Escherichia coli; ESBL = Extended-Spectrum Beta-Lactamase;
S.agalactiae = Streptococcus agalactiae; Rt. = right; Lt. = le; - = no concordance; + = moderate concordance; ++ = strong concordance
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
98
TABLE 2. Common aerobic bacteria isolated from pus and mucosal cultures.
Pathogens Pus Mucosa Pus
(Kirtsreesakul V et al.)
12
Coagulase negative Staphylococcus aureus - - 8
Pseudomonas aeruginosa 7 4 2
Staphylococcus aureus (Methicillin-susceptible) 3 4 3
Commensalora 2 2 -
Klebsiella pneumoniae 2 1 2
Non-fermentative gram-negative rods 2 1 -
Escherichia coli 1 1 2
Enterobacter aerogenes 1 1 1
Pasteurella multocida 1 1 -
Morganella morganii 1 0 -
Citrobacter diversus - - 1
Streptococcus agalactiae 0 1 -
Alpha-hemolytic Streptococcus not group D - - 3
Beta-hemolytic Streptococcus not group A, B, D - - 1
Micrococcus spp. - - 1
No growth 12 14 1
TABLE 3. Comparison of pus and mucosal culture methods.
Culture methods Positive pus cultures Negative pus cultures Total
Positive mucosal cultures 8 2 10
Negative mucosal cultures 6 11 17
Total 14 13 27
Specicity = 84.62%; sensitivity = 57.14%; positive predictive value = 80%; negative predictive value = 64.71%
N.B.: Negative mucosal cultures = no pathogenic bacterial growth or dierent pathogenic bacterial growth from pus culture
DISCUSSION
e 40.74% rate for the mucosal cultures was not
very dierent from the 51.85% rate for the pus cultures
(p = 0.4531, 95% CI: -0.292204 to 0.084671), as shown
in Table 1. Similar isolated pathogenic bacteria were
recovered from the mucosa and pus, as shown in Table 2.
A good concordance rate of 70.37% was achieved for the
two aerobic bacterial culture techniques. e mucosal
cultures provided a specicity of 84.62, a sensitivity of
57.14%, a predictive value of a positive result of 80%, a
predictive value of a negative result of 64.71%, a positive
likelihood ratio of 3.71, and a negative likelihood ratio
of 0.51 compared with the pus cultures collected by the
maxillary antral tap (Table 3).
Antibiotic treatment is essential in patients with chronic
rhinosinusitis with acute exacerbation. Otolaryngologists
usually treat rhinosinusitis patients by empirical therapy.
If patients do not respond to antibiotics, then correct
identication of the bacterial organisms is very important.
e standard sampling method for sinus cultures is a
maxillary sinus tap through the canine fossa or inferior
meatus, or open sinus surgery utilizing the Caldwell–Luc
approach.
1-6
However, pus collection from the maxillary
sinus during surgery is sometimes unsuccessful because
Bedavanija et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
99
Original Article
SMJ
of small pus amounts, an absence of pus in the sinus, or
the accidental removal of all pus. e study by Karma
et al.
7
found that intraoperative mucosal tissue cultures
can be used as one method to get true pathogens and
yielded a higher proportion of positive bacterial cultures
than positive pus cultures, but the researchers did not
compare the types of organisms obtained from the tissue
and pus cultures. e current study is probably the rst,
at least in ailand, to compare the outcomes of cultures
derived from tissue biopsies and pus collected from the
maxillary sinus. A good concordance of the bacterial
organisms of both groups, as well as a high specicity,
and a high positive predictive value in the mucosal group
compared to the pus cultures, were found in the present
study. erefore, if we cannot collect pus, a mucosal
culture in chronic maxillary rhinosinusitis can be used
to identify the pathogenic organisms.
However, the interpretation of mucosal culture data
should be made cautiously if a negative culture is obtained
because of its low sensitivity and low negative predictive
value. e proportion of positive cultures from the sinus
mucosa in this study was slightly lower than that from
the pus cultures, which contrasts with the earlier study by
Karma et al.
7-9
is might be because bacterial pathogens
are hidden in the biolm on the surface of the mucosa and
are dicult to culture. Biolm bacteria are usually found
on the surface of the mucosa, but planktonic bacteria are
found in pus content. Biolms of Staphylococcus aureus,
Pseudomonas aeruginosa, Streptococcus pneumoniae, and
Haemophilus inuenzae are commonly found in patients
with chronic rhinosinusitis that is resisting antibiotic
treatment.
14-18
Biolm bacteria metabolize more slowly,
reproduce less frequently, and show dierent phenotypes
than planktonic bacteria, and standard microbiological
culture techniques are incapable of identifying all species
present in pus cultures.
19
Nevertheless, since the mucosal
cultures in the present study showed a good specicity
and positive predictive value, mucosal cultures may be
used as a guide to the prescribing of target antibiotics
for chronic rhinosinusitis patients who do not have any
pus in their maxillary sinuses.
We have applied this nding in our department,
especially with immunocompromised patients, e.g.,
those with hematologic malignancy who have chronic
rhinosinusitis with acute exacerbation. With such cases,
the active sinus infection tends to progress rapidly with
a high risk of serious complications. erefore, the need
for sinus surgery is usually indicated early in order to
restore drainage and obtain pus for culture because
correct antibiotics selection is very important. However,
as such patients sometimes have no, or only a small
amount of, pus in the maxillary sinus, especially during
chemotherapy, a mucosal culture can be used instead.
Moreover, we can use this method with other chronic
rhinosinusitis patients with acute exacerbation if a pus
specimen is not available.
e common pathogens were the same in both
groups, as shown in Table 1&2. is nding is another
reason why a mucosal culture is a good alternative method
for identifying causative pathogens. In a comparison
with other studies in ailand as well as with a recent
systematic review by anasumpun and Batra
23
(Table 4),
the majority of causative aerobic bacteria are usually
gram-negative rods, such as Pseudomonas aeruginosa
and Klebsiella pneumoniae, and Staphylococcus aureus.
is data should be used as a guide to the prescribing of
the correct antibiotics for chronic rhinosinusitis patients
in ailand.
Although novel molecular methods can be used to
identify bacteria, pus cultures from maxillary sinuses
are a reference standard. Culture methods from pus or
tissue are cheap, easy to perform, reliable to identify
pathogenic bacteria, and suitable to use in our country.
e drawback of this study is that it did not include
anaerobic bacterial cultures. Because anaerobic bacteria
are also common organisms in chronic rhinosinusitis,
further studies that include anaerobic bacterial cultures
are needed to develop more complete data.
CONCLUSION
The types of pathogenic bacteria recovered
from the mucosa and pus from the chronic maxillary
rhinosinusitis patients were similar. e mucosal cultures
should be a reference standard and can be used as an
alternative method when pus is unavailable, especially
with immunocompromised patients. is is because of,
rstly, the high degree of similarity between the types of
bacteria found in the mucosal and pus cultures; secondly,
the good concordance rate of the two aerobic bacterial
culture techniques; and nally, the high specicity and
positive predictive value of the mucosal cultures, compared
with the pus cultures.
ACKNOWLEDGMENTS
is study was supported by a grant from the Faculty
of Medicine Siriraj Hospital, Bangkok, ailand. e
authors thank Mr. Suthipol Udompunturak for his help
with the statistical analyses, Prof.Dr.med.Dr.hc.mult.
Wolf J. Mann for his very useful and valuable advice in
the manuscript, and Ms. Ngamrat Treerassapanich for
her support compiling the data.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
100
REFERENCES
1. Carenfelt C, Lundberg C, Nord CE, Wretlind B. Bacteriology
of maxillary sinusitis in relation to quality of the retained
secretion. Acta Otolaryngol. 1978; 86:298-302.
2. Gwaltney JM Jr, Sydnor A Jr, Sande MA. Etiology and
antimicrobial treatment of acute sinusitis. Ann Otol Rhinol
Laryngol Suppl. 1981;90:68-71.
3. Evans FO Jr, Sydnor JB, Moore WE, Moore GR, Manwaring
TABLE 4. Results of aerobic bacteriological cultures from chronic maxillary rhinosinusitis patients from several
studies in ailand and a systematic review.
Study authors Year of study Participants Results
(Number of cases) (Positive pus cultures)
Jareoncharsri P, et al.
20
1997–1999 31 74.19% (23/31)
H. inuenzae 10.77%
Non-fermentative gram-negative rod 7.69%
K. pneumoniae 4.62%
Tantilipikorn P, et al.
21
2004 162 50.61% (82/162)
P. aeruginosa 16.2%
Non-fermentative gram-negative rod 10.8%
CNS 9.9%
K. pneumoniae 9.9%
This study 2010–2011 27 51.85% (14/27)
P. aeruginosa 21.88%
S. aureus 9.38%
K. pneumoniae 6.25%
Non-fermentative gram-negative rod 6.25%
Udomsiri N and 2011–2012 85 51.76% (44/85)
Bedavanija A, et al.
22
P. aeruginosa 18.82%
S. aureus 5.88%
K. pneumoniae 4.71%
Non-fermentative gram-negative rod 2.35%
Thanasumpun T and 2015 6005 63.71% (3826/6005)
Batra PS
23
(systematic review) CNS 10.24%
S. aureus 7.82%
H. inuenzae 3.98%
Viridans streptococci 3.79%
P. aeruginosa 3.22%
N.B.: H. inuenzae=Haemophilus inuenzae, K. pneumoniae= Klebsiella pneumoniae, CNS= Coagulase negative Staphylococcus aureus,
S. aureus=Staphylococcus aureus
JL, Brill AH, et al. Sinusitis of the maxillary antrum. N Engl J
Med. 1975;273:735-9.
4. Frederick J, Braude AI. Anaerobic infection of the paranasal
sinuses. N Engl J Med. 1974; 290:135-7.
5. Bjorkwall T. Bacteriological examinations in maxillary sinusitis.
Acta Otolaryngol Suppl. 1950; 83:9-58.
6. Benninger MS, Appelbaum PC, Denneny JC, Osguthorpe
DJ, Stankiewicz JA. Maxillary sinus puncture and culture in
Bedavanija et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
101
Original Article
SMJ
the diagnosis of acute rhinosinusitis: the case for pursuing
alternative culture methods. Otolaryngol Head Neck Surg.
2002;127:7-12.
7. Karma P, Jokipii L, Sipila P, Luotonen J, Jokipii AM. Bacteria
in chronic maxillary sinusitis. Arch Otolaryngol. 1979;105:386-
90.
8. Jiang RS, Liang KL, Jang JW, Hsu CY. Bacteriology of endoscopically
normal maxillary sinuses. J Laryngol Otol. 1999;113:825-8.
9. Su WY, Liu C, Hung SY, Tsai WF. Bacteriological study in
chronic maxillary sinusitis. Laryngoscope. 1983;93:931-4.
10. Fokkens W, Lund V, Mullol J. European position paper on
rhinosinusitis and nasal polyps 2007. Rhinol Suppl. 2007;20:1-
136.
11. Fokkens W, Lund V, Mullol J, Bachert C, Alobid I, Baroody
F, et al. European position paper on rhinosinusitis and nasal
polyps 2012. Rhinol Suppl. 2012;23:1-298.
12. Kirtsreesakul V, Chatwiwat Y, Laohaprertthisan V. A comparison
between endoscopically middle meatal aspiration culture using
modied aspiration instrument and direct maxillary antral tap
culture in chronic rhinosinusitis. J Med Assoc ai. 2005;88:
1591-7.
13. Vogan JC, Bolger WE, Ketes AS. Endoscopically guided sinonasal
cultures: a direct comparison with maxillary sinus aspirate
cultures. Otolaryngol Head Neck Surg. 2000; 122:370-3.
14. Desrosiers M, Bendouah Z, Barbeau J. Eectiveness of topical
antibiotics on Staphylococcus aureus biolm in vitro. Am J
Rhinol. 2007;21:149-53.
15. Kilty SJ, Desrosiers MY. e role of bacterial biolms and the
pathophysiology of chronic rhinosinusitis. Curr Allergy Asthma
Resp. 2008;8:227-33.
16. Costerton JW, Stewart PS, Greenbreg EP. Bacterial biolms:
a common cause of persistent infections. Science. 1999;284:
1318-22.
17. Dlugaszewska J, Leszczynska M, Lenkowski M, Tatarska A,
Pastusiak T, Szyer W. e pathophysiological role of bacterial
biolms in chronic sinusitis. Eur Arch Otorhinolaryngol.
2016;273:1989-94.
18. Fastenberg JH, Hsueh WD, Mustafa A, Akbar NA, Abuzeid WM.
Biolms in chronic rhinosinusitis:pathophysiology and therapeutic
strategies. World J Otorhinolaryngol Head Neck Surg. 2016;2:
219-29.
19. Metcalf DG, Bowler PG, Hurlow J. A clinical algorithm for
wound biolm identication. J Wound Care. 2014;23:137-8,
140-2.
20. Jareoncharsri P, Bunnag C, Tunsuriyawong P, Voraprayoon S,
Srifuengfung S, Bedavanija A. Bacteriologic prole of acute
and chronic maxillary sinusitis. J Infect Dis Antimicrob Agents.
2001;18:96-102.
21. Tantilipikorn P, Bunnag C, Srifuengfung S, Dhiraputra C,
Tiensasitorn C, Jareoncharsri P, et al. A surveillance study of
bacteriologic prole in rhinosinusitis. Siriraj Med J. 2007;
59:177-80.
22. Udomsiri N. Prevalence study of Anaerobic bacteria in chronic
rhinosinusitis [dissertation]. Bangkok: Mahidol University;
2012.
23. anasumpun T, Batra PS. Endoscopic-derived bacterial cultures
in chronic rhinosinusitis: a systematic review. Amer J Otolaryngol
Head Neck Med Surg. 2015;36:686-91.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
102
Bannapuch Pinkaew, M.Sc.*, Paraya Assanasen, M.D.*, Olaf Michel, M.D., Ph.D.**, Kanokporn Talek, B.Sc.*,
aratham Phonmanee, B.Ed.*, Jeerapa Kerdnoppakhun, B.A.*
*Department of Otorhinolaryngology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand, **Department of
Otorhinolaryngology, University Hospital Brussels, Vrije Universiteit Brussel, Brussels, Belgium.
Impact Assessment of Smell and Taste Disorders
on Quality of Life in Thais Using the SF-36 Health
Survey (Thai version)
Corresponding author: Bannapuch Pinkaew
E-mail: bangon.pin@mahidol.ac.th
Received 24 July 2018 Revised 17 January 2019 Accepted 4 February 2019
ORCID ID: http://orcid.org/0000-0002-2304-9507
http://dx.doi.org/10.33192/Smj.2019.16
ABSTRACT
Objective: Smell and taste defects adversely aect both physical and mental health. e Short Form (SF)-36 Health
Survey is a widely used tool for the quality of life (QoL) assessment. e aim of this study was to investigate the
impact of smell and taste disorders on QoL in ais using the SF-36 Health Survey (ai version).
Methods: is retrospective chart review included the patients with smell and taste disorders that attended our
clinic during 2011 to 2016. Smell ability was evaluated by phenyl ethyl alcohol odor detection threshold, and smell
discrimination and identication tests. Taste ability was evaluated by electrogustometry, regional testing, and
modied taste strips. SF-36 was used to assess QoL.
Results: ree hundred y-ve patients were included in the nal analysis. e mean age was 50.8±15.5 years,
and 64.2% were female. Most patients (78.59%) had smell disorder only, 15.78% had taste disorder only, and 5.63%
had both disorders. Specic to taste disorders, QoL was signicantly lower in the patient group than in healthy
population for the following 6 domains: physical function, role-physical, bodily pain, general health, vitality, and
role-emotional (all p<0.05).
Conclusion: e four major causes of smell and taste disorders are nasal/sinonasal diseases, idiopathic causes,
post-URI, and head trauma. Women are more oen aected than men. Although smell and taste disorders both
adversely aect physical and mental health, the taste disorders cause more adverse eect. An assessment tool that is
specic to smell and taste disorders may facilitate more detailed elucidation of the eect of these conditions on QoL.
Keywords: Impact assessment; smell and taste disorders; quality of life; QoL; ai population; SF-36 Health Survey
(ai version) (Siriraj Med J 2019;71: 102-109)
INTRODUCTION
Smell and taste sensations are important neurosensory
functions that are known to be closely related. Disorder
of one, the other, or both sensations can adversely aect
mental health and quality of life
1,2
due to the loss of ability
to enjoy food and drink, as well as to enjoy and appreciate
pleasant aromas.
3,4
In addition, loss of ability to notice
toxic chemicals and gases also aects life safety. Moreover,
these disorders have been associated with aging
5
and a
broad range of diseases, including Alzheimer’s disease and
Parkinsonism.
6,7
Causes of olfactory dysfunction include
head/surgical trauma, chronic rhinosinusitis, iatrogenic
Pinkaew et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
103
Original Article
SMJ
causes, viral infection, nasal obstruction, neurologic
disorders, medications, endocrine disturbances, and
normal aging.
8,9
However, despite being common symptoms, smell
and taste problems are usually underestimated by patients
and overlooked by physicians.
10-12
Interestingly, few studies
have investigated the eect of olfactory and gustation
disorders on quality of life (QoL). Temmel, et al.,
13
reported
that a high percentage of patients with olfactory disorders
complained about daily life problems, such as cooking,
eating, detecting their own body odor, and detecting
contaminated or spoiled food. A study in 750 patients
with abnormal smell or taste perception conducted at
the University of Pennsylvania Smell and Taste Center
reported that 68% of participants experienced poor QoL,
46% had decreased appetite and body weight, and 56%
had worse daily living and psychological well-being.
8
e impact on QoL, including safety issues and interpersonal
relations, as well as eating habits and nutritional intake,
may be severely altered in a large proportion of patients
with olfactory disorders.
13-14
Furthermore, complaints
of decreased taste function actually reect decreased
smell function.
8
e Short Form (SF)-36 Health Survey questionnaire
is a widely used tool for the assessment of QoL in various
diseases.
15-16
Our review of the literature revealed very few
studies that investigated the impact of smell disorder on
QoL using the SF-36.
17-19
All 3 of those studies reported
a markedly lower QoL in patients with smell problems,
especially in the domains of general health (GH) and
vitality (VT). e SF-36 was translated into ai with
permission from the original developer in 2000
20
, and
the SF-36 (ai version) has since been evaluated several
times
21
. To our knowledge, no study has evaluated the eect
of smell and taste disorders on QoL in ai population.
Accordingly, the aim of this study was to investigate
the impact of smell and taste disorders on QoL in ai
population using the SF-36 Health Survey questionnaire
(ai version 1.0).
MATERIALS AND METHODS
Subjects
is retrospective chart review included patients
with smell and taste disorders that attended the Smell and
Taste Clinic of the Division of Allergy and Rhinology,
Department of Otorhinolaryngology, Faculty of Medicine
Siriraj Hospital, Mahidol University, Bangkok, ailand
during the August 2011 to June 2016 study period. Siriraj
Hospital is ailand’s largest national tertiary referral
center. e protocol for this study was approved by the
Siriraj Institutional Review Board (Si 387/2013).
Patient data were collected, including age, gender,
smoking habits, medication history, and possible causes of
smell and/or taste disorders. Smell and/or taste tests and
ENT examination were then performed. e validated
ai version (version 1.0) of the SF-36 Health Survey was
used (with permission) to assess patient QoL.
20
e smell
tests included phenyl ethyl alcohol (PEA) odor detection
threshold, and smell discrimination and identication
tests. e taste tests included electrogustometry (EGM),
regional (spatial) testing of taste function, and modied
taste strips.
Smell testing
e following 3 smell tests were performed: smell
detection, discrimination, and identication. e smell
detection threshold test consisted of a forced-choice
single-staircase method using PEA, which is a rose-like
odorant in dierent diluents in a concentration series of
light mineral oil. PEA has been proven to activate the
olfactory nerve, but relatively low activation of the trigeminal
nerve.
22
e objective of the smell detection threshold
test was to identify the lowest concentration of PEA that
the patient could detect.
23,24
e smell discrimination
test (SDT) is used to determine a person’s ability to
distinguish between odorous substances (e.g., coee) and
water, which is an odorless substance.
25
For the smell
identication test (SIT), a total of 10 common odorous
substances concentrated to suprathreshold concentration
levels were selected. e aim of the SIT is to determine a
person’s ability to correctly identify and name odorants.
25
Since smell identication is strongly aected by culture
and eating behavior, the tested odorants must appropriately
match the culture and dietary habits of the person being
tested.
Taste testing
Electrogustometry (EGM) is a measurement of
taste threshold that was performed using a commercially
available electrogustometer (Hortmann Neuro-Otometric
Myoton, Alberti Technikhandel GmbH, Bad Neuenahr-
Ahrweiler, Germany). Regional (spatial) taste function
testing was performed using a test developed by the
Connecticut Chemosensory Clinic Research Center
(CCCRC) (Farmington, New Mexico, USA).
26
e aim of
the regional taste function test is to examine taste ability
at areas innervated by the greater supercial petrosal
branch (right and le so palate) and the chorda tympani
branch (right and le anterior tongue) of cranial nerve
VII and cranial nerve IX (right and le posterior tongue).
e 4 suprathreshold tastants used for testing included
1.0 mol sodium chloride (NaCl) for salty avor, 1.0 mol
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
104
sucrose for sweet avor, 0.03 mol citric acid for sour
avor, and 0.001 mol quinine hydrochloride (QHCl) for
bitter avor. Patients were asked not to smoke, brush
their teeth, or eat or drink anything other than water for
1 hour before the test. Each tastant was applied to each
area of the palate (areas 1-2) and tongue (areas 3, 4, 5,
and 6) (Fig 1) with a sterile cotton bud according to a
randomized order, for a total of 24 tests (4 tastants x 6
tested areas). e patients had to identify the taste as either
salty, sweet, sour, or bitter. If the answer was correct, a
score of 1 was given. Patients were also asked to rate the
intensity of each tastant on a visual analogue scale (VAS:
0-10), with a higher score indicating a stronger level of
taste intensity. Aer each of the 24 tests, the patient was
asked to rinse his/her mouth prior to the next test.
Fig 1. Images showing the palate and tongue areas that are swabbed
during regional testing. (Drawing by Miss Boonyisa Pinkaew)
Modied taste strips were applied using a sterile
cotton bud (length: 15 cm) with a tip area of 1 cm in
length (small size). Each of the 16 cotton buds was
impregnated with one of the 4 tastants (salty, sweet, sour,
or bitter). Each type of tastant was tested at 4 dierent
concentrations, as follows: salty – 1.0, 0.75, 0.50, and
0.25 mol/L sodium chloride; sweet – 1.0, 0.75, 0.50, and
0.25 mol/L sucrose; sour – 0.03, 0.0225, 0.015, and 0.0075
mol/L citric acid; and, bitter – 0.001, 0.00075, 0.0005,
and 0.00025 mol/L quinine hydrochloride. ese tests
were performed at all areas of both the right side and
le side of the tongue (Fig 2). e patient had to identify
the taste as salty, sweet, sour, or bitter. A correct answer
was given a score of 1. e test score ranged from 0 to
16 for each side of the tongue.
Quality of life (QoL) questionnaire
e ai version (version 1.0) of the SF-36 Health
Survey is a 36-item patient-reported survey of health that
consists of scaled scores for the following 8 domains:
Fig 2. Image showing the areas of the tongue that are swabbed during
modied taste strip testing. (Drawing by Miss Boonyisa Pinkaew)
physical functioning (PF); role-physical (RP); bodily pain
(BP); general health (GH), which measures physical health;
vitality (VT); social functioning (SF); role-emotional (RE);
and, mental health (MH), which measures mental health.
e SF-36 (ai version) was adapted from English, and
was shown to have proven validity and reliability.
20-21
Raw scores are transformed to scores ranging from 0 to
100, with higher scores indicating better health status.
Data analysis
All statistical analyses were performed using SPSS
Statistics version 18.0 (SPSS, Inc., Chicago, IL, USA).
Descriptive statistics were used to summarize participant
demographic and lifestyle characteristics, and the causes
of smell and/or taste disorders. Data are reported as
percentage, number and percentage, or mean ± standard
deviation (SD). Normally distributed continuous data
were evaluated by One-Sample Kolmogorov-Smirnov
Test, and are reported as mean ± SD. Demographic data,
including age and duration of disorder, were compared
between groups using one-way analysis of variance
(ANOVA). e number of cases was compared between
groups using Pearson’s chi-square test. QoL scores were
analyzed using Kruskal-Wallis test for non-normally
distributed data (PF, RP, BP, VT, SF, RE, and MH),
and by ANOVA for normally distributed data (GH).
A p-value less than 0.05 was considered statistically
signicant for all tests.
RESULTS
Of 372 patients with complaints of smell and/or
taste disorders who were seen at our clinic during the
study period, 17 (4.57%) patients were excluded for
Pinkaew et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
105
Original Article
SMJ
having either incomplete data or normal smell and/or
taste test result. e remaining 355 patients were enrolled
and included in the nal analysis. We categorized three
groups of disorders for further evaluation smell only, taste
only and combined smell and taste disorder determined
by phenyl ethyl alcohol (PEA) detection threshold, smell
discrimination and identication tests and taste ability
including electrogustometry (EGM), regional (spatial)
testing of taste function and modied taste strips. Two
hundred and seventy-nine patients (78.6%) had problems
with smell only (mean age: 49.9 ±16.0 years, range: 8-87;
102 males and 177 females), 56 patients (15.8%) had
problems with taste only (mean age: 56.5±13.9 years,
range: 15-81; 16 males and 40 females), and 20 patients
(5.6%) had problems with both smell and taste (mean
age: 48.7±12.1, range: 30-68; 5 males and 15 females).
Demographic characteristics, and causes of smell and/
or taste disorders are shown in Table 1. More women
had smell and/or taste problems than men (64.2% vs.
35.8%, respectively). Almost all patients (82.3%) were
non-smokers, with 9.6% being ex-smokers and 8.2%
being current smokers. e diagnosed cause of smell
and/or taste disorders was nasal/sinonasal diseases (SND;
50.7%), unknown (idiopathic) cause (21.7%), post-URI
(10.1%), head/surgical trauma (10.1%), chemical exposure
(1.7%), and other causes (5.6%).
SF-36 (ai version) scoring compared among
Thai patients with smell and/or taste disorders and
healthy ai population is shown in Table 2. ere was
no signicant dierence among the smell only, taste
only, and smell and taste disorder groups, except for
the Physical function (PF) dimension (p=0.013), and the
signicant dierence was between the smell only and the
smell and taste defect groups. A signicant dierence
was found between smell only and taste only, and also
between both smell and taste and taste only. However,
patients with taste disorders tended to have poorer QoL
among the 3 groups followed by patients with smell
disorders and patients with smell and taste disorders,
respectively. Nevertheless, in comparison between each
group of this study and the data of a healthy population
21
,
QoL was signicantly lower in the domain of general
health (GH) which evaluates overall health, including
current health, health outlook and resistance to illness
in all groups. In addition, there was also signicantly
lower score in some domains especially, vitality (VT)
which evaluates feeling energetic and full of pep versus
feeling tired and worn out (Table 2).
TABLE 1. Demographic characteristics, and causes of smell and/or taste disorders.
Smell Taste Smell and
All groups
Characteristics disorder disorder taste disorders
(N=355)
(N=279) (N=56) (N=20)
Age (yr), (mean±SD) 49.9±16.0 56.5±13.9 48.7±12.1 50.8±15.5
Age range (yr) 8-87 15-81 30-68 8-87
Gender, n (%)
Male 102 (36.6%) 16 (28.6%) 5 (25.0%) 127 (35.8%)
Female 177 (36.4%) 40 (71.4%) 15 (75.0%) 228 (64.2%)
Smoking status, n (%)
Never smoked 228 (81.7%) 48 (85.7%) 16 (80.0%) 292 (82.3%)
Current smoker 23 (8.2%) 5 (8.9%) 1 (5.0%) 29 (8.2%)
Stopped smoking 28 (10.0%) 3 (5.4%) 3 (15.0%) 34 (9.6%)
Duration of disorder (mo), (mean±SD) 34.7±54.4 17.9±34.5 19.7±24.7 33.2±52.6
Causes, n (%)
Nasal/sinonasal diseases 162 (58.1%) 8 (14.3%) 10 (50.0%) 180 (50.7%)
Post-URI 34 (12.2%) 2 (3.6%) 0 (0.0%) 36 (10.1%)
Head/surgical trauma 33 (11.8%) 0 (0.0%) 3 (15.0%) 36 (10.1%)
Chemical exposure 5 (1.8%) 0 (0.0%) 1 (5.0%) 6 (1.7%)
Idiopathic causes 34 (12.2%) 37 (66.1%) 6 (30.0%) 77 (21.7%)
Others 11 (3.9%) 9 (16.1%) 0 (0.0%) 20 (5.6%)
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
106
TABLE 2. SF-36 score compared among ai patients with smell and/or taste disorders and healthy ai population.
Smell Taste Smell and taste Healthy
SF-36 scores disorder disorder disorders population
(mean±SD) (mean±SD) (mean±SD) (mean±SD)
(N=279) (N=56) (N=20) (N=1,345)
21
Physical function (PF) 75.53±20.17 67.23±21.55
#
77.50±17.43 77.50±17.40
Role-physical (RP) 71.14±38.94* 63.84±43.93
#
72.50±37.96 82.20±28.50
Bodily pain (BP) 68.85±19.56* 66.27±19.23
#
69.75±21.79 74.90±18.20
General health (GH) 50.37±16.35* 44.93±19.50
#
50.05±15.99*
#
65.20±17.40
Vitality (VT) 57.75±16.38* 51.88±19.44
#
58.50±13.48 61.80±13.50
Social functioning (SF) 72.76±22.88* 73.88±21.22 71.88±15.64 78.30±18.50
Role-emotional (RE) 64.16±42.29* 60.12±44.69
#
81.67±35.00 80.20±31.80
Mental health (MH) 65.02±18.04 62.64±20.36 67.60±14.90 65.50±13.00
A P-value<0.05 indicates statistical signicance
*Statistically signicant by unpaired t-test between smell disorder and healthy ai population
#
Statistically signicant by unpaired t-test between taste disorder and healthy ai population
*
#
Statistically signicant by unpaired t-test between smell and taste disorders and healthy ai population.
DISCUSSION
Few studies have investigated the impact of smell
and/or taste impairment on QoL. Moreover, the study
of smell and taste disorders in ai patients are rather
new and rare. We are able to report the results from a
large study population because our clinic is the rst Smell
and Taste Clinic to be established in ailand. Clinical
investigations have shown poor general QoL, depression
and mood changes, and diculties in daily life in this
patient population. is is the rst investigation of QoL
impairment due to smell and/or taste disorders in ai
patients using the SF-36 Health Survey (ai version).
Our ndings revealed the most common etiologies
to be nasal/sinonasal diseases (SND) (50.7%), idiopathic
cause (21.7%), post-URI (10.1%), head/surgical trauma
(10.1%), chemical exposure (1.7%), and other causes
(5.6%), including aging, diabetes, hypertension, genetic
disorders (Kallmann syndrome), Parkinsonism, and other
neurologic disorders. ese results are similar to those
reported from other studies (Table 3). Kaolawanich,
et al.,
27
reported possible causes of smell disorder in a
clinical population of ais to be SND (66.7%), head
injury (12.1%), idiopathic cause (10.6%), post-viral upper
respiratory infection (URI) (6.8%), congenital cause (3%),
and other (0.8%). Damm, et al., found inammatory
diseases of the nose/paranasal sinus (53%), postviral
conditions (11%), idiopathic causes (6%), head trauma
(5%), chemical exposure (2%), and other (23%) to be
the most common causes of smell defect.
28
Miwa, et al.,
4
reported dierent proportions of the same causes reported
by Damm, et al. Nordin and Bramerson
29
reported the
most common etiologies of smell loss to be URI (18-
45% of the clinical population) and nasal/sinus disease
(7-56%), followed by head trauma (8-20%), exposure to
toxins/drugs (2-6%), and congenital anosmia (0-4%).
Amongst the range of etiologies that have been reported
from numerous studies
8,13
, the most commonly reported
cause of smell disorders was URI, followed by idiopathic
cause and head injury. ese variations in reported
incidences among etiologies may be due to dierences
in race, environment, socioeconomic status, lifestyle, and
culture- all of which may play an important role in smell
and taste disorders.
27
As a result, the causes reported
from various studies are very similar, with dierences
only in the proportions of patients aected within each
study.
e data from this study revealed taste disorders
as having the most impact on QoL in all domains of the
SF-36 Health Survey (ai version). Signicantly lower
scores were found in 6 domains when compared to healthy
population, including physical function (p<0.0001),
role-physical (p<0.0001), bodily pain (p=0.0005), general
health (p<0.0001), vitality (p<0.0001), and role-emotional
(p<0.0001). Smell only and combined smell and taste
Pinkaew et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
107
Original Article
SMJ
TABLE 3. Etiology of olfactory disorders from this study and prior studies.
Etiology
Pinkaew, et al.(2019) Miwa, et al. (2011) Kaolawanich, et al. (2009) Damm, et al. (2004)
% % % %
Nasal/sinonasal diseases 50.7% 21.4% 66.7% 53.0%
Idiopathic causes 21.7% 28.4% 10.6% 6.0%
Post-URI 10.1% 17.1% 6.8% 11.0%
Head/surgical trauma 10.1% 17.1% 12.1% 5.0%
Chemical exposure 1.7% 3.2% 0.0% 2.0%
Other causes 5.6% 11.6% 3.8% 23.0%
Abbreviation: URI =upper respiratory infection
disorders resulted in a higher score. is may be due to
the ai cultural habit of consuming hot and spicy food.
Trachootham, et al.,
30
reported that ai people have a
strong preference for spicy food, with a preference for
mild-moderate spicy food in 70%, and very spicy food
in 10% of respondents. at study also found that 70%
of ai people consumed spicy food on a weekly basis.
ese ndings suggest that populations with a preference
for hot and spicy foods, such as ailand, may have
much poorer taste sensitivity and perception.
30
Another
reason that SF-36 (ai version) scores were signicantly
lower in patients with taste defect may be due to the
small number of patients in the taste disorder group.
Future study in a larger study population is, therefore,
recommended-especially since problems with taste were
found to have the most adverse eect on QoL.
In groups of taste only and combined, higher scores
were observed. is was striking, but could also a bias
due to the small-sized number of patients in both groups.
Moreover, in a comparison among the 3 study groups,
the lowest scores were observed in the General health
(GH) and Vitality (VT) domains. Additionally, when
compared with 2019 data from healthy ai population
21
,
the scores were signicantly lower in all domains and in
all groups. Prior studies in the impact of smell disorders
on QoL using the SF-36 Health Survey are shown in
Table 4 and Fig 3. ese studies show a similar pattern in
each domain. Consistent with the ndings of the present
study, the lowest scores were found in the General health
(GH) and Vitality (VT) domains.
17-19
e GH domain
assesses QoL relative to general physical health, and the
VT domain assesses QoL relative to the patient’s level of
feeling drowsy or sedated.
16
Interestingly, the ndings
of the present study are lower than those reported in
all previous reports
17-19
, which may signify the impact
of dierences in race, behavior, culture, environment,
socioeconomic status, and lifestyle among different
cultures.
is study has some mentionable limitations. First
and consistent with the retrospective nature of this
study, some patient data may have been missing or
incomplete. Second, the size of two of our three study
groups was relatively small. As a result, our study may
have lacked sucient power to identify all signicant
dierences and associations. ird, the patients enrolled
in this study were from a single center, which is located
within a large urban metropolis. Fourth, our center is
ailand’s largest tertiary referral hospital, which means
that we are oen referred patients with complicated
and intransigent conditions. As such, it is possible that
our ndings may not be generalizable to patients with
the same condition in other settings. Importantly, the
strength of this study is that this data is representative
of treatment outcomes in a real-world setting from the
rst Smell and Taste Clinic to be established in ailand.
CONCLUSION
e present study revealed that ai women suer
from smell and taste disorders more than ai men. e
four major causes of smell and taste disorders are nasal/
sinonasal diseases (50.7%), idiopathic causes (21.7%),
post-URI (10.1%), and head trauma (10.1%). Smell and
taste disorders adversely aect both physical and mental
health, with taste disorders adversely eectuating more
causation than smell disorders. Although the SF-36 is
the most widely used assessment for evaluating QoL, a
questionnaire that is specic to smell and taste disorders
may facilitate more detailed elucidation of the eect of
these conditions on QoL.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
108
TABLE 4. Mean SF-36 scores by domain in patients with smell disorders from this study and prior studies.
Pinkaew, et al. Katotomichelakis, Neuland, et al. Seems, et al.
SF-36 scores (2019) et al. (2013)
17
(2011)
18
(2009)
19
(mean±SD) (mean±SD) (mean±SD) (mean±SD)
(N=279) (N=89) (N=280) (N>82)
Physical function (PF) 75.53±20.17 80.00±19.42 71.53±27.50 81.44±22.78
Role-physical (RP) 71.14±38.94 69.66±28.52 56.52±42.77 75.00±37.55
Bodily pain (BP) 68.85±19.56 87.67±16.54 61.51±30.31 84.38±20.63
General health (GH) 50.37±16.35 60.07±19.43 52.73±20.08 66.44±20.73
Vitality (VT) 57.75±16.38 68.09±21.88 52.43±20.69 64.43±17.74
Social functioning (SF) 72.76±22.88 83.57±18.42 74.07±25.72 91.52±21.44
Role-emotional (RE) 64.16±42.29 74.53±28.34 63.10±42.84 84.67±29.99
Mental health (MH) 65.02±18.04 68.40±19.46 66.36±19.77 73.24±16.13
Fig 3. Comparison of the mean SF36 scores for each domain among 4 studies.
Physical function domain (PF) Role-physical domain (RP)
Bodily pain domain (BP) General health domain (GH)
Vitality domain (VT) Social functioning domain (SF)
Role-emotional domain (RE) Mental health domain (MH)
ACKNOWLEDGMENTS
e authors gratefully acknowledge Assoc. Prof.
Dr. Watcharee Leurmarnkul of the Department of
Pharmacology, Faculty of Pharmacy, Silpakorn University,
Nakhon Pathom, ailand for granting us permission
to use the SF-36 Health Survey (ai version 1.0); Miss
Booyisa Pinkaew for her assistance with anatomical
illustrations; Mr. itiphol Chomsook for assistance with
data acquisition; and Mr. Suthipol Udompunthurak of
the Division of Clinical Epidemiology, Department of
Research and Development, Faculty of Medicine Siriraj
Hospital, Mahidol University for assistance with statistical
analysis.
Pinkaew et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
109
Original Article
SMJ
Conict of interest declaration: All authors declare no
personal or professional conicts of interest relating to
any aspect of this report.
Funding disclosure: is was an unfunded study.
REFERENCES
1. Lombion-Pouthier S, Vandel P, Nezelof S, Haen E, Millot JL.
Odor perception in patients with mood disorders. J Aect
Disord 2006;90:187-91.
2. Sergerie K, Chochol C, Armony JL. e role of the amygdala
in emotional processing: A quantitative meta-analysis of
functional neuroimaging studies. Neurosci Biobehav Rev
2008;32:811-30.
3. Mattes RD, Cowart BJ. Dietary assessment of patients with
chemosensotyr disorders. J Am Diet Assoc 1994;94:50-56.
4. Miwa T, Furukawa M, Tsukatani T, Costanzo RM, DiNardo
LJ, Reiter ER. Impact of olfactory impairment on quality of life
and disability. Arch Otolaryngol Head Neck Surg 2001;127:
497-503.
5. Doty RL, Shaman P, Applebaum SL, Giberson R, Siksorski L,
Rosenberg L. Smell identication ability: changes with age.
Science 1984;226:1441-3.
6. Schiman SS. Taste and smell in disease (rst of two parts).
N Engl J Med. 1983;308:1275-9.
7. Chalke HD, Dewhurst JR. Accidental coal-gas poisoning; loss
of sense of smell as a possible contributory factor with old
people. Br Med J. 1957;2:915-7.
8. Deems DA, Doty RL, Settle RG, Moore-Gillon V, Shaman P,
Mester AF, et al. Smell and taste disorders, a study of 750
patients from the University of Pennsylvania Smell and Taste
Center. Arch Otolaryngol Head Neck Surg 1991;117:519-28.
9. Schiman SS. Taste and smell losses in normal aging and
disease. JAMA 1997;278:1357-62
10. Marschner H, Gudziol H, Guntinas-Lichius O. Olfactory
dysfunctions are substantially more frequent than they are
complained. Laryngorhinootologie 2010;89:654-9.
11. Landis BN, Hummel T, Hugentobler M, Giger R, Lacroix JS.
Ratings of overall olfactory function. Chem Senses 2003;28:
691-4.
12. Welge-Luessen A, Hummel T, Stojan T, Wolfensberger M.
What is the correlation between ratings and measures of
olfactory function in patients with olfactory loss? Am J Rhinol
2005;19:567-71.
13. Temmel AF, Quint C, Schickinger-Fischer B, Klimek L, Stoller
E, Hummel T. Characteristics of olfactory disorders in relation
to major causes of olfactory loss. Arch Otolaryngol Head Neck
Surg 2002;128:635-41.
14. Hummel T, Nordin S. Olfactory disorders and their consequences
for quality of life. Acta Otolaryngol 2005;125:116-21.
15. Stewart AL, Hays RD, Ware JE, Jr. e MOS short-form general
health survey. Reliability and validity in a patient population.
Med Care 1988;26:724-35.
16. Ware JE, Jr., Sherbourne CD. e MOS 36-item short-form
health survey (SF-36). I. Conceptual framework and item
selection. Med Care 1992;30:473-83.
17. Katotomichelakis M, Simopoulos E, Tripsianis G, Prokopakis
E, Danielides G, Velegrakis SG, et al. Improvement of olfactory
function for quality of life recovery. Laryngoscope. 2013;123:
E10-6.
18. Neuland C, Bitter T, Marschner H, Gudziol H, Guntinas-
Lichius O. Health-related and specic olfaction-related quality
of life in patients with chronic functional anosmia or severe
hyposmia. Laryngoscope. 2011;121(4):867-72.
19. Smeets MA, Veldhuizen MG, Galle S, Gouweloos J, de Haan
AM, Vernooij J, et al. Sense of smell disorder and health-related
quality of life. Rehabil Psychol 2009;54:404-12.
20. Leurmankul W. Meetam P. Development of a quality of life
questionnaire: SF-36 (ai version). ai J Pharm Sci 2000;24:
92-111.
21. Lim LLY, Seubsman S-a, Sleigh A. ai SF-36 health survey:
tests of data quality, scaling assumptions, reliability and validity
in healthy men and women. Health Qual Life Outcomes
2008;6:52.
22. Doty RL, Gregor TP, Settle RGJCS. Inuence of intertrial
interval and sni-bottle volume on phenyl ethyl alcohol odor
detection thresholds. Chem Senses. 1986;11(2):259-64.
23. Doty RL. Practical approaches to clinical olfactory testing.
In: Seiden A, ed. Taste and smell disorders. New York: ieme;
1996.p.38-51.
24. Assanasen P, Tunsuriyawong P, Pholpornphisit W, Chatameteekul
M, Bunnag C. Smell detection threshold in ai adults. J Med
Assoc ai 2009;92:813-6.
25. Pinkaew B, Assanasen P, Bunnag C. Smell discrimination
and identication scores in ai adults with normosmia. Asian
Biomedicine.2017;9:789.
26. Bartoshuk L. Clinical evaluation of the sense of taste. Ear Nose
roat J 1989;68:331-7.
27. Kaolawanich A, Assanasen P, Tunsuriyawong P, Bunnag C,
Tantilipikorn P. Smell disorders: a study of 132 patients
from the rst Smell and Taste Clinic of ailand. J Med Assoc
ai 2009;92:1057-62.
28. Damm M, Temmel A, Welge-Lussen A, Eckel HE, Kre MP,
Klussmann JP, et al. Olfactory dysfunctions. Epidemiology and
therapy in Germany, Austria and Switzerland. HNO 2004;52:
112-20.
29. Nordin S, Bramerson A. Complaints of olfactory disorders:
epidemiology, assessment and clinical implications. Curr Opin
Allergy Clin Immunol 2008;8:10-15.
30. Trachootham D, Satoh-Kuriwada S, Lam-Ubol A, Promkam
C, Chotechuang N, Sasano T, et al. Dierences in Taste Perception
and Spicy Preference: A ai-Japanese Cross-cultural Study.
Chem Senses 2017;43:65-74.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
110
Natakorn I-Tuporn, M.N.S.*, Amornrat Ratanasiri, Ph.D.**, itima Nutravong, Ph.D.***, Kultalee Boonprasert,
M.D.****, Tasanee Na Pikul, M.N.S.****
*Faculty of Nursing, Chiang Rai College, Chiang Rai 57000, **Department of Community Medicine, ***Department of Microbiology, Faculty of
Medicine, Khon Kaen University, Khon Kaen 40002, ****Chiang Rai Regional Hospital, Chiang Rai 57000, ailand.
Risk Scoring System for the Prediction of
Postpartum Blood Loss over 300 mL at Chiang Rai
Regional Hospital
Corresponding author: itima Nutravong
E-mail: thicha@kku.ac.th
Received 21 November 2018 Revised 14 January 2019 Accepted 13 February 2019
ORCID ID: http://orcid.org/0000-0001-8039-6401
http://dx.doi.org/10.33192/Smj.2019.17
ABSTRACT
Objective: To develop an assessment tool for the prediction of postpartum blood loss over 300 mL and 500 mL
aer vaginal delivery.
Methods: A retrospective case-control study of 504 (72 cases: 432 controls) pregnant women with gestational age
of 28 weeks or above who underwent vaginal delivery at Chiang Rai Regional Hospital between 1
st
October 2015
and 30
th
September 2016 was conducted. e predictors were selected using “Sign OK” selection. e risk scoring
system was developed according to the seven steps of clinical prediction model development of Steyerberg.
Results: is form included eight predictors: maternal age of 35 years old or above, gestational age over 40 weeks,
being nulliparous, history of curettage in prior pregnancy, gestational hypertensive disorder, hemoglobin level
equal to or less than 10 g/dL, fundal height of 38 centimeters or above, and the use of pethidine for pain relief in
the 1st stage of labor. e sensitivity of predicting postpartum blood loss over 300 mL aer vaginal delivery by the
risk scoring system was 80.7%, specicity of 60.8%, and the ROC curve was equal to 0.71 at the optimal cut-o
score of four points or above. To predict the postpartum blood loss of 500 mL or above aer vaginal delivery, the
sensitivity was 88.1%, specicity of 54.6%, and the ROC curve was equal to 0.71 at the optimal cut-o score of three
points or above.
Conclusion: is form had an acceptable performance in predicting postpartum blood loss over 300 mL and 500
mL at the optimal cut-o scores of four points or above and three points or above, respectively.
Keywords: Risk score; postpartum blood loss; vaginal delivery; Chiang Rai Regional Hospital (Siriraj Med J 2019;71:
110-116)
I-Tuporn et al.
INTRODUCTION
In ailand, the incidence of postpartum hemorrhage
(PPH) was 2.30%, 2.37%, 2.44%, 2.40%, 2.39%, 2.54%,
and 2.65%, from 2009 to 2015.
1
Similarly, in Chiang
Rai province, the incidence of PPH increases by 1.12%,
1.15%, 1.34%, and 2.07%; the pregnant women who had
PPH and hysterectomy were 3.05%, 4.58%, 1.82%, and
1.23%, from 2012 to 2015.
2
Moreover, in scal year 2014
and 2015, the pregnant women who died from PPH
numbered 2 cases, and 1 case, respectively.
2
For early diagnosis of PPH, the healthcare providers
in Chiang Rai Regional Hospital used a collector bag
to measure postpartum blood loss over 300 mL aer
vaginal delivery following the LABOR strategy. Similarly,
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
111
Original Article
SMJ
many settings used collector bags as an eective tool for
measuring postpartum blood loss aer vaginal delivery
in the third stage of labor.
3-6
However, the risk of PPH
could be assessed in the antepartum or the rst stage of
labor as indicated in the study of Sittiparn and Siwadune
that identied the risk factors associated with PPH and
incorporated those risk factors into a risk score to predict
PPH in normal labor. It could predict PPH in normal
labor with sensitivity of 81.30% and specicity of 50.80%
at the optimal cut-o score of four points or above.
7
The purpose of this study was to set a tool for
use in the rst stage of labor for the early detection of
postpartum blood loss over 300 mL aer vaginal delivery
compared with using a collector bag for measuring
postpartum blood loss aer vaginal delivery in the third
stage of labor. is study was part of the risk prediction
model development for postpartum hemorrhage at the
hospitals in Chiang Rai province.
MATERIALS AND METHODS
Before the investigation of the retrospective case-control
study, twenty articles from PubMed, Science Direct, Pro
Quest, and EBSCO host related to risk factors for PPH
in studies from 2005 to 2017, in ailand and globally,
were analyzed by MedCalc Version 18.2.1 (Free Trail
Version 15 days)
12
to create the data collection form. is
study was approved by the Khon Kaen University Ethics
Committee (HE601234) and the Chiang Rai Regional
Hospital Ethics Committee on 21
st
July 2017.
Study design and population
e retrospective case-control study included 2,957
pregnant women with gestational age of 28 weeks or above
who underwent vaginal delivery between 1
st
October 2015
and 30
th
September 2016 at Chiang Rai Regional Hospital.
Five hundred and four of them who had blood loss in
the range of 50-3,000 ml were enrolled in this study. e
case study included 72 pregnant women with PPH and
the control was 432 women with no PPH. e control
groups were selected on maternal age, gestational age,
gravida, and parity similar to the cases. e ratio of case
: control was 1 : 6.
Identication of risk factors for postpartum blood
loss aer vaginal delivery
e following seven steps for development and an
ABCD for validation of the clinical prediction model
8,9
include:
Step 1: problem denition and data inspection;
step 2: coding of predictors, candidate risk factors for
postpartum blood loss of 500 mL or above aer vaginal
delivery were identied from the secondary analysis of
twenty articles from PubMed, Science Direct, Pro Quest,
and EBSCO host related to the risk factors for PPH in
studies from 2005 to 2017, in ailand and globally. ey
categorized that individual factors, obstetric factors, and
placental factors followed the 4T’s causes of PPH
10,11
to
create the data collection form.
Step 3: model specication; step 4: model estimation,
the data collection form created in steps 1-2 was used in
the investigation of the retrospective case-control study
of 72 pregnant women with PPH and 432 pregnant
women with no PPH. ey had gestational age of 28
weeks or above and underwent vaginal delivery between
1
st
October 2015 and 30
th
September 2016 at Chiang Rai
Regional Hospital. First, the samples were categorized
to 93 pregnant women with postpartum blood loss over
300 mL and 411 pregnant women with postpartum blood
loss equal to or less than 300 mL. Second, the association
between the factors and postpartum blood loss over 300
mL was determined by univariate logistic regression.
en, the factors associated with blood loss over 300 mL,
statically signicant at a p-value < 0.05, were selected
for analysis by multiple logistic regression. ere were
eight predictors for postpartum blood loss over 300
mL aer vaginal delivery. Lastly, these predictors were
conrmed by experts’ opinion agreement, namely two
obstetricians, three registered nurses, and ve nursing
instructors. is predictor identication method is called
“Sign OK” selection and included statistical analysis and
experts’ opinion agreement.
Step 5: model performance; step 6: model validity,
eight predictors approved by “Sign OK” selection were
incorporated into a risk scoring system for the prediction
of postpartum blood loss over 300 mL and 500 mL aer
vaginal delivery, with calculation of sensitivity, specicity,
and a receiver operating characteristic (ROC) curve by
the bootstrap resampling technique. However, this step
had only 415 samples for data analysis (missing data 89
records).
Step 7: model presentation; the model was presented
in the Postpartum Hemorrhage Risk Score Record Form.
Data analysis
Data were analyzed using MedCalc Version 18.2.1
(Free Trail Version 15 days)
12
in steps 1-3 and IBM SPSS
soware version 22 in steps 4-6. e variables that were
signicantly associated with postpartum blood loss over
300 mL aer vaginal delivery according to the Chi-square
test were entered into logistic regression model and used
to calculate the index of item objective congruence (IOC).
A risk scoring system was developed from the sum of
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
112
each signicant adjusted odds ratio value. e ROC curve
was calculated to determine the optimal score for the
prediction of postpartum blood loss over 300 mL and
500 mL aer vaginal delivery. Sensitivity and specicity
with an associated 95% condence interval (CI) of each
cut-o point were presented. e statistically signicant
dierence was set at a p-value < 0.05.
RESULTS
In steps 1-2, candidate risk factors for postpartum
blood loss of 500 mL or above aer vaginal delivery from
hospitalization at the labor room unit to the rst stage
of labor were as follows: individual factors as teenage
pregnancy, advanced age pregnancy, body mass index
(OR > 1.0-1.5), and hemoglobin level equal to or less
than 10 g/dL (OR > 2.0); obstetric factors as nulliparous,
primiparous, grand-multiparous, pre-term gestational
age, gestational diabetes meatus (OR > 1.0-1.5), post-term
gestation age, gestational hypertensive disorder, abnormal
presentation (OR > 1.5-2.0), prior PPH, previous abortion,
and large gestational age / multiple pregnancy / broid /
hydramnios (OR > 2.0); placental factors as placenta
previa, placenta abruption, and placenta accrete (OR
> 2.0); complications and procedures of the rst stage
of labor as induction or augmentation of labor, use of
anesthetic drugs (OR > 1.0-1.5), and chorioamnionitis
(OR > 1.5-2.0) (data not shown).
In steps 3 – 4, eight predictors for postpartum blood
loss over 300 mL aer vaginal delivery were selected using
“Sign OK” selection as being maternal age of 35 years
old or above, hemoglobin level equal to or less than 10
g/dL, being nulliparous, gestational age over 40 weeks,
curettage in prior pregnancy, gestational hypertensive
disorder, fundal height of 38 centimeters or above, and
having received pethidine for pain relief in the rst
stage of labor. In addition, their adjusted odds ratio was
transformed to a score that is presented in Table 1.
In steps 5-6, the total of the risk score ranged from
0 to 30. e sensitivity and specicity with an associated
95% CI for the risk score at dierent cut-o values are
presented in Tables 2 and 3. e ROC curve of a risk
score for the prediction of postpartum blood loss over
300 mL aer vaginal delivery demonstrated overall an
area under the curve (AUC) of 0.71 (95% CI 0.65 to 0.77)
(Table 2; Fig 1). e cut-o score was three points or
greater, which had a sensitivity of 85.5% and specicity
of 56.0%. In addition, the cut-o score of four points
or greater had a sensitivity of 80.7% and specicity of
60.8%. On the other hand, the ROC curve of the risk
score for the prediction of postpartum blood loss of 500
mL or above aer vaginal delivery demonstrated overall
an AUC of 0.71 (95% CI 0.65 to 0.77) (Table 3; Fig 2).
e cut-o score was three points or greater, which had
a sensitivity of 88.1% and specicity of 54.6%.
In step 7, the model was presented in the Postpartum
Hemorrhage Risk Score Record Form (Fig 3).
TABLE 1. Risk factors associated with postpartum blood loss over 300 mL aer vaginal delivery were identied
by “Sign OK” selection.
Risk factors
Statistical analysis Expert’s opinion
Score
(multiple logistic regression) agreement
Adjusted odds ratio Item objective
[95% CI], P-value congruence
Maternal age ≥ 35 years old 3.33 [1.44 – 7,67], .005 0.60 3
Hemoglobin ≤ 10 gram per deciliters 8.19 [3.05 – 22.00], <.001 0.90 8
Nulliparous 2.42 [1.26 – 4.63], .008 0.60 2
Gestational age >40 weeks 2.25 [1.19 – 4.24], .012 1.00 2
Curettage in prior pregnancy 3.81 [1.58 – 9.22], .003 1.00 4
Gestational hypertensive disorder 7.59 [2.49 – 23.08], <.001 0.90 7
Fundal high ≥ 38 centimeters 2.67 [1.29 – 5.35], .008 0.80 2
Having received pethidine for relief pain 2.11 [1.16 – 3.82], .014 0.60 2
in the 1
st
of labor
Statistically signicant at p-value <.05
I-Tuporn et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
113
Original Article
SMJ
TABLE 2. Performances of a risk score at dierent cut-o values for the prediction of postpartum blood loss over
300 mL aer vaginal delivery.
Cut-off Numberof ROC %sensitivity %specicity 95%CI P-value
value women with score curve
at cut-off level
0 415 0.50 100.00 NA 0.43 – 0.57 1.000
1 336 0.61 98.80 23.50 0.55 – 0.67 .002
2 336 0.61 98.80 23.50 0.55 – 0.67 .002
3 217 0.71 85.50 56.00 0.65 – 0.77 <.001
4 197 0.71 80.70 60.80 0.65 – 0.77 <.001
5 96 0.68 51.80 84.00 0.60 – 0.75 <.001
6 85 0.67 47.00 86.10 0.60 – 074 <.001
7 60 0.63 34.90 90.70 0.56 – 0.70 <.001
8 50 0.64 33.70 93.40 0.56 – 0.71 <.001
9 35 0.61 26.50 96.10 0.54 – 0.69 .001
10 28 0.59 21.70 97.00 0.52 – 0.67 .008
11 19 0.58 16.90 98.50 0.50 – 0.65 .030
12 14 0.55 12.00 98.80 0.48 – 0.63 .126
13 5 0.52 4.80 99.70 0.45 – 0.59 .524
14 5 0.52 4.80 99.70 0.45 – 0.59 .524
15 3 0.52 3.60 100.00 0.45 – 0.59 .610
16 3 0.52 3.60 100.00 0.45 – 0.59 .610
17 1 0.51 1.20 100.00 0.44 – 0.58 .865
18 0 0.50 NA 100.00 0.43 – 0.57 1.000
Abbreviations: ROC = receiver operating characteristic, NA = not analysis
Fig 1. e ROC curve of a risk score for the prediction
of postpartum blood loss over 300 mL aer vaginal
delivery.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
114
TABLE 3. Performances of a risk score at dierent cut-o values for the prediction of postpartum blood loss of
500 ml or above aer vaginal delivery.
Cut-off Numberofwomen ROC %sensitivity %specicity 95%CI P-value
value with score curve
at cut-off level
0 415 0.50 100.00 NA 0.42 – 0.58 1.000
1 336 0.61 98.50 22.40 0.54 – 0.67 .007
2 336 0.61 98.50 22.40 0.54 – 0.67 .007
3 217 0.71 88.10 54.60 0.65 – 0.77 <.001
4 197 0.71 82.10 59.20 0.64 – 0.77 <.001
5 96 0.69 55.20 83.00 0.62 – 0.77 <.001
6 85 0.67 49.30 85.10 0.59 – 0.75 <.001
7 60 0.64 37.30 89.90 0.56 – 0.72 <.001
8 50 0.64 35.80 92.50 0.56 – 0.72 <.001
9 35 0.63 29.90 95.70 0.55 – 0.71 .001
10 28 0.60 23.90 96.60 0.52 – 0.68 .008
11 19 0.58 17.90 98.00 0.50 – 0.66 .039
12 14 0.55 11.90 98.30 0.47 – 0.63 .185
13 5 0.53 6.00 99.70 0.45 – 0.61 .461
14 5 0.53 6.00 99.70 0.45 – 0.61 .461
15 3 0.52 4.50 100.00 0.45 – 0.60 .562
16 3 0.52 4.50 100.00 0.45 – 0.60 .562
17 1 0.51 1.50 100.00 0.43 – 0.58 .847
18 0 0.50 NA 100.00 0.42 – 0.58 1.000
Abbreviations: ROC = receiver operating characteristic, NA = not analysis
I-Tuporn et al.
Fig 2. e ROC curve of a risk score for the prediction of
postpartum blood loss of 500 mL or above aer vaginal
delivery.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
115
Original Article
SMJ
Fig 3. e Postpartum Hemorrhage Risk Score Record Form for prediction of postpartum blood loss over 300 mL aer vaginal delivery.
DISCUSSION
Eight predictors incorporated into the Postpartum
Hemorrhage Risk Score Record form are the risk factors
of 4T’s causing PPH. Being of maternal age of 35 years old
or above, gestational age over 40 weeks, fundal height of
38 centimeters or above, and having received pethidine
for pain relief in the rst stage of labor may aect uterine
atony, failure of contractions, and retraction of myometrial
muscle bers.
10,11
Curettage in prior pregnancy may
aect retained placenta.
10,11
Being nulliparous may aect
uterine atony and cause damage to the birth canal or
the genital tract.
10,11
Hemoglobin level equal to or less
than 10 g/dL may lead to dysfunction of the coagulation
system.
10,11
All of these may lead to excessive bleeding
and PPH. Prior studies found that being of maternal
age of 35 years old or above,
7
hemoglobin level equal
to or less than 10 d/dL,
13
and gestational hypertensive
disorder
7,14-18
were risk factors associated with PPH. On
the other hand, being nulliparous, gestational age over 40
weeks, curettage in prior pregnancy, and fundal height
of 38 centimeters or above were predictors of PPH in
this study which are dierent from those of the previous
studies. ese ndings may result from demographic
dierences. However, they were similar to results in the
secondary analysis of the twenty articles in steps 1 – 2.
For performance on the Postpartum Hemorrhage
Risk Score Record Form, it had sensitivity (80.7%) lower
than the sensitivity (88.1%) of the tool of Sittiparn and
Siwadune,
7
but it had specicity (60.8%) and the ROC
curve (0.71) higher than the specicity (54.6%) and the
ROC curve (0.66) of the tool of Sittiparn and Siwadune.
7
However, the Postpartum Hemorrhage Risk Score Record
Form incorporated many predictors (eight predictors)
and covered all of the causes of PPH (4T’s), while the
tool of Sittiparn and Siwadune
7
included only three risk
factors, which are being of maternal age of 35 years old or
above, gestational hypertensive disorder, and gestational
diabetes meatus. ese three factors cover only one of
the four causes of PPH, specically tone or uterine atony.
Although the cut-o scores were three points or greater
and four points or greater to predict postpartum blood
loss over 300 mL aer vaginal delivery had the same ROC
curve (0.71) and summation of sensitivity and specicity
(Table 2), the cut-o score of four points or greater had
higher accuracy (64.82%) and positive predictive value
(PPV) (34.01%) than the cut-o score of three points
or greater (61.93% and 32.72%, respectively). erefore,
the cut-o score of three points or greater had a high
sensitivity (85.5%) and low specicity (56.0%) and maybe
resulted in false positives leading to overtreatment. Early
detection or diagnosis, theoretically, must have a high
specicity and screening must have a high sensitivity.
9
e cut-o score of four points or greater, which had
80.7% sensitivity, 60.8% specicity, and 64.82% accuracy,
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
116
was appropriate for early detection or diagnosis in the
rst stage of labor of postpartum blood loss over 300 mL
compared to the cut-o score of three points or greater.
In practice, early diagnosis of PPH in the third stage
of labor used a collector bag to measure postpartum blood
loss over 300 mL and the healthcare providers used a
high dose uterotonic drug to prevent excessive bleeding
to 500 mL when it occurred.
1
Due to the circumstances
whereby some settings in Chiang Rai province are far
from Chiang Rai Regional Hospital, it is hard to refer the
pregnant women with PPH. us, the cut-o score of three
points or greater to predict postpartum blood loss of 500
mL or above aer vaginal delivery is inappropriate for
patients who are far from Chiang Rai Regional Hospital
but appropriate for those who are nearby. So, the cut-o
score of four points or greater to predict postpartum
blood loss over 300 mL aer vaginal delivery is more
appropriate for the early detection of PPH. e pregnant
women with PPH are immediately referred from remote
areas to Chiang Rai Regional Hospital.
To conclude, the Postpartum Hemorrhage Risk
Score Record Form is the rst useful clinical assessment
tool for the early detection or diagnosis of PPH. It has
high sensitivity and specicity for the prediction of
postpartum blood loss over 300 mL aer vaginal delivery
according to the LABOR strategy for the early detection
or diagnosis of PPH. However, it had the limitation
that some factors had a high score. For application, the
healthcare providers should take into consideration
the setting and characteristics of the pregnant women.
Moreover, there should be conrmation of the score of
each predictor and the total score by a retrospective or
prospective cohort study.
ACKNOWLEDGMENTS
e authors thank the ai Society of Maternal and
Fetal Medicine for funding this study.
REFERENCES
1. Limpanyalert P. How to reduce morbidity and mortality of
postpartum hemorrhage? Paper presented at Workshop Proceeding
of Community Practice for Patient Safety in Labor Room;
2015 Aug 7; Bangkok, ailand.
2. Ajalapong J. Problem solving for reduce maternal death from
postpartum hemorrhage [Presentation]. Chiangrai: Chiangrai
Prachanukroh Hospital; 2015.
3. Abbaspoor Z, Vaziri L. e eectiveness of a collector bag for
measurement of postpartum hemorrhage. Afr J Health 2017;
21:99-103.
4. Bamberg C, Niepraschk-von Dollen K, Mickley L, Henkelmann
A, Hinkson L, Kaufner L, et al. Evaluation of measured postpartum
blood loss aer vaginal delivery using a collector bag in relation
to postpartum hemorrhage management strategies: a prospective
observational study. J Perinat Med 2016;44:433-9.
5. Jandapradit R, Anusornteerakul S. e eectiveness of using
plastic lm collector bag for the measurement postpartum
blood loss. Srinagarind Med J 2015;30:270-5.
6. Zhang WH, Deneux-araux C, Brocklehurst P, Juszczak E,
Joslin M, Alexander S, et al. Eect of a collector bag for measurement
of postpartum blood loss aer vaginal delivery: cluster randomized
trial in 13 European countries. BMJ 2010;340:c293.
7. Sittiparn W, Siwadune T. Risk score for prediction of postpartum
Hemorrhages in normal labor at Chonburi Hospital. J Med
Assoc ai 2017;100:382-8.
8. Steyerberg EW, Vergouwe Y. Towards better clinical prediction
model: Seven steps for development and an ABCD for validation.
EUR Heart J 2014;35:1925-31.
9. Steyerberg EW. Clinical prediction models: a practical approach
to development, validation, and updating. NY: Springer; 2009.
10. Smith JR. Postpartum hemorrhage [Internet] 2017 Jul 21.
Available from: URL: http://goo.gl/YuRp4B [Cited 2018 Jan
30]
11. Cunningham FG, Williams JW, editors. Obstetrical hemorrhage.
In: Williams Obstetrics. 23rd ed. NY: McGraw-Hill Medical;
2010.p.757-803.
12. MedCalc Version 18.2.1 (Free Trail Version 15 days). Available
from: URL: https://www.medcalc.org/download.php [Cited
2017 May 3]
13. Frass KA. Postpartum hemorrhage is related to the hemoglobin
levels at Labor: observational study. Alex J Med 2015;51:33-7.
14. Durmaz A, Komurcu N. Relationship between maternal
characteristics and postpartum hemorrhage: a meta-analysis
study. e journal of nursing research: JNR. 2017 Dec.
15. Lertbunnaphong T, Leetheeragul J, itadilok W. Risk factors
of primary postpartum hemorrhage in Siriraj Hospital. Siriraj
Med J 2017;62:195-8.
16. Ngwenya S. Postpartum hemorrhage: incidence, risk factors,
and outcome in low-resource setting. Int J Womens Health
2016;8:647-50.
17. Puccilli P. e risk factors of postpartum hemorrhage during
labor. Dissertation for the degree of Doctorate of Nursing
Practice. Durham, NH: University of New Hampshire Graduate
School; 2015.
18. Sheiner E, Sarid L, Levy A, Seidman DS, Hallak M. Obstetric
risk factors and outcome of pregnancies complicated with
early postpartum hemorrhage: a population-based study. J
Matern Fetal Neonatal Med 2005;18:149-54.
I-Tuporn et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
117
Original Article
SMJ
Chanon Ngamsombat, M.D.*, Thanapat Dechasasawat, M.D.*, Jitsupa Wongsripuemtet, M.D.*, Panida
Charnchaowanish, B.Sc.*, Weerasak Muangpaisan, M.D.**, Orasa Chawalparit, M.D.*
*Department of Radiology, **Department of Preventive and Social Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok
10700, ailand.
The Evaluation of Posterior Cingulate Gyrus by
Diffusion Tensor Imaging in Alzheimer’s Disease
Patients Compared with Normal Control Subjects
Corresponding author: Orasa Chawalparit
E-mail: oak_art@yahoo.com, orasa.cha@gmail.com
Received 25 July 2017 Revised 1 November 2017 Accepted 20 November 2017
ORCID ID: http://orcid.org/0000-0001-7073-3750
http://dx.doi.org/10.33192/Smj.2019.18
ABSTRACT
Objective: Posterior cingulate gyrus atrophy is found in early clinical stage of Alzheimer’s disease (AD) patients.
1
Diusion tensor imaging (DTI) can be used for evaluating microstructure change in brain parenchyma.
2
Our
objective was to compare the microstructural change at posterior cingulate gyrus between AD patients and normal
control subjects by using DTI.
Methods: e retrospective review of 23 AD patients, diagnosed by NINCDS-ADRDA with available MRI data
including DTI, and 19 normal control subjects was performed. e DTI parameters of posterior cingulate gyrus of
each group were analyzed and compared.
Results: e mean diusivity (MD), axial diusivity and radial diusivity (RD) of posterior cingulate gyrus were
signicantly increased in AD patients compared with normal control subjects (p value <0.001, <0.001, <0.001,
respectively). e fractional anisotropy (FA) was slightly decreased in AD patients compared with normal control
subjects but did not reach statistical signicance (p value=0.71).
Conclusion: Microstructural change at posterior cingulate gyrus demonstrated by DTI parameters including MD,
axial diusivity and RD were signicantly dierent between AD patients and normal control subjects.ese results
were probably helpful for early diagnosis, evaluation, and follow up of the AD patients as correlate with clinical
ndings.
Keywords: Diusion tensor imaging; Posterior cingulate gyrus; Alzheimer’s disease (Siriraj Med J 2019;71: 117-122)
INTRODUCTION
Alzheimer’s disease (AD) is a major cause of dementia
which brings disability to older adults worldwide. Early
detection of AD can provide early treatment for this
disease and thus delay progression and disability. Many
investigations are now emerging to fulll this purpose
Abbreviations: DTI: Diusion tensor imaging; FA: Fractional anisotropy; ROI: Region of interest; MD: Mean diusivity;
RD: Radial diusivity; AD: Alzheimer’s disease
such as CSF tau level, genetic test, FDG or amyloid PET.
However, AD patients are still diagnosed late in disease
course.
1-6
Furthermore, other causes of dementia such as
dementia with Lewy bodies (DLB), frontotemporal dementia
(FTD) and idiopathic normal pressure hydrocephalus
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
118
(iNPH) also share common symptoms and signs with
AD, but need dierent specic treatments. A test that
can detect AD early and also provide dierentiation from
other causes will bring many benets to these patients.
7-9
e previous study by Pengas et al.,
10
showed that
atrophy of posterior cingulate cortex and hippocampus
was a feature of early AD. e studies by Minoshima
et al., and Nestor et al., also showed hypometabolism
in both areas.
11-13
Diusion tensor imaging (DTI) is one of the imaging
tools for studying microstructural change of white matter
which is believed to account for early AD. Previous
studies
14-16
demonstrated abnormal FA and MD in posterior
cingulate gyrus in AD patients. Our objective was to
compare microstructural change at posterior cingulate
gyrus between AD patients and normal control subjects
by using DTI.
MATERIALS AND METHODS
e study was approved by Siriraj Institutional
Review Board (Si 651/2557). Twenty-three AD patients
(11 males and 12 females, mean age 78.82 years; range
66-90 years) diagnosed according to NINCDS-ADRDA
criteria 2011
17,18,19
with available MRI scan including
DTI data were included in the disease group (Table 1).
Nineteen normal cognitive subjects (10 males and 9
females, mean age 59.52 years old; range 47-80 years)
were enrolled as the control group and recruited from
the patients who underwent the MRI study due to non-
specic symptoms such as headache or vertigo with
normal cognitive function, no abnormal neurological
examination and no detectable gross MRI abnormality
such as infarction, hemorrhage or mass lesion.
MR Imaging Data Acquisition
MRI acquisitions were done on two machines; the
rst one was a 3.0 tesla MR system (Archieva, Philips,
e Netherlands) with an 8-channel head coil and DTI
protocol was a single shot, spin echo EPI; 32 diusion
encoding directions; b value = 0 and 800 s/mm
2
; acquisition
matrix 112x112; FOV 22.4 cm; voxel size = 2 mm (RL)
x 2 mm (AP) with 60 contiguous slices, slice thickness
2.3 mm, and acquisition time was 12:39 min. e other
machine was a 3.0 tesla MR system (Ingenia, Philips
Medical System, Best, the Netherlands) with 16-channel
head coil and DTI protocol was a single shot, spin echo
EPI; 32 diusion encoding directions; b-value = 0 and
800 s/mm
2
; acquisition matrix 112x112; FOV 22.4 cm;
voxel size = 2 mm (RL) x 2 mm (AP) with 60 contiguous
slices, slice thickness = 2.0 mm, and acquisition time
was 12:14 min. e 3D-T1 weighted image and uid-
attenuated inversion recovery images were performed in
all patients. In AD group, coronal oblique T1W at right
angles to the longitudinal axis of the hippocampus was
also obtained.
DTI Processing
Processing of the diusion data including brain
extraction and correction for eddy current distortions
by using FSL (FMRIB Soware Library, version 5.0.8)
were done.
20
en dtit for local tting of diusion
tensors was performed. e standard template for ROI
measurement (Harvard-Oxford Cortical Structural Atlas)
of the posterior cingulate gyrus was placed. Finally, the
diusion tensor parameters including FA, MD, axial
diusivity, and RD were derived and compared between
the AD group and control group. (Fig 1)
Statistical analysis
Statistical analysis was performed using PASW (SPSS)
version 18. Continuous data analysis was measured and
presented as means, standard deviations (SD), min and
max parameters. Categorical data analysis was divided by
each group and converted into percentage. Sex, education
and Scheltens score between AD patients and normal
control subjects were compared using Chi-square test.
Age at scanning between each group was compared
using unpaired t-test.
e diusion tensor parameters (FA, MD, Axial
diusivity, and RD) were compared between the AD group
and control group using Mann-Whitney U test. P-value
less than 0.05 was considered statistically signicant.
RESULTS
We investigated 23 AD patients (11 males, 12
females, mean age at scanning: 78.8±6.1 years) and 19
normal control subjects (10 males, 9 females, mean age
at scanning: 55.0±9.0 years). e age of scan in AD group
was higher than that in control group (p< 0.001). No
gender dierence between the two groups (p= 0.757)
was observed.
Regarding education in AD group, there were
2 patients (8.7%) with no study, 11 patients (47.8%)
with primary school graduation, 5 patients (21%) with
secondary school graduation and 5 patients (21%) with
college degree. e mean age of onset was 77.6±7.0 years
and mean TMSE scores (0-30) was 20.83±4 .5 points in
AD group.
Scheltens score (range from 0-4 points) in AD group
were 2 points in six patients (26.1%), 3 points in een
patients (65.2%) and 4 points in two patients (8.7%)
whereas the Scheltens score in control group was 1 point
Ngamsombat et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
119
Original Article
SMJ
TABLE 1. Shows demographic data as sex, age, TMSE score, education, Scheltens score, age at MRI scanning and
underlying diseases of AD patient group.
Number Sex Age TMSE Education Scheltens Age at MRI Underlying
(years) (0-30) score scanning diseases
(0-4) (years)
AD1 Male 87 24 Primary school 3 87 DM, HT, DLP
AD2 Female 78 21 Primary school 3 77 DM, HT, DLP
AD3 Male 66 19 Secondary school 3 66 None
AD4 Female 87 22 Primary school 3 87 HT
AD5 Male 82 21 Secondary school 3 82 HT, DLP
AD6 Female 79 25 Primary school 3 79 HT
AD7 Female 84 19 Primary school 3 84 HT, DLP
AD8 Female 90 19 No study 3 90 DM, HT
AD9 Male 83 14 University 3 83 DM
AD10 Male 71 23 No study 2 71 HT, DLP
AD11 Male 72 23 University 2 73 HT, DLP
AD12 Female 80 18 Secondary school 3 81 HT, DLP
AD13 Female 84 21 Primary school 3 84 DM, HT, DLP
AD14 Female 79 26 University 3 79 HT, DLP
AD15 Male 67 22 Secondary school 2 70 DM, HT, DLP
AD16 Male 74 8 Secondary school 2 74 HT, DLP
AD17 Female 76 16 Primary school 3 78 HT, DLP
AD18 Female 81 15 Primary school 4 82 DM, HT
AD19 Female 67 23 Primary school 2 70 HT, DLP
AD20 Female 79 27 Primary school 3 80 HT, DLP
AD21 Male 71 24 Primary school 3 79 HT, DLP
AD22 Male 81 27 University 4 82 None
AD23 Male 68 22 University 2 75 None
Abbreviations: DM=diabetes mellitus, HT=hypertension, DLP= dyslipidemia
Fig 1. Showed region of interest
analysis of the posterior cingulate
gyrus demonstrated in axial,
sagittal and coronal views in FA
map in AD patient.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
120
TABLE 2. Shows mean FA, mean MD, mean axial diusivity, and mean RD of AD patients and normal control
subjects.
Parameters AD patients Normal control subjects P value
(±SD and range) (±SD and range)
Fraction anisotropy (FA) 0.216 ± 0.035 0.218 ± 0.027 0.71
(0.181-0.251) (0.191-0.245)
Mean diffusivity 1.396 ± 0.195 1.178 ± 0.137 <0.0001
(MD, 10
-3
mm
2
/s) (1.201-1.591) (1.041-1.315)
Axial diffusivity 1.659 ± 0.194 1.459 ± 0.263 <0.0001
(AD, 10
-3
mm
2
/s ) (1.465-1.853) (1.196-1.722)
Radial diffusivity 1.267 ± 0.200 1.064 ± 0.143 <0.0001
(RD, 10
-3
mm
2
/s) (1.067-1.467) (0.921-1.207)
in eighteen patients (94.7%), and 2 points in one patient
(5.3%). Mean Scheltens score in AD group was signicantly
higher than that in control group (p < 0.001). (Table 1)
FA
e mean FA at posterior cingulate gyrus in AD
group was slightly lower than that in control group
(0.216 ± 0.035 vs 0.218 ± 0.027, respectively). However,
there was no statistically signicant dierence between
the two groups (p-value is 0.71). (Table 2)
MD
MD at posterior cingulate gyrus in AD group (1.396
± 0.195 x 10
-3
mm
2
/s) was statistically signicantly higher
than that in control group (1.178 ± 0.137 x 10
-3
mm
2
/s),
p value <0.001. (Table 2)
Axial diusivity
Axial diusivity at posterior cingulate gyrus in AD
group (1.659 ± 0.194 x 10
-3
mm
2
/s) was signicantly
higher than that in control group (1.459 ± 0.263 x 10
-3
mm
2
/s), p value <0.001. (Table 2)
Radial diusivity (RD)
RD at posterior cingulate gyrus in AD group (1.267
± 0.200 x 10
-3
mm
2
/s) was signicantly higher than that
in control group (1.064 ± 0.143 x 10
-3
mm
2
/s), p value
<0.001. (Table 2)
DISCUSSION
In early AD, atrophy of hippocampus and posterior
cingulate gyrus is a distinct imaging feature.
13
Several
studies
10-12
have found that the earliest hypometabolic (F18
FDG-PET) region in AD was the posterior cingulate cortex
which showed more signicantly decreased metabolism
than the mesial temporal structure.
Additional study revealed that there was decreased
FA coupled with increased diusivity (MD) in cingulum
ber which connects the hippocampus and posterior
cingulate gyrus reecting axonal loss or demyelinating
process.
16,21-25
Moreover, Salat D et al., revealed that diusion
measurement was related to indices of disease severity
and cognitive disability and specically associated with
episodic memory impairment. ese ndings represented
a potential clinical role for DTI to index white matter
degeneration and track AD symptoms.
26
Several previous studies
26-29,31,33
have widely used
post-processing analysis soware consisting of FSL for
DTI analysis at posterior cingulate gyrus. All of these
studies supported association of abnormal DTI value at
posterior cingulate cortex with early AD. In our study, there
was decreased FA with increased mean diusivity (MD),
axial and radial diusivity in AD patients compared with
control subjects. However, only MD, axial diusivity and
RD were signicantly statistically dierent between the
two groups. Our result was consistent with the studies by
Yoshiura et al., Nakata, et al., and Fellgiebel, et al., which
found that increased mean, axial and radial diusivity
in posterior cingulate correlated with AD.
Yoshiura T et al.,
14
found that mean axial and radial
diusivity, but not FA, in the posterior cingulate white
matter correlated with MMSE which reect progression
of AD-related histopathological changes. Fellgiebel A et al.,
15
Ngamsombat et al.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
121
Original Article
SMJ
found signicantly decreased FA and increased MD at
posterior cingulate white matter in AD compared to
normal controls. Nakata Y et al.,
27
found that MD in
the posterior cingulum signicantly correlated with
the MMSE score. However, no signicant correlation
was seen between FA and MMSE score. e measuring
methods were probably dierent in detail in each study,
such as the method of ROI measurement, and also DTI
scanning protocol that probably eected the measurement
of DTI values.
For correlation with pathophysiology of disease,
explanation of the association of increased diusivity,
hypometabolism and atrophy at posterior cingulate gyrus
in AD patients was established. MD is a measurement of
translational diusion which increases in the presence of
tissue damage.
30
erefore, MD is expected to increase
at the posterior cingulate gyrus in AD patients which
indicates pathology in limbic system contributing to
memory loss.
In addition, the dissociation between radial diusivity
and axial diusivity was concordant with the previous study
by Song SK et al.,
31-32
which pointed out that signicantly
increased radial diusivity reected myelin injury in AD.
However, increased axial diusivity was not consistent
with axonal injury in AD.
In the aspect of FA, there was no signicant dierence
between AD patients and normal control subjects in our
study. e FA demonstrated directionality of diusion
and in the white matter, high FA can be found in highly
organized tissue with parallel structure. Damage to white
matter disrupts the organized structure leading to a
decrease in FA. ree decreased FA regions involving
medial temporal lobe, temporal lobe proper, and posterior
cingulate gyrus were reported to be relatively consistent
with AD.
30
FA in our study was not signicantly lower in the
AD group which was similar to a study by Yoshiura
et al. is can be contributed to rotationally variant
during three orthogonal directions measurement of
FA. erefore, the relative orientation of the patient’s
head concerning the xed geometry of the MR imaging
system gradients can aect FA parameters.
14
A Study by Nakata et al., in 2008 using tract-specic
analysis of posterior cingulate ber tracts showed signicant
lower FA in AD patients.
28
However, another study by
Nakata et al., in 2009
27
showed no signicant correlation
between FA and MMSE which used MMSE as an indicator
of disease progression in AD. His conclusion suggested
that MD in the posterior cingulum is a more sensitive
indicator of progression of AD than FA at the posterior
cingulum and hippocampal volume.
All of the aforementioned studies reected the
variability of FA and implied that FA is not a reproducible
parameter. us, factors that can alter FA values are
detectable by DTI technique such as parameter, ROI
placement,
26,33
voxel size and number of collinear gradients.
Limitations of our study
1. Abnormal DTI ndings could be related to other
pathology such as dementia other than AD and/or
cerebral vascular disease due to non specic in nature
of DTI metrics.
2. Mean age at scanning of control group were lower
than the AD group. e age-related degeneration of
white matter may aect the DTI parameters.
CONCLUSION
e microstructural change at the posterior cingulate
gyrus demonstrated by DTI parameters including MD,
AD and RD were signicantly dierent between AD
patients and normal control subjects. Our findings
suggested that DTI parameters including MD, AD and
RD at posterior cingulate gyrus are non-invasive markers
of AD pathology. erefore, these are probably helpful
for early diagnosis, evaluation, and follow-up of AD
patients to correlate with clinical ndings. For further
studies, advanced imaging techniques that can dierentiate
between AD and mild cognitive impairment (MCI) will
be helpful in clinical practice.
ACKNOWLEDGMENT
e authors thank Mrs. Angkana Jongsawaddipatana
and all sta members of the primary care unit of Siriraj
Hospital for their contribution to data collection.
REFERENCES
1. Hong YJ,Yoon B,Shim YS,Cho AH,Lim SC,Ahn KJ,Yang
DW. Dierences in microstructural Alterations of the hippocampus
in Alzheimer disease and idiopathic normal pressure hydrocephalus:
A diusion tensor imaging study. AJNR Am J Neuroradiol
2010;31:1867-72.
2. Alzheimer’s Association. 2012 Alzheimer’s disease facts and
gures. Alzheimers Dement 2012;8:131-68.
3. Kumman P, Srijittra B. Alzheimer’s disease: Age or disease,
Bangkok, 2000.
4. Feldman HH, Jacova C, Robillard A Garcia A, Chow T, Borrie
M, et al . Diagnosis and treatment of dementia: 2.Diagnosis.
CMAJ 2008;178: 825-36.
5. Alves L, Correia AS, Miguel R, Alegria P, Bugalho P. Alzheimer’s
disease: a clinical practice-oriented review. Front Neurol
2012;3:63.
6. Trojanowski JQ, Vandeerstichele H, Korecka M, Clark CM,
Aisen PS, Petersen RC, et al. Update on the biomarker core of
the Alzheimer’s disease neuroimaging initiative subjects.
Alzheimers Dement 2010;6:230-8.
Volume 71, No.1: 2019 Siriraj Medical Journal
www.smj.si.mahidol.ac.th
122
7. Hattori T1, Sato R, Aoki S, Yuasa T, Mizusawa H. Dierent
patterns of fornix damage in idiopathic normal pressure
hydrocephalus and Alzheimer disease. AJNR Am J Neuroradiol
2012;33:274-9.
8. Daouk J, Chaarani B, Zmudka J, Capel C, Fichten A, Bouzerar
R, et al. Relationship between cerebrospinal uid ow, ventricles
morphology, and DTI properties in internal capsules: Dierences
between Alzheimer’s disease and normal-pressure hydrocephalus.
Acta Radiol 2014;55:992-9.
9. Kilimann I1, Likitjaroen Y, Hampel H, Teipel S. Diusion
tensor imaging to determine eects of antidementive treatment
on cerebral structural connectivity in Alzheimer’s disease.
Curr Pharm Des 2013;19:6416-25.
10. Pengas G, Hodges JR, Watson P, Nestor PJ. Focal posterior
cingulate atrophy in incipient Alzheimer’s disease. Neurobiology
Aging 2010;31:25-33.
11. Minoshima S, Giordani B, Berent S, Frey KA, Foster NL, Kuhl
DE. Metabolic reduction in the posterior cingulate cortex in
very early Alzheimer’s disease. Ann Neurol 1997;42(1):85-94.
12. Nestor PJ, Fryer TD,Smielewski P,Hodges JR. Limbic
hypometabolism in Alzheimer’s disease and mild cognitive
impairment. Ann Neurol 2003;54:343-51.
13. Nestor PJ,Fryer TD,Hodges JR. Declarative memory impairments
in Alzheimer’s disease and semantic dementia. Neuroimage
2006;30:1010-20.
14. Yoshiura T, Mihara F, Ogomori K, Tanaka A, Kaneko K, Masuda
K. Diusion tensor in posterior cingulate gyrus: correlation with
cognitive decline in Alzheimer’s disease. Neuroreport 2002;13:
2299-302.
15. Fellgiebel A, Muller MJ, Wille P, Dellani PR, Scheurich A,
Schmidt LG, et al. Color-coded diusion-tensor-imaging
of posterior cingulate ber tracts in mild cognitive impairment.
Neurobiol Aging 2005;26:1193-8.
16. Takahashi S, Yonezawa H, Takahashi J, Kudo M, Inoue T,
Tohgi H. Selective reduction of diusion anisotropy in white
matter of Alzheimer disease brains measured by 3.0 Tesla
magnetic resonance imaging. Neurosci Lett 2002;332:45-48.
17. American Psychiatric Association: Diagnostic and Statistical
Manual of Mental Disorders, Fourth Edition, Text Revision.
Washington, DC, American Psychiatric Association, 2000.p.
157-8.
18. McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack
CR Jr, Kawas CH, et al. e diagnosis of dementia due to
Alzheimer’s disease: recommendations from the National
Institute on Aging-Alzheimer’s Association workgroups
on diagnostic guidelines for Alzheimer’s disease. Alzheimers
Dement2011;7:263-9.
19. Scheltens P,Leys D,Barkhof F,Huglo D,Weinstein HC,Vermersch
P,et al. Atrophy of medial temporal lobes on MRI in Probable
Alzheimer’s disease and normal ageing: diagnostic clue and
neuropsychological correlates. J Neurol Neurosurg Psychiatry
1992;55:967-72.
20. Jenkinson M,Beckmann CF,Behrens TE,Woolrich MW,Smith
SM. FSL. Neuroimage 2012;62:782-90.
21. Zhang Y,Schu N,Jahng GH,Bayne W,Mori S,Schad L,
et al. Diusion tensor imaging of cingulum bers in mild
cognitive impairment and Alzheimer disease, Neurology
2007;68:13-19.
22. Basser PJ, Pierpaoli C. Microstructural and physiological
features of tissues elucidated by quantitative-diusion-tensor
MRI. J Magn Reson B 1996;111:209-19.
23. Sullivan EV, Pfeerbaum A. Diusion tensor imaging in normal
aging and neuropsychiatric disorders. Eur J Radiol 2003;45:
244-55.
24. Wakana S, Jiang H, Nagae-Poetscher LM, van Zijl PC, Mori S.
Fiber tract-based atlas of human white matter anatomy. Radiology
2004;230:77-87.
25. Catani M, Howard RJ, Pajevic S, Jones DK. Virtual in vivo
interactive dissection of white matter fasciculi in the human
brain. Neuroimage 2002;17:77-94.
26. Salat DH, Tuch DS, van der Kowe AJW, Greve DN, Pappu
V, Lee SY, et al. White matter pathology isolates the hippocampal
formation in Alzheimer’s disease. Neurobiol Aging 2010;31:
244-56.
27. Nakata Y, Sato N, Nemoto K, Abe O, Shikakura S, Arima K,
et al. Diusion abnormality in the posterior cingulum and
hippocampal volume: correlation with disease progression in
Alzheimer’s disease. Magn Reson Imaging 2009;27:347-54.
28. Nakata Y, Sato N, Abe O, Shikakura S, Arima K, Furuta N.
Diusion abnormality in posterior cingulate ber tracts in
Alzheimer’s disease: tract-specific analysis. Radiat Med
2008;26:466-73.
29. Zhou Y, Dougherty JH Jr, Hubner KF, Bai B, Cannon RL,
Hutson RK. Abnormal connectivity in the posterior cingulate
and hippocampus in early Alzheimer’s disease and mild cognitive
impairment. Alzheimers Dement 2008;4:265-70.
30. Stebbins GT,Murphy CM. Diusion tensor imaging in Alzheimer’s
disease and mild cognitive impairment. Behav Neurol 2009;21:
39-49.
31. Song SK, Sun SW, Ramsbottom MJ, Chang C, Russell J, Cross
AH. Dysmyelination revealed through MRI as increased radial
(but unchanged axial) diusion of water. Neuroimage 2002;17:
1429-36.
32. Song SK, Sun SW, Ju WK, Lin SJ, Cross AH, Neueld AH.
Diffusion tensor imaging detects and differentiates axon
and myelin degeration in mouse optic nerve after retinal
ischemia. Neuroimage 2003;20:1714-22.
33. Kalus P, Slotboom J, Gallinat J, Mahlber R, Cattapan-Ludewig
K, Weist R, et al. Examining the gateway to the limbic system
with diusion tensor imaging: the perforant pathway in dementia.
Neuroimage2006;30:713-20.
Ngamsombat et al.
