Volume 71, Number 4, July-August 2019

Siriraj Medical Journal

SMJ

ISSN 2629-995X

ORIGINAL ARTICLE

253

Agreement between Emergency Physicians and a Cardiologist

on Cardiac Function Evaluation after Short Training

Apichaya Monsomboon, et al.

261 Incidence and Risk Factors of Necrotizing Enterocolitis

Following Gastroschisis Repair in Correlation with Modes

of Abdominal Wall Closure and Umbilical Management at

Siriraj Hospital: an 11-Year Retrospective Review

Mongkol Laohapensang, et al.

268

The Relationship among A Palliative Care Service, Patient’s

Factors, and Quality of Life of Post Treatment Cervical

Cancer Patients: a Causal Model Approach

Malee Ngamprasert, et al.

278 The Accuracy of Cervical Length for Prediction of Delivery

in Term Pregnancy Patients Presenting with Labor Pain

Anantachai Jaisaby, et al.

284 Diagnostic Performance of Short MR-Neurography

Protocol for Brachial Plexus Injuries

Siriwan Piyapittayanan, et al.

290 The Determinants of Quality of Life in Thai Family

Caregivers of Stroke Survivors

Ruttanaporn Kongkar, et al.

297 Clinical Characteristics and Prognosis of Morphea

(Localized Scleroderma) in Adults: a Retrospective Study

Narumol Silpa-archa, et al.

302 A Two-Year Outcome of Intrastromal Corneal Ring

Segment Implantation in Keratoconus: Initial Report in

Thai Patients

Suksri Chotikavanich, et al.

310 The Prevalence and Factors Associated with Mistreatment

Perception among Thai Medical Students in a Southern

Medical School

Jarurin Pitanupong, et al.

CASE REPORT

318 Surgical Treatment of Myopic Strabismus Fixus by Loop

Myopexy Augmented with Scleral Fixation: a Case Report

Pittaya Phamonvaechavan, et al.

LETTER TO EDITOR

322 Pseudohyperkalemia Caused by EDTA Contamination:

a Not Uncommon Pre-analytical Error

Sathima Laiwejpithaya, 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 Siriwan Piyapittayanan, et al.

SIRIRAJ MEDICAL JOURNAL

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

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

Somboon Kunathikom

Executive Editor: Prasit Watanapa Editorial Director: Manee Rattanachaiyanont

Managing Editor: Gulapar Srisawasdi, Chenchit Chayachinda

Editor-in-Chief: awatchai Akaraviputh

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

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

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

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

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

Philip Board (Australian National University, Australia)

Richard J. Deckelbaum (Columbia University, USA)

Yozo Miyake (Aichi Medical University, Japan)

Yik Ying Teo (National University of Singapore, Singapore)

Harland Winter (Massachusetts General Hospital, USA)

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

Noritaka Isogai (Kinki University, Japan)

Yuji Murata (Aizenbashi Hospital, Japan)

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

Shuji Shimizu (Kyushu University Hospital, Japan)

David S. Sheps (University of Florida, USA)

Robin CN Williamson (Royal Postgraduate Medical School, UK)

Tai-Soon Yong (Yonsei University, Korea)

Anusak Yiengpruksawan (e Valley Robotic Institute, USA)

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

Kyoichi Takaori (Kyoto University Hospital, Japan)

Tomohisa Uchida (Oita University, Japan)

Yoshiki Hirooka (Nagoya University Hospital, Japan)

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

Kazuo Hara (Aichi Cancer Center Hospital, Japan)

Shomei Ryozawa (Saitama Medical University, Japan)

Christopher Khor (Singapore General Hospital, Singapore)

Yasushi Sano (Director of Gastrointestinal Center, Japan)

Mitsuhiro Kida (Kitasato University & Hospital, Japan)

Seigo Kitano (Oita University, Japan)

Ichizo Nishino (National Institute of Neuroscience NCNP, Japan)

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

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

George S. Baillie (University of Glasgow, UK)

G. Allen Finley (Delhousie University, Canada)

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

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

Marianne Hokland (University of Aarhus, Denmark)

Marcela Hermoso Ramello (University of Chile, Chile)

Ciro Isidoro (University of Novara, Italy)

Moses Rodriguez (Mayo Clinic, USA)

Robert W. Mann (University of Hawaii, USA)

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

Frans Laurens Moll (University Medical Center Ultrecht, Netherlands)

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

John Hunter (Oregon Health & Science University, USA)

Nima Rezaei (Tehran University of Medical Sciences, Iran)

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

Folker Meyer (Argonne National Laboratory, USA)

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

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

Victor Manuel Charoenrook de la Fuente 

(Centro de Oalmologia Barraquer, Spain)

Karl omas Moritz

(Swedish University of Agricultural Sciences, Sweden)

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

www.smj.si.mahidol.ac.th

SMJ

Statistician: Saowalak Hunnangkul (Mahidol University, ailand)

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

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

International Editorial Board

Editorial Board

Watchara Kasinrerk (Chiang Mai University, ailand)

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

Wiroon Laupattrakasem (Khon Kaen University, ailand)

Anuwat Pongkunakorn (Lampang Hospital, ailand)

Nopporn Sittisombut (Chiang Mai University, ailand)

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

Yuen Tanniradorm (Chulalongkorn University, ailand)

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

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

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

Jaturat Kanpittaya (Khon Kaen University, ailand)

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

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

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

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

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

Suttipong Wacharasindhu (Chulalongkorn University, ailand)

Prapon Wilairat (Mahidol University, ailand)

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

Ampaiwan Chuansumrit

(Ramathibodi Hospital, Mahidol University, ailand)

Sayomporn Sirinavin

(Ramathibodi Hospital, Mahidol University, ailand)

Vitoon Chinswangwatanakul

(Siriraj Hospital, Mahidol University, ailand)

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

253

Original Article

SMJ

Apichaya Monsomboon, M.D.*, iti Patarateeranon, M.D.**, Surat Tongyoo, M.D.***, Usapan Surabenjawong,

M.D.*, Wansiri Chaisirin, M.D.*, Tipa Chakorn, M.D.*, Nattakarn Praphruetkit, M.D.*, Onlak Ruangsomboon,

M.D.*, Tanyaporn Nakornchai, M.D.*

*Department of Emergency Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, **Emergency Medicine Unit, Taksin

Hospital, Medical Service Department, Bangkok 10600, ***Department of Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok

10700, ailand.

Agreement between Emergency Physicians and

a Cardiologist on Cardiac Function Evaluation

after Short Training

ABSTRACT

Objective: Delayed diagnosis and treatment of shock patients may lead to multiorgan dysfunction syndrome and death.

Volume status assessment in shock patients is crucial for guiding early management. Focused echocardiography has

become an important tool for assessing volume status because it is non-invasive and easy to perform. We aimed to

ascertain the degree of agreement between emergency medicine (EM) residents and a cardiologist on cardiac function

evaluations using echocardiography. We also assessed the extent of agreement on pericardial eusion diagnoses.

Methods: A cross sectional study was conducted at the Emergency Department, Siriraj Hospital. e EM residents

who had limited experience in ultrasound examination underwent a 3-hour echocardiography training course

consisting of a lecture and a workshop before starting the study. Patients with shock or suspected hypervolemia

were included. Echocardiography was performed by EM residents to evaluate ventricular function of each patients.

With visual estimation, they classied the le ventricular function (LVF) into 3 categories: good, moderate and poor.

e video les were recorded and re-evaluated by a cardiologist oine. e correlation of le ventricular function

estimation and the diagnosis of pericardial eusion between the two operators were determined.

Results: Ninety-two patients were enrolled between October and December 2014. e overall agreement of ventricular

function assessment between the EM residents and the cardiologist was 79.4% (weighted kappa = 0.73). e degree

of agreements of LVF classied as poor, moderate and good LVF were 87.5%, 37.5% and 95% respectively. Moreover,

the residents diagnosed the pericardial eusion with 100% accuracy, compared to the cardiologist.

Conclusion: Following a short educational training, the EM residents eciently assessed the le ventricular function

with a high level of agreement with a cardiologist.

Keywords: Cardiac function evaluation; limited echocardiography; volume assessment (Siriraj Med J 2019; 71: 253-260)

INTRODUCTION

Delayed diagnosis and treatment of shock can lead

to multi-organ dysfunction syndrome and possibly death.

In general, shock can be classied into four categories

hypovolemic, distributive, cardiogenic, and obstructive.

e dierent types of shock need dierent approaches to

the management of volume resuscitation. Most types of

shock improve with uid bolus except cardiogenic shock,

for which uid bolus may be harmful. erefore, a le

ventricular function (LVF) assessment is one of the important

Corresponding author: Tanyaporn Nakornchai

E-mail: tanyaporn.ploy@gmail.com

Received 12 October 2018 Revised 19 April 2019 Accepted 14 May 2019

ORCID ID: http://orcid.org/0000-0003-1513-2818

http://dx.doi.org/10.33192/Smj.2019.39

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

254

Monsomboon et al.

initial steps in deciding whether to give uid. Cardiogenic

shock should be considered if the LVF is poor, but uid

should not be given for that condition. Furthermore,

performing an accurate volume status assessment is one

of the critical steps in eectively and safely managing

shock, and it results in a decrease in mortality.

2,3

However,

volume assessment using only a physical examination

has some limitations, for example, in overweight patients

or when it is performed by inexperienced physicians.

Moreover, a physical examination alone is inadequate

for dierentiating hypovolemic from cardiogenic shock.

A chest radiograph is another measure used to evaluate

volume status; however, it might not be feasible for use

with some overweight patients or those who cannot

inspire deeply. Even when interpreted by a radiologist,

the sensitivity of chest radiographs has been found to be

only 53%, with an accuracy of only 68% for diagnosing

acute congestive heart failure.

Nowadays, point-of-care ultrasound (PoCUS)

has become an important tool for assessing the LVF

and volume status because it is non-invasive, is easy

to perform, and can be used to perform a dynamic

assessment of uid responsiveness in critically-ill patients.

Moreover, the combination of PoCUS applications, which

are inferior vena cava variation, lung ultrasound, and

focused echocardiography, may enable physicians to give

more accurate uid therapy.

Many studies have shown

that echocardiography is comparable to other invasive

methods, such as central venous catheterizations and

pulmonary artery catheters, in terms of cardiac function

assessment and uid treatment guidance.

9,10

Furthermore,

focused echocardiography performed within 15 minutes

of Emergency Department (ED) arrival signicantly

reduces errors in diagnosing shock patients.

Previous studies have shown that emergency

physicians (EPs) have been able to assess the LVF by

visual estimation with a high degree of agreement with

cardiologists (> 80%).

12,13

However, the investigators

in those studies were experienced EPs or emergency

medicine (EM) residents who had used PoCUS for at

least 150 cases each. At our institution, which is the

largest tertiary hospital in ailand, EM is still a relatively

new department, having been established for less than

10 years. Although the usage of ultrasound in the ED is

becoming more common, the number of experienced

operators is limited.

This study aimed to evaluate the ability of EM

residents with limited experience to evaluate the LVF using

echocardiography. To this end, we aimed to ascertain

the level of agreement between the EM residents and a

cardiologist on cardiac function estimations. Furthermore,

we assessed their agreement on pericardial eusion

diagnoses as well as evaluated the appropriateness of

the echocardiographic views and the time taken for the

residents to complete the scans.

MATERIALS AND METHODS

A cross sectional study was conducted between

October and December 2014 at the Emergency Department,

Siriraj Hospital. e hospital, located in Bangkok, is

the largest tertiary care university hospital in ailand.

We enrolled patients aged over 18 years who needed a

volume status assessment. e conditions with which

the patients would require an assessment of the LV

function were 1) shock (systolic blood pressure < 90

mmHg or diastolic blood pressure < 60 mmHg); and

2) suspected heart failure, dened as having at least 1

of the Framingham criteria.

Patients were excluded if

they were uncooperative or had an ST-segment elevation

myocardial infarction. Intubated patients were excluded

because of the diculty of placing the patients in an

appropriate position for scanning. Patients for whom

focused echocardiography might delay their standard

treatment were also excluded. is study was approved

by Siriraj Institutional Review Board (Si 546/2014) and

complied with the Code of Ethics of the World Medical

Association.

e study investigators who performed the ultrasounds

in this study were second- and third-year EM residents.

e residents had been working in the ED for less than 3

years. ey had not participated in any formal ultrasound

training courses because emergency ultrasound was not

part of the EM curriculum.

We arranged our study into two phases: training

and data collection. In the training phase, we provided

a focused echocardiography course for the residents.

is course consisted of a 2-hour lecture and a 1-hour

hands-on workshop conducted by a cardiologist. Aer

the course, each participant had to practice on at least

two patients in the ED. All recorded images and video

files of the practice were subsequently sent to the

cardiologist to assess the residents’ ability to perform

focused echocardiography. e cardiologist had to approve

each participant’s performance before he or she was

permitted to proceed to the next phase.

During the data collection phase, the attending EP

sta assessed the eligibility of patients. If the patients

met the inclusion criteria, written, informed consent

was obtained. ey then notied a study investigator

(i.e., one of the 2

or 3

year EM residents who had

attended the training course) who was not the primary

treating physician of the patients to perform focused

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

255

Original Article

SMJ

echocardiography. e patients were scanned in the

supine position for as long as they could tolerate. A

Philips HD15 PureWave ultrasound machine was used in

this study. e LVF was assessed using parasternal long

axis, parasternal short axis, apical 4-chamber, and apical

5-chamber views with an S5-2 phased array transducer

(5-1 MHz). e LVF was categorized into three groups,

based on the le ventricular ejection fraction (LVEF):

normal/good (LVEF > 55%), moderate (LVEF 30%-

55%) and poor (LVEF < 30%). Pericardial effusion

was also assessed using the subxiphoid view with a

C 5-1 convex array transducer (5-1 MHz). e patients’

baseline characteristics and details of the medications

or inotropic agents administered to the patients in the

ED were also recorded. All images and video les of the

focused echocardiographic ndings were sent to the

cardiologist to assess the LVF and appropriateness of the

echocardiographic views. e cardiologist was blinded

to the patients’ clinical information, the identity of the

study investigators, and their results.

Sample size calculation

From a previous study by Randazzo

, the kappa

statistic of agreement between EPs and cardiologists on

cardiac function evaluation was 0.71. With a condence

interval of 95% and a margin of error of 5%, a sample

size of 88 patients was required (nQuery Advisor). Aer

adding 10% to adjust for potential missing data, the nal

total sample size was 97.

Statistical analysis

We used SPSS Statistics for Windows, version

18.0 (SPSS Inc., Chicago, Ill., USA) to analyze the data.

Frequency and percentage were used to describe the

categorical data. Mean and standard deviation were used

to describe continuous data with normal distribution.

Median and interquartile range were used to describe data

with non-normal distribution. e kappa statistic and

95% condence interval were used to assess the degree

of agreement between the EPs and the cardiologist on

the LVF evaluations. Fisher’s exact test was used to nd

the factors that aected the correlation.

RESULTS

A total of 101 patients were assessed for eligibility

between October and December 2014. Of those, 9 (8.9%)

were excluded due to incomplete data, leaving 92 patients.

Forty-one patients (44.6%) presented with hypotension,

and the remaining 51 patients (55.4%) were suspected to

have hypervolemia based on the Framingham criteria. e

patients’ mean age was 67 years, and the mean systolic

blood pressure was 108 mmHg. e baseline characteristics

are shown in Table 1. e focused echocardiography was

performed by four second-year and three third-year EM

residents.

The residents classified the patients’ LVF as

“good contraction” in 60 patients (65.2%), “moderate

contraction” in 24 (26.1%), and “poor contraction” in 8

(8.7%). e cardiologist classied the patients as having

good, moderate, and poor contraction in 69 (75%),

12 (13%), and 11 (12%) patients, respectively (Table 2).

e overall agreement on the LVF assessments by the

EM residents and the cardiologist was 79.4%, with

a weighted kappa of 0.73 (95% CI 0.58-0.89). e patients

with a moderate contraction represented the majority

in the disagreement category (Fig 1). No factors aected

the correlation of the cardiac function evaluations. e

studied factors included a BMI of more than 23 kg/m

and inotropic drug use. e agreement between the EM

residents and the cardiologist on the pericardial eusion

diagnoses was 100%.

e most appropriately performed views were the

subxiphoid and inferior vena cava views (94.6%). e

least properly performed view was the apical 5-chamber

(Fig 2). The EM residents were able to achieve all

appropriate views in 47 patients (51.1%; Fig 3). None of

the following factors had an eect on the appropriateness

of the echocardiographic views: being overweight, the

operators, and patients having congestive heart failure.

e nal diagnoses are detailed in Table 3. e

most common cause of shock was sepsis or septic shock

(92.6%). e lowest level of agreement on LVF evaluation

was found in patients with pneumonia (42.9%); however,

there was no statistically signicant dierence from

other diagnoses. e mean time taken for the residents

to perform the focused echocardiographies was 19.8+7.1

minutes.

DISCUSSION

In our study, EM residents with limited experience

in focused echocardiography were able to assess the LVF

by visual estimation alone with a high level of agreement

with a cardiologist (weighted kappa = 0.73) and aer

only a short training course. is result is similar to the

ndings of previous studies, which also reported a high

level of agreement between experienced and inexperienced

scanners.

9,12

We categorized the LVFs into three groups using

visual estimation because this method was easy and

not time-consuming. Moreover, previous studies have

shown no clinical dierence between LVF assessments

by visual estimation and other complex methods.

15,16

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

256

TABLE 1. Baseline characteristics.

Baseline characteristics N = 92 (%)

Age (mean ± SD; years) 68 ± 14

Sex (males) 39 (42.4)

BMI

†

≥ 23 kg/m

36 (39.1)

Comorbidities

Hypertension 50 (54.3)

Diabetes mellitus 34 (37.0)

Heart disease 26 (28.3)

Others 9 (9.8)

SBP‡ (mean ± SD; mmHg) 108 ± 30 (50–190)

HR‡ (mean ± SD; bpm) 90 ± 21 (49–153)

Inotropic drugs

Dopamine 2 (2.2)

Norepinephrine 6 (6.5)

Indication

Hypotension 41 (44.6)

Suspected hypervolemia 51 (55.4)

Performer

Resident 2 27 (29.3)

Resident 3 65 (70.7)

†BMI: Body Mass Index; BMI ≥ 23 kg/m

represents being overweight

‡while performing focused echocardiography

TABLE 2. LVF categorizations by EM residents and cardiologist.

Cardiologist

Emergency Physicians Good contraction Moderate contraction Poor contraction

no. (%) no. (%) no. (%)

Good contraction no. (%) 57 (95.0) 3 (5.0) 0 (0)

Moderate contraction no. (%) 11 (45.8) 9 (37.5) 4 (16.7)

Poor contraction no. (%) 1 (12.5) 0 (0) 7 (87.5)

Monsomboon et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

257

Original Article

SMJ

TABLE 3. Final diagnoses and correlations for LVF evaluations.

Final Diagnosis Correlation (%) No correlation (%)

Hypotension (n = 45)

Severe sepsis/septic shock 26 (92.9) 2 (7.1)

Hypovolemic shock 5 (62.5) 3 (37.5)

Cardiogenic shock 2 (100.0) 0

Unknown 7 (100.0) 0

Suspected hypervolemia (n = 47)

Congestive heart failure 28 (77.8) 8 (28.6)

Pneumonia 3 (42.9) 4 (57.1)

Others 3 (75.0) 1 (25.0)

Fig 1. Percentage of correlation between EM residents and cardiologist in LVF assessment.

Fig 2. Appropriateness of each echocardiographic view.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

258

Fig 3. Percentage of appropriateness of echocardiographic views.

e credentials for performing PoCUS dier from

country to country. In the United States, the American

College of Emergency Physicians requires a physician to

perform PoCUS in at least 25 cases in order to be qualied.

In contrast, the British Society of Echocardiography

requires at least 250 scans and a written examination to

be passed for emergency physicians who want to perform

echocardiography.

In Australia, the Australasian College

for Emergency Medicine requires 50 examinations of

bedside echocardiography and a pass in a practical bedside

examination.

In ailand, the Emergency Medicine

Residency training program was established in 2004;

to date, however, no specic requirement regarding

PoCUS training has been included in the curriculum. As

a result, the EM residents in our study had various levels

of experience in focused echocardiography. Accordingly,

our results might indicate that aer proper training,

even performers with limited experience in PoCUS

are able to assess the LVF with comparable accuracy to

a cardiologist.

e duration of the didactic sessions and practice

has varied from study to study, ranging from web-based

learning to 12-hour, didactic lectures.

12,13,20,21

Similarly,

the practice session duration has ranged from 5 scans of

real patients to a 10-hour observation and practice.

12,13,20,21

In our study, the training session comprised only 2

hours of lecture and 1 hour of practice. erefore, our

ndings suggest that a short training session could also

enable physicians with limited PoCUS skills to achieve

a high level of agreement with an expert. However, it

might have been because the EM residents in our study

were not blinded to the patients’ clinical presentations;

the general appearance of the patients might therefore

have biased the residents’ echocardiographic ndings.

e best correlation for the LVF evaluations was seen

in the good contraction group. is nding was also similar

to those of previous studies, which demonstrated that

the strongest correlation was found in the good and poor

contraction groups.

12,13

However, the EPs in the current

study could not evaluate the LVF as accurately in patients

with a moderate contraction. is was also concordant

with previous studies and might have been due to the less

apparent ndings. Nonetheless, echocardiography might

help EPs to identify the cause of shock more rapidly,

which could lead to a faster provision of appropriate

treatment.

Furthermore, the accuracy of the pericardial eusion

diagnoses was 100%, which was as high as that found

by a previous study.

is might have been because

our study investigators were familiar with the FAST

(Focused Assessment with Sonography in Trauma) exam,

which uses the subxiphoid view to evaluate pericardial

eusion. e subxiphoid view was thus the most properly

performed view by the EPs in this study. However, this

nding was discordant with some previous studies,

20,23

which reported that the parasternal short and long axes

were the most properly performed views. is might have

been because those studies included performers who had

some experience in focused echocardiography, making

it easy for them to acquire the proper parasternal short

and long axis views. In comparison, the subxiphoid

view might be the easiest view to achieve in the case

of inexperienced performers. Furthermore, the apical

5-chamber view was the least properly performed view

in the present study, which might have been due to the

need both cooperation from the patient and a skillful

operator. Nevertheless, as the apical 5-chamber view is

employed to evaluate cardiac output, we should improve

providers’ skill performing it.

Monsomboon et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

259

Original Article

SMJ

The mean time taken to perform the focused

echocardiographies was 19 minutes, which was comparable

to the times reported by previous studies, which had

mean durations ranging from 5 to 25 minutes.

9,11,12,24

Many factors can aect the scan time. e rst inuential

factor is the performers’ experience. If the investigators

in the present study had performed many scans, they

would have had more experience and would therefore

have been able to complete the scans faster than the

more inexperienced scanners.

In addition, the apical

5-chamber view, which is dicult to achieve, might

have caused an increase in the scan time. Other factors

that might have aected the scan time in the current

study were machine familiarity, the cooperation of the

patients, and the patients’ habitus; however, those data

were not collected.

Our study had some limitations. Firstly, the cardiologist

did not perform the echoes on real patients but instead

conducted a review of recorded pictures or video. is

might have led to some misinterpretations of the LVFs

due to inappropriate views. Moreover, no inter-rater

reliability between experts was obtained because there was

only one participating cardiologist. As well, we included

some patients who could not tolerate a supine position

for a long duration for various reasons. is could have

made their focused echocardiographies more dicult

and might have aected the appropriateness of their

echocardiographic views. Fourthly, some conclusions

could not be made due to the small number of patients

in the subgroups, such as patients with diagnoses other

than sepsis and patients with inotropic use. Finally, the

number of supervised cases was very small. erefore,

further studies with increased supervised practice and

real time feedback should be conducted in order to

achieve a higher correlation.

From this study, we conclude that emergency physicians

with only short training in focused echocardiography

can assess the LVF by visual estimation, and they are able

to achieve a high level of agreement with a cardiologist.

is might help emergency physicians to more quickly

and accurately identify the causes of shock, which in

turn should lead to the more rapid and appropriate

management of emergency patients.

ACKNOWLEDGMENTS

The authors thank Chulaluk Komoltri, MPH,

DrPH, statisticians, for data analysis as well as Onlak

Ruangsomboon, MD, for manuscript editing.

REFERENCES

1. Voyce SJ, McCaree DR. In: Irwin RS, ed. Intensive care

medicine. Lippincott Williams & Wilkins. Philadelphia.

2003.p.45-67.

2. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich

B, et al. Early goal-directed therapy in the treatment of severe

sepsis and septic shock. N Engl J Med 2001;345:1368-77.

3. Sebat F, Musthafa AA, Johnson D, Kramer AA, Shoner D,

Eliason M, et al. Eect of a rapid response system for patients

in shock on time to treatment and mortality during 5 years.

Crit Care Med 2007;35:2568-75.

4. Low D, Vlasschaert M, Novak K, Chee A, Ma IW. An argument

for using additional bedside tools, such as bedside ultrasound,

for volume status assessment in hospitalized medical patients:

a needs assessment survey. J Hosp Med 2014;9:727-30.

5. McGee S, Abernethy WB, Simel DL. Is this patient hypovolemic?

JAMA 1999;281:1022-29.

6. Mueller-Lenke N, Rudez J, Staub D, Laule‐Kilian K, Klima T,

Perruchoud AP, et al. Use of chest radiography in the emergency

diagnosis of acute congestive heart failure. Heart 2006;92(5):

695-6.

7. Evans D, Ferraioli G, Snellings J, Levitov A. Volume responsiveness

in critically ill patients: use of sonography to guide management.

J Ultrasound Med 2014;33(1):3-7.

8. Atkinson P, Bowra J, Milne J, Lewis D, Lambert M, Jarman B,

et al. International Federation for Emergency Medicine Consensus

Statement: sonography in hypotension and cardiac arrest

(SHoC): an international consensus on the use of point of

care ultrasound for undierentiated hypotension and during

cardiac arrest. CJEM 2017;19(6):459-70.

9. Gunst M, Ghaemmaghami V, Sperry J, Robinson M, O’Keee

T, Friese R, et al. Accuracy of cardiac function and volume

status estimates using the bedside echocardiographic assessment

in trauma/critical care. J Trauma 2008;65:509-16.

10. Kaul S, Stratienko AA, Pollock SG, Marieb MA, Keller MW,

Sabia PJ. Value of two-dimensional echocardiography for

determining the basis of hemodynamic compromise in critically

ill patients: a prospective study. J Am Soc Echocardiogr

1994;7:598-606.

11. Jones AE, Tayal VS, Sullivan DM, Kline JA. Randomized,

controlled trial of immediate versus delayed goal-directed

ultrasound to identify the cause of nontraumatic hypotension

in emergency department patients. Crit Care Med 2004:32:

1703-8.

12. Moore CL, Rose GA, Tayal VS, Sullivan DM, Arrowood JA,

Kline JA. Determination of le ventricular function by emergency

physician echocardiography of hypotensive patients. Acad

Emerg Med 2002;9:186-93.

13. Randazzo MR, Snoey ER, Levitt MA, Binder K. Accuracy of

emergency physician assessment of le ventricular ejection

fraction and central venous pressure using echocardiography.

Acad Emerg Med 2003;10:973-7.

14. Maestre A, Gil V, Gallego J, Aznar J, Mora A, MartínHidalgo

A. Diagnostic accuracy of clinical criteria for identifying

systolic and diastolic heart failure: crosssectional study. J Eval

Clin Pract 2009;15(1):55-61.

15. Amico AF, Lichtenberg GS, Reisner SA, Stone CK, Schwartz RC,

Meltzer RC. Superiority of visual versus computerized

echocardiographic estimation of radionuclide le ventricular

ejection fraction. Am Heart J 1989;118:1259-65.

16. Shahgaldi K, Dudmundsson P, Manouras A, Brodin L, Winter

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

260

R. Visually estimated ejection fraction by two dimensional

and triplane echocardiography is closely correlated with quantitative

ejection fraction by real-time three-dimensional echocardiography.

Cardiovasc Ultrasound 2009;7:41.

17. American College of Emergency Physicians. Emergency

ultrasound guidelines. Ann Emerg Med 2009;53:550-70.

18. British Society of Echocardiography. Accreditation. Available

at: https://www.bsecho.org/accreditation/types-of-accreditation/

(Cited February 14, 2018).

19. International Federation for Emergency Medicine (IFEM)

Point of Care Emergency Ultrasound Curriculum Guidelines;

2014. Available at: https://www.ifem.cc/wp-content/uploads/

2016/07/IFEM-Point-of-Care-Ultrasound-Curriculum-

Guidelines-2014.pdf (Cited February 14, 2018).

20. Chrisholm CB, Dodge WR, Balise RR, Williams SR,

Gharahbaghian L, Beraud AS. Focused cardiac ultrasound

training: how much is enough? J Emerg Med 2013;44:818-22.

21. Bustam A, Azhar MN, Veriah RS, Arumugam K, Loch A.

Performance of emergency physicians in point-of-care

echocardiography following limited training. Emerg Med J

2014;31:369-73.

22. Mandavia DP, Hoffner RJ, Mahaney K, Henderson SO.

Bedside echocardiography by Emergency Physicians. Ann

Emerg Med 2001;38:377-82.

23. Mark DG, Ku BS, Carr BG, Everett WW, Okusanya O,

Horan A, et al. Directed bedside transthoracic echocardiography:

preferred cardiac window for le ventricular ejection fraction

estimation in critically ill patients. Am J Emerg Med 2007;25:

894-900.

24. Volpicelli G, Lamorte A, Tullio M, Cardinale L, Giraudo M,

Stefanone V, et al. Point-of-care multiorgan ultrasonography

for the evaluation of undifferentiated hypotension in the

emergency department. Intensive Care Med 2013;39:1290-98.

Monsomboon et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

261

Original Article

SMJ

Mongkol Laohapensang, M.D.*, Duangkamol Puthakunraksa, M.D.**, Niramol Tantemsapya, M.D.*

*Department of Surgery, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, **Department of Surgery, Chareonkrung Pracharak

Hospital, Bangkok 10120, ailand.

Incidence and Risk Factors of Necrotizing

Enterocolitis Following Gastroschisis Repair in

Correlation with Modes of Abdominal Wall

Closure and Umbilical Management at Siriraj

Hospital: an 11-Year Retrospective Review

Corresponding author: Mongkol Laohapensang

E-mail: mongkol.lao@mahidol.ac.th

Received 6 February 2019 Revised 24 April 2019 Accepted 14 May 2019

ORCID ID: http://orcid.org/0000-0002-0774-5705

http://dx.doi.org/10.33192/Smj.2019.40

ABSTRACT

Objective: We investigated the correlation between umbilical management and NEC in infants with gastroschisis

as well as the incidence and potential risk factors of NEC in patients with gastroschisis at Siriraj Hospital from

2005 to 2016.

Methods: A retrospective chart review was conducted of patients with gastroschisis who were surgically repaired

at Siriraj Hospital from January 2005 to January 2016. e baseline characteristics, umbilical management, and

short-term outcomes were analyzed in relation to NEC complications to determine the associated correlations and

potential risk factors.

Results: Overall, 106 patients were enrolled. e incidence of NEC following gastroschisis repair was 16% (17/106).

Umbilical preservation was a signicant potential risk factor for NEC (p = 0.009; hazard ratio = 5.14; 95% CI =

1.51-17.42). ere were no signicant dierences between the NEC and non-NEC group for gender, median Apgar

scores, gestational ages, and birth weight. e short-term outcomes were signicantly higher for the NEC than the

non-NEC group, with a time to rst oral feeding of 15 vs. 9 days (p = 0.006), duration of total parenteral nutrition,

22 vs. 12 days (p < 0.001), and length of stay, 32 vs. 23 days (p = 0.01) respectively.

Conclusion: Umbilical preservation following gastroschisis repair was associated with a higher incidence of NEC,

even in term infants. us, NEC should be carefully monitored aer abdominal fascial closure with umbilical

preservation.

Keywords: NEC; gastroschisis; umbilical preservation (Siriraj Med J 2019; 71: 261-267)

INTRODUCTION

Gastroschisis is a congenital, abdominal-wall defect

with eviscerated abdominal content. Usually, the defect

is less than 4 cm in width and is located on the right of

the umbilicus. e incidence varies from 2-4.9 per 10,000

live births.

1–3

Necrotizing enterocolitis (NEC) is a well-

recognized and serious complication, particularly aer

gastroschisis repair, and is responsible for increased

morbidity and occasional mortality.

4–6

e incidence

of NEC has ranged from 4%–18.5% in infants with

gastroschisis following abdominal wall closure.

4–7

Previous studies have reported the association of

NEC in infants with gastroschisis with a lower birth

weight

5,8

and prematurity

, although the association is

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

262

unclear following the initiation of enteral feedings.

4–5

e predisposing factors for NEC in normal term infants

have been speculated to be the combination of reduced

mesenteric perfusion and feeding with articial formula.

e risk factors for NEC in infants with gastroschisis

have rarely been documented.

The current definitive surgical management of

gastroschisis is by early return of the eviscerated abdominal

content into the abdomen through the primary abdominal

fascia and skin closure.

3,4

If primary abdominal closure

is impossible due to visceral-abdominal disproportion or

increased intraabdominal hypertension, an articial pouch

or silo is placed with a subsequent serial reduction of the

extra-abdominal content

3,4

and a succeeding denitive

closure.

e dierence in umbilical cord management during

and after abdominal closure depends on individual

surgeons’ preferences. Several previous studies have

demonstrated the advantages of umbilical preservation

in cosmetic outcomes, and most did not report any

major complications such as cellulitis or omphalitis.

10–12

Conversely, research by Komuro et al. revealed a possible

association between umbilical preservation and wound

infection.

Concerns about periumbilical necrotizing fasciitis

led to umbilical cord and urachal remnant excision, as

reported by Kosloske and Bartow.

e infection was

speculated to spread along the umbilical arteries and vein

as well as the urachal remnant. However, the association

between umbilical preservation and NEC has not yet

been dened.

Objectives

e purpose of this study was to determine the

association between umbilical cord preservation and

NEC in infants with gastroschisis. e incidence and

risk factors of NEC in infants with gastroschisis were

also evaluated.

Keywords and variables in the study

Preterm refers to a baby born before 37 weeks of pregnancy

have been completed. Preterm infants experience a greater

risk of morbidity and mortality than term infants due

to immaturity-related complications.

Low Apgar score is dened as a score of less than 7 at

1 minute.

Low birth weight is dened as a birth weight of less than

1,500 grams.

Modes of surgical treatment for gastroschisis were

classied into:

• Primary fascial closure with umbilical preservation:

closure of the abdominal fascial wall, either immediately

or in the rst operation, with preservation of the

umbilical cord.

• Primary fascial closure with umbilical cord removal:

closure of the abdominal fascial wall, either immediately

or in the rst operation, with an umbilical cord

excision followed by an umbilicoplasty.

• Staged abdominal wall closure: a salvage technique

providing temporary intestinal coverage, which

allows for the gradual reduction of the contents,

followed by a denitive closure and an umbilical

cord excision.

Umbilical cord excision: the excision of the umbilical

cord from a newborn by cutting the umbilical arteries,

umbilical vein, and urachal remnant at the level of the

peritoneum.

NEC Staging: classied according to the modied Bell’s

clinical staging criteria.

MATERIALS AND METHODS

A retrospective clinical-chart review was conducted

of patients with gastroschisis who had been surgically

repaired at Siriraj Hospital January 1, 2005-January

31, 2016. e inclusion criterion was neonates with

gastroschisis operated at Siriraj Hospital during the

11-year study period. Excluded were patients who had

undergone a denitive operation at another hospital or

who had incomplete medical records.

In all, 110 patients with gastroschisis were identied

as having been treated at Siriraj Hospital from 2005 to

2016. Complete medical records of 108 of the patients

were available and formed the basis of the retrospective

review. However, as two of those patients were subsequently

found to have undergone denitive surgical repair at other

hospitals, they were excluded. e patients’ demographic

and clinical data that were collected comprised gender,

gestational age, birth weight, Apgar score, type of delivery,

mode of surgical treatment, comorbidities, operative

ndings, and umbilical management. e variables of

the NEC and non-NEC groups were compared. e

signicant variables were reanalyzed to nd the potential

risk factors for NEC. e short-term outcomes (time to

rst oral feeding, duration of total parenteral nutrition

[TPN], and length of stay) were reviewed with respect to

overall complications, sepsis, wound infections, necrotizing

enterocolitis, morbidities, and mortalities.

Laohapensang et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

263

Original Article

SMJ

e statistical analyses were performed using SPSS

Statistics for Windows, version 17 (SPSS Inc., Chicago,

Ill., USA). e data were presented as mean, median,

mode, min, max, and standard deviation. Student’s t-test

was used for normally distributed data, and the Mann–

Whitney U test was employed for deviated distribution

or non-parametric tests in comparative quantitative data

analysis. e qualitative data were compared using the

Chi-square test, and if the sample size was deemed too

small, a two-tailed Fisher’s exact test was used. Statistical

signicance was determined by a p-value of less than

0.05. e signicant factors from the univariate analysis

were reanalyzed in a stepwise selection of variable and

logistic regression processes.

Ethics approval

e institute’s ethics review board approved the

review of the medical records undertaken by the present

research (Si 341/2015).

RESULTS

A total of 106 infants with gastroschisis were studied.

ere were 50 male and 56 female patients (47% vs.

53%), as summarized in Table 1. Sixty five percent

of the cases were preterm infants, mainly weighing

more than 1,500 grams (92.5%) and mostly with good

Apgar scores (73.6%). e most common associated

anomaly was bowel atresia with or without volvulus,

with an incidence of 12.3% of the study population. In

this study, 70 out of 106 patients (66.1%) underwent

primary fascial closure, which included 17% of the cases

with umbilical preservation and 49% with umbilical

cord excision. irty six out of 106 patients underwent

staged fascial closure, accounting for 34% of the cases

with delayed umbilical cord removal. e incidence of

NEC in this study was 16% (17/106), and the mortality

rate was 7.5% (8/106). All mortalities were not related

to NEC (1 abdominal compartment syndrome, 2 bowel

atresia with complications, 1 midgut volvulus, 2 sepsis,

1 pneumothorax, and 1 congenital heart disease and

associated hydranencephaly).

ere were 13 complicated patients with gastroschisis

with associated bowel atresia or midgut volvulus. Twelve

patients had associated intestinal atresia, of which 9

underwent fascial closure with constructed ostomy and

3 underwent fascial closure with primary anastomosis.

ere were two mortalities from the ostomy group. One

died from very short bowel syndrome, with a jejunal

length of 12 cm. Another died of complicated anastomotic

dysfunction aer operative ileostomy closure.

A univariate analysis to identify the potential risk

factors of NEC in infants with gastroschisis is presented in

Table 2.1. ere were no signicant dierences between

some of the parameters of the NEC and non-NEC groups

(gender, median gestational age, Apgar score, and birth

weight). However, the univariate analysis showed that

umbilical management (including umbilical preservation),

umbilical excision, and delayed umbilical resection during

staged abdominal wall closure were signicantly dierent

between the NEC and non-NEC groups, with a p-value

of 0.001. e signicant variables were reanalyzed in

a stepwise logistic regression, as detailed in Table 2.2.

Umbilical preservation was identied as a potential risk

factor for NEC in infants with gastroschisis, with a p-value

of 0.009 (hazard ratio = 5.13; 95% CI = 1.512-17.42), and

staged closure showed a tendency towards a preventive

eect of NEC, but with no statistical signicance (adjusted

odds ratio = 0.378; 95% CI = 0.078-1.937).

Excluding 8 mortalities, the short-term outcomes

were compared between the NEC and non-NEC groups.

Table 3 details the comparison of the outcomes (the

median time to rst oral feeding, duration of TPN, and

length of stay), all of which were signicantly higher in

the NEC than the non-NEC group (p-values = 0.006,

0.000, and 0.01, respectively).

e overall postoperative complications were intestinal

obstruction, respiratory complications, and NEC (Table 4).

ere were no signicant dierences between the groups

with dierent umbilical management in terms of the overall

complication rate, sepsis, and wound complications.

e infants with gastroschisis who underwent a staged

abdominal wall closure with delayed umbilical resection

had the highest overall complication rate (58.3%). is

incidence was similar to the gure of 55.6% for a primary

fascial closure with a preserved umbilicus group. e

primary fascial closure with umbilical removal group

had the lowest overall complications (36.5%) and wound

complication rate (9.6%). e incidence of sepsis was

highest in the staged closure with a delayed umbilical

resection group (27.8%).

e incidence of NEC was analyzed based on umbilical

management, which was classified according to the

modied Bell’s clinical staging criteria. e analysis

found signicant dierences between the 3 umbilical

management groups, with a p-value of 0.01 (Table 5).

e primary fascial closure with preserved umbilicus

group had the highest NEC incidence of 44.4%. e

umbilical removal approaches (primary fascial closure

or staged closure) had signicantly lower incidences of

NEC (13.5% and 5.6%, respectively).

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

264

TABLE 1. Demographic data of 106 neonates with gastroschisis treated 2005–2016 (n = 106).

Parameter Category Incidence %

Gender Male/Female 50/56 47.1/52.8

Mean GA (weeks) < 37/≥ 37 69/37 65/34.9

Birth weight (g) ≤ 1,500/> 1,500 8/98 7.5/92.5

Apgar < 7/≥ 7 28/78 26.4/73.6

Comorbidities Bowel atresia/volvulus 13 12.3

Umbilical management Primary closure

Preserve umbilicus 18 17

Remove umbilicus 52 49

Staged closure (remove umbilicus) 36 34

NEC 17 16

Mortality 8 7.5

TABLE 2.1. Univariate analysis of potential risk factors for NEC in infants with gastroschisis (n = 106).

TABLE 2.2. Logistic regression analysis of potential risk factors for NEC in infants with gastroschisis (n = 106).

Variable NEC (n = 17) Non-NEC (n = 89) P-value

Gender (M:F) 7:10 43:46 0.59

Median GA (weeks; range) 37 (34–39) 36 (30–44) 0.71

Median Apgar score (range) 9 (7–10) 10 (5–10) 0.97

Median birth weight (g; range) 2,340 (1,730–3,647) 2,060 (1,200–3,700) 0.08

Variable NEC Non-NEC

P-value

Adjusted odds ratio Hazard ratio

(n = 17) (n=89) (95% CI) (95% CI)

Umbilical management 0.001

Preserve umbilicus 8 (47.7%) 10 (11.2%) 0.009

5.143

(1.512–17.42)

Remove umbilicus 7 (41.2%) 45 (50.6%) 1

Staged closure 2 (11.8%) 34 (38.2%) 0.24

0.378

(remove umbilicus) (0.078–1.937)

Laohapensang et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

265

Original Article

SMJ

TABLE 3. Comparison of short-term outcomes of NEC and non-NEC groups (excluding 8 mortality cases; n = 98).

TABLE 4. e complication rates of each mode of treatment and umbilical management (n = 106).

TABLE 5. e incidence of NEC for each mode of treatment, classied according to the modied Bell’s clinical

staging criteria

(n = 106).

Variable NEC Non-NEC

(n = 17) (n = 81)

P-value

Time to feeding Median (min-max) 15 (6–45) 9 (4–73) 0.006

TPN duration Median (min-max) 22 (12–45) 12 (0–100) 0.000

LOS Median (min-max) 32 (19–67) 23 (9–365) 0.01

Abbreviations: TPN = total parenteral nutrition, LOS = length of stay, P-value of < 0.05 considered statistically signicant

Primary closure

Staged closure

Preserve umbilicus Remove umbilicus

(n = 36)

P-value

(n = 18) (n = 52)

Complications 10 (55.6%) 19 (36.5%) 21 (58.3%) 0.97

Sepsis 3 (16.7%) 9 (17.3%) 10 (27.8%) 0.44

Wound complications 4 (22.2%) 5 (9.6%) 8 (22.2%) 0.21

P-value of < 0.05 considered statistically signicant

Primary closure

Staged closure

Preserve umbilicus Remove umbilicus

(n = 36)

P-value

(n = 18) (n = 52)

NEC stage 8 (44.4%) 7 (13.5%) 2 (5.6%) 0.011

IA 4 3 1

IB 3 4 1

IIB 1 0 0

P-value of < 0.05 considered statistically signicant

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

266

DISCUSSION

Necrotizing enterocolitis occurring in infants aer

gastroschisis repair is not uncommon. It tends to occur

later in the clinical course, although it can oen be

successfully managed nonoperatively.

In this study,

the incidence of NEC was 16%, which is similar to the

ndings of previous studies that reported incidences

ranging from 4%-18.5%.

4–7

Most cases were diagnosed

early in stages I-II, and they were successfully treated

nonoperatively by fasting and intravenous empirical

antibiotics. Comprehensive studies by Snyder

and

Eggink et al.

reported lower NEC incidences of 4%

and 9%, respectively, which illustrates the reduction is a

consequence of more advanced stage inclusion criteria.

NEC in infants with gastroschisis is a distinctive

condition in that no signicant relationship has been

established with any known predisposing conditions,

such as prematurity, low birth weight, or low Apgar

score.

4,18

Conforming to previous reports, our study

found no significant association between NEC and

gestational age, birth weight, or Apgar score. However,

substantial studies by Oldham et al.

and Suttiwongsing

et al.

acknowledged the combination of low birth weight,

compromised intestinal vascular perfusion, and associated

gastrointestinal anomalies as predisposing factors for

NEC aer gastroschisis repair.

A study by Komuro et al.

reported the disadvantages

of umbilical preservation as a possible cause of wound

infections, omphalitis, cellulitis, and the development

of ventral hernia. is corresponds to earlier reports by

Kosloske and Bartow,

which emphasized the importance

of umbilical cord and urachal excision in complicated

periumbilical necrotizing fasciitis infants where there were

concerns of an infection spreading along the umbilical

vessels and urachal remnant. Based on those reports, it

can be speculated that the concept of infection spreading

along the umbilical vessels and urachal remnant may

be related to the development of NEC in infants with

gastroschisis. e hypothesis is that umbilical preservation

results in bacterial colonization in the area of the umbilical

insertion; consequently, spread through the fascial defect,

together with operative fascial closure procedures, may

lead to impaired tissue perfusion, reduced mucosal blood

ow, and cellular hypoxia. erefore, intestinal mucosal

injury and break-down precipitates the development of

NEC. Our study showed that umbilical preservation was

highly correlated to the development of NEC in infants

with gastroschisis, accounting for 47% of the NEC cases.

Accordingly, the incidence of NEC in the 18 patients

who underwent primary fascial closure with umbilical

preservation was as high as 44.4%, which is the highest

compared to both primary and staged umbilical resection,

with 13.5% and 5.6% respectively. As a result, umbilical

preservation was identied as a predisposing factor for

NEC in infants with gastroschisis aer surgical repair

(p-value = 0.009; hazard ratio = 5.13; 95% CI = 1.512-

17.42). Further additional pathophysiological studies

may be required.

Alternatively, various studies

10,11

have advocated

umbilical preservation as it oers better cosmetic results.

Nagaya et al.

reported the approach in later gastroschisis

cases, following complaints about the subsequent absence

of the umbilicus in earlier cases. In the 5 patients who

underwent fascial closure with umbilical preservation,

the procedure provided excellent cosmetic results with

no complicated omphalitis or cellulitis. ose studies

10,11

did not identify the incidence of NEC or its correlation

to umbilical preservation; in addition, they had fairly

small sample sizes. Our study found no dierence in the

incidences of overall complications, including sepsis and

wound complications, in the dierent umbilical management

groups. Nevertheless, the highest overall complication

rate of 58.3% in the staged fascial closure group may be

a consequence of a more critical population group, for

which a primary closure is impractical. Furthermore,

wound infections are common and likely to occur with a

staged fascial closure. As to the research by Fonkalsrud,

it found a much smaller percentage of patients (31%)

undergoing a primary fascial closure (compared to 66%

in the current study) and cited a higher incidence of

complications in those with a staged silo repaired. It was

postulated by Fonkalsrud that the signicant visceral-

abdominal disproportion precluded a primary closure.

With respect to the incidence of sepsis, aside from the

higher risk of wound infections in staged fascial closure

wounds, a delay in the time to the rst oral feeding results

in the prolonged use of a central venous catheter, which

in turn can ultimately predispose the patient to septicemia

more than any other reason.

In the present study, the

signicantly longer time to rst oral feeding, prolonged

duration of TPN, and length of hospital stay of the NEC

group were due to the early detection and treatment of the

condition. When NEC was suspected, the patient would

be fasted and promptly provided empirical intravenous

antibiotics as well as parenteral nutrition. Not only was

NEC diagnosed early (no later than stage IIB), all cases

were successfully managed nonoperatively. Only one

severe case developed pneumatosis intestinalis, and it

was in the umbilical preservation group.

Several earlier studies

4,7

included either only denite

NEC cases or cases from stages IIA upwards. is contrasts

with the current study, in which the majority of cases

Laohapensang et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

267

Original Article

SMJ

were in stages IA and IB, with only 1 case in stage IIB.

As to the diagnosis of stage I, some of the cases of NEC

may actually have been a delayed gut function, which

is common in patients with gastroschisis; however, this

is one of the limitations of conducting a retrospective

review. Despite the fact that most studies

4,7,8

reported

higher incidences (18%-18.5%) of NEC in infants with

gastroschisis, including only denite NEC cases, the declining

rate of NEC

over the past 2 decades has presumably been

the result of early detection and the availability of more

ecient antimicrobial agents. e lower incidence of NEC

in the staged fascial closure group in the present study

(5.6%) was consistent with the conclusion of Schlatter

et al.

that the improved outcome was attributable to lower

intraabdominal pressure following the initial preformed

silo coverage, allowing sucient intestinal perfusion.

e current study identied only one signicant

predisposing factor for NEC in infants with gastroschisis,

which is the preservation of the umbilicus. e patients

in the primary fascial closure with preserved umbilicus

group had 5 times the risk of developing NEC than

those in the umbilical removal groups over the study

period. A staged fascial closure conveyed the lowest

incidence of NEC, despite having the highest incidence

of wound infections, sepsis, and overall complications.

Nevertheless, the decision to perform any method of

fascial closure depends on a patient’s condition rather

than the prevention of wound complications. erefore,

NEC should be closely monitored when preserving the

umbilicus for a patient.

CONCLUSION

Umbilical preservation following gastroschisis

repair was strongly associated with the development of

NEC, with no correlation to gestational age. us, NEC

should be carefully monitored aer abdominal fascial

closure with umbilical preservation. NEC may lead to

a prolonged time to the rst oral feeding, an extended

duration of TPN, and a lengthened hospital stay. Staged

fascial closure showed a tendency towards a preventive

eect of NEC, but with no statistical signicance.

ACKNOWLEDGMENTS

The limitations of this study were those of a

retrospective analysis, a single center, and a relatively

small sample size. In addition, the study provided only

short-term outcomes of patients with gastroschisis; a

long-term study period may provide a more extensive

illustration of the outcomes and complications.

REFERENCES

1. Michael DK. Congenital defects of the abdominal wall. In:

Grosfeld JL, O’Neill JA, Fonkalsrud EW, Coran AG, editors.

Pediatric Surgery. Vol 2. 6

ed. St. Louis, MO: Mosby; 2006.

p. 1157-71.

2. Minkes RK. Abdominal wall defects. In: Oldham KT, Colombani

PM, Foglia RP, Skinner MA, editors. Principles and Practice

of Pediatric Surgery. Vol 2. 3

ed. Philadelphia: Lippincott,

Williams and Wilkins; 2005. p. 1103-19.

3. Chiengkriwate P, Sangkhathat S, Pathapinyokul S. Sutureless

elastic ring silo for the gastroschisis. Asian Biomed 2010;4:747-55.

4. Oldham KT, Coran AG, Drongowski RA. e development of

necrotizing enterocolitis following repair of gastroschisis: a

surprising high incidence. J Pediatr Surg 1988;23: 945-9.

5. Mungnirundr A, Khorana J, Ruangtrakool R. Correlation

between duration of postoperative parenteral nutrition and

incidence of postoperative complication in gastroschisis patients.

J Med Assoc ai 2010;93:443-8.

6. Jayanthi S, Seymour P, Puntis JW, Stringer MD. Necrotizing

enterocolitis aer gastroschisis repair: A preventable complication?

J Pediatr Surg 1998;33: 705-7.

7. Snyder CL. Outcome analysis for gastroschisis. J Pediatr Surg

1999;34:1253-6.

8. Suttiwongsing A, Sriworarak R, Buranakitjaroen V, Niramis

R. Related factors in Necrotizing enterocolitis aer gastroschisis

repair. ai J Surg 2011;32:113-9.

9. Lambert DK, Chistensen RD, Henry E, Besner GE, Baer VL,

Wiedmeier SE, et al. Necrotizing enterocolitis in term neonates:

Data from a multi-hospital health-care system. J Perinatology

2007;27:437-43.

10. Wessen DE, Baesl TJ. Repair of gastroschisis with preservation

of the umbilicus. J Pediar Surg 1986;21:764-5.

11. Nagaya M, Ando H, Tsuda M, Hiraiwa K, Akatsuka H. Preservation

of the umbilical cord at the primary fascial closure in infants

with gastroschisis. J Pediatr Surg 1993;28: 1471-2.

12. Patrapinyokul S. Abdominal wall defects. In: Patrapinyokul

S, editor. Pediatric surgery: Songklanagarindra. Songkhla:

Pediatric Surgery Unit, Department of Surgery, Prince of

Songkla University; 2002.p.105-20.

13. Komuro H, Imaizumi S, Hirata A, Matsumoto M. Staged silo

repair of gastroschisis with preservation of the umbilical cord.

J Pediatr Surg 1998;33:485-8.

14. Kosloske AM, Bartow SA. Debridement of periumbilical

necrotizing fasciitis: Importance of excision of the umbilical

vessels and urachal remnant. J Pediatr Surg 1991;26:808-10.

15. Engle WA. A recommendation for the denition of “late

preterm” (near-term) and the birth weight-gestational age

classication system. Semin Perinatol 2006;30:2-7.

16. Sykes GS, Molloy PM, Johnson P, Gu W, Ashworth F, Stirrat

GM, et al. Do Apgar scores indicate asphyxia? Lancet 1982;1:494-6.

17. Walsh MC, Kliegman RM. Necrotizing enterocolitis: treatment

based on staging criteria. Pediatr Clin North Am 1986;33:179-201.

18. Eggink BH, Richardson CJ, Malloy MH, Angel CA. Outcome

of gastroschisis: a 20-year case review of infants with gastroschisis

born in Galveston, Texas. J Pediatr Surg 2006;41:1103-8.

19. Fonkalsrud EW. Selective repair of neonatal gastroschisis

based on degree of visceroabdominal disproportion. Ann Surg

1980;191:139-44.

20. Baird R, Puligandla P, Skarsgard E, Laberge JM; Canadian

Pediatric Surgical Network. Infectious complications in the

management of gastroschisis. Pediatr Surg Int 2012;28: 399-404.

21. Schlatter M, Norris K, Uitvlugt N, DeCou J, Connors R. Improved

outcomes in the treatment of gastroschisis using a preformed

silo and delayed repair approach. J Pediatr Surg 2003;38:459-64.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

268

Malee Ngamprasert, D.N.S. Candidate*, Ketsarin Utriyaprasit, Ph.D.**, Siriorn Sindhu, Ph.D.**, Ameporn

Ratinthorn, Ph.D.***, Chukiat Viwatwongkasem, Ph.D.****

*Doctor of Nursing Science Program, **Department of Surgical Nursing, ***Department of Obstetric & Gynecological Nursing, Faculty of Nursing,

Mahidol University, Bangkok 10700, ****Department of Biostatistics, Faculty of Public Health, Mahidol University, Bangkok, ailand.

The Relationship among a Palliative Care Service,

Patient’s Factors, and Quality of Life of Post

Treatment Cervical Cancer Patients: a Causal Model

Approach

Corresponding author: Ketsarin Utriyaprasit

E-mail: ketsarin.utr@mahidol.ac.th

Received 26 February 2019 Revised 1 April 2019 Accepted 22 April 2019

ORCID ID: http://orcid.org/0000-0001-7932-3395

http://dx.doi.org/10.33192/Smj.2019.41

ABSTRACT

Objective: To test a causal pathway of palliative care (PC) service delivery system on quality of life (QOL) of patients

with post-treatment stage II-III cervical cancer (PT2-3CC-PTs).

Methods: e cross-sectional design with two-staged sampling was implemented to recruit 447 PT2-3CC-PTs and

16 healthcare providers from 12 cancer center hospitals. Standardized tools and procedures were used, followed by

the structural equation model (SEM) analysis to measure variables of interest.

Results: e average age was 54.07 (SD 11.72) years, most of the patients (90.2%) received radiotherapy and

chemotherapy, and more than two-thirds (68.7%) had comorbidities. e results showed that the patients had

problems with the physical, role, emotional, and social functioning due to urinary and vaginal symptoms which

made them anxious about sexual health. SEM analysis also revealed that the PC service delivery system did not

improve the patients’ QOL signicantly. Finally, it was found that healthcare providers had low competencies in

PC, and their caring did not respond to the patients’ needs concerning a woman’s specic role and responsibility.

Conclusion: e PC service delivery system in ailand mainly provides physical care without encouraging family

to collaborate in care of the patients. PC training to increase competency of providers to care for PT2-3CC-PTs

should take the patients’ age, severity of comorbidities, treatment dierences, and a woman’s role, especially sexual

health into account. us, to enhance QOL (general and sexual health) in PT2-3CC-PTs, healthcare providers

should focus on patients’ self-management and provider support.

Keywords: Quality of life; post-treatment cervical cancer; palliative care delivery service; sexual health; self-

management; social support (Siriraj Med J 2019; 71: 268-277)

INTRODUCTION

Quality of life (QOL) is one of the measurements

used to evaluate outcome of the patients’ cervical cancer

(CC) care, self-management (SM), and CC health-related

palliative care (PC). Patient care is found to be associated

with patients’ QOL

and impacted by functioning problems

including physical functioning

, role functioning

, social

functioning

4,5

, emotional functioning

, cognitive functioning

Ngamprasert et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

269

Original Article

SMJ

sexual functioning

, and symptoms experience aecting

daily lives.

8-11

Patients with CC need PC for symptom

and psychosocial management so as to maintain all

dimensional functioning. Healthcare providers are

responsible for providing information and practical

care so they should acknowledge patients’ competency

knowledge on disease and treatment, and they should

possess eective communication skills and understanding

of women with CC.

9,11,12

Previous studies have reported

that QOL was lower in ACC-PT post-treatment compared

to that of other women.

2-4,13

PC is one of health delivery

services that support symptom management, symptom

control, and QOL of post-treatment cervical cancer

patients (PTCC-PT).

14-16

PC could be applied early on

in the course of illness or in conjunction with other

therapies intended to prolong life.

However, several

studies of PC have been undertaken only with the advance

stage cervical cancer patients (ACC-PT) or stage 4.

ere was a few studies on PC during post-treatment

stages 2-3 cervical cancer patients (PT2-3CC-PTs).

2,11,13,24

In fact, the study of PC among PT2-3CC-PTs is needed

due to the highest prevalence of this disease among the

female population. It was anticipated that the results of

the present study could be utilized to provide knowledge

of PC for PT2-3CC-PTs.

is study aimed to promote QOL for the PTCC-

PT depending on multidimensional factors, including

patients’ dimensional factors and PC dimensional factors

that enhance patients’ self-management behaviors to

reduce physical, mental, and social health problems.

Patients’ dimensional factors investigated in this study

included demographic factors (age of the patients

9,18

, social

support

4,5,19

, and patients’ self-management)

7,12,15,20

, as well

as medical factors (the stage of disease

, radiation therapy/

combination of radiation therapy and chemotherapy

(RT/CCRT)

9,21

, and comorbidities

). PC dimensional

factors included model of care

22,23

, provider competence

and provider support.

7,24

In addition, the relationship

between the patients’ dimensional factors and the PC

dimensional factors and QOL of PT2-3CC-PTs

14,15

has rarely been studied in the country. erefore, this

study also examined the association of the patients’

dimensional factors and the PC dimensional factors on

QOL of PT2-3CC-PTs.

Research Questions

1. What factors in the patients’ dimensional factors

(age, stage of disease, RT/CCRT, comorbidity, social

support, self-management, and symptom experience)

have a direct eect on QOL of ai PT2-3CC-PTs?

2. What factors in the PC dimensional factors (model

of care, provider competence, and provider support)

have an eect on QOL of ai PT2-3CC-PTs?

3. What factors between the patients’ dimensional

factors (age, stage of disease, RT/CCRT, comorbidity,

social support, self-management, and symptom experience)

and the PC dimensional factors (model of care, provider

competence, and provider support) have an eect on QOL

of ai PT2-3CC-PTs in tertiary care cancer centers in

four regional service areas in ailand?

MATERIALS AND METHODS

Design

A cross-sectional study design was conducted from

May 2016 to April 2017. e data were presented focusing

on the results regarding the relationships between patients’

dimensional factors and PC dimensional factors on QOL

of ai PT2-3CC-PTs.

Sample and Recruitment Methods

e two-stage sampling was used to select service

areas and recruit participants of this study. Field data

collection comprised four regional service areas in ailand

categorized by health service networks from the guideline

of the Ministry of Public Health. Finally, the four areas

were selected including the fourth, h, sixth, and 13

service areas. ese four areas had the highest registry

number, 61.1% of 3,771 new CC patients in the year

2013 (1,844 of whom were PT2-3CC-PTs).

e research was implemented in 16 follow-up

clinics including ve obstetric-gynecological clinics and 11

radiological clinics of 12 tertiary HC centers. e power

analysis was implemented to calculate the sample size for

SEM, the sample size required in this study was 480.

en, the distribution of patient participants in each

hospital setting was calculated based on the probability

proportional to size sampling

that was appropriate for

an unequal number of CC population in each clinic. ere

were 16 provider participants (one provider participant

per clinic) in the study.

e participants of this study were divided into

two groups including patients and HC providers. e

rst group was patients with 2-3CC-PT who underwent

RT/CRT for three to 12 months as prescribed by the

physician to ensure completeness of response for cancer

treatment. en, the patients received care provided

at the onco-gynecological clinic, radiotherapy clinic,

and obstetric-gynecological clinic in the out-patient

department of cancer centers. Next, the patients age was

18 years and over, and they understood or could read or

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

270

write in ai. Patients who had recurrence or metastasis,

had been diagnosed with cognitive impairment, or were

unable to communicate were excluded from this study.

e second group was HC providers including

physicians, radio-oncologists, gynecological-oncologists,

and registered nurses (oncology nurse specialists and

registered nurses in the radiotherapy clinic, onco-

gynecological clinic, or obstetric-gynecological clinic

in the out-patient department of the cancer centers). e

HC providers were selected based on the settings where

the patients received care. Temporary HC providers in

the clinics were excluded.

Ethics

Ethical approval for the study was obtained from

the Ethics Committee of the Faculty of Medicine, Siriraj

Hospital, Mahidol University (Si 281/2016) and Human

Research Ethics Committee of each 11 cancer center

hospitals.

Measures

e participants’ self-reported questionnaires were

used to collect data including dimensional factors, PC

dimensional factors, and QOL.

Patient’s dimensional factors were measured using

the Patients Demographic and Medical Information

Questionnaire (PDMIQ) to collect the demographic

data and medical information of the PT2-3CC-PTs.

Next, the Cervical Cancer Patient Self-management

Questionnaire (CCSMQ) was used to assess the patients’

ability to use available resources, information, decision

making, and SM skill by sharing decision making and

utilizing coping strategies. e reliability of the tool was

determined using internal consistency, with Cronbach’s

alpha equal to 0.92. en, the Medical Outcome Study-

Social Support Survey (MOS-SSS) was employed to measure

perceptions of the availability of support or the support

from partners, family, and peers including its adequacy.

As for reliability of the instrument, internal consistency

was calculated and Cronbach’s alpha was 0.90. Lastly,

data regarding comorbidity, which was dened as one

or more additional diseases or disorders co-occurring

with a primary disease or disorder that the patients had

before cancer diagnosis, were retrieved from the patients’

health record based on the Charlson Comorbidity Index

(CCI).

Each comorbidity was reassigned into the number

weighting by the severity of diseases.

Furthermore, data regarding PC dimensional factors

were collected by means of the Structure of Palliative

Care Delivery Service Questionnaire (SPCDQ) and the

Palliative Care Delivery Service Questionnaire (PCDQ)

which were developed by the researchers. Firstly, the

SPCDQ consisted of the palliative care delivery structure

questionnaire (PCDSQ) and the provider information

questionnaire (PIQ). e PCDSQ questionnaire was

used to collect data regarding the type of care, number of

providers in the clinic, and number of patients who received

care per day, etc. e provider information questionnaire

(PIQ) was used to gather provider information including

gender, age, educational background, work experience,

duration of work experience, specialist certication,

and role in the clinic. Secondly, the PCDQ consisted of

the provider competence questionnaire (PCQ) and the

provider support questionnaire (PSQ). e PCQ assessed

HC provider competence. e competency included

symptom monitoring and management, use of simple

language with the patients, capability to enhance good

relationships with patients and other providers, being

an eective advocate for the patient, and referral to the

specialist. e reliability of the tool was determined in

terms of internal consistency, with Cronbach’s alpha equal

to 0.95. Also, the provider support questionnaire (PSQ)

was used to measure four behaviors of care provided for

patients including coaching, training, goal setting, and

performance feedback. e reliability of the tool was

determined using internal consistency, and Cronbach’s

alpha was 0.97.

The European Organization for Research and

Treatment of Cancer Quality of Life Questionnaire C30

(EORTC QOQ-C30) was a 30-item core cancer specic

questionnaire measuring QOL in cancer patients. e

instrument was provided by the EORTC Quality of

Life Group.

e reliability of the tool using internal

consistency by means of Cronbach’s alpha was 0.78. Finally,

the EORTC QOQ CX 24 was a 24-item questionnaire to

assess patients’ perception of symptoms related to CC.

e reliability of the tool was determined using internal

consistency, with Cronbach’s alpha equal to 0.78.

Data Collection Procedures

e study protocol was submitted to and approved

by Human Research Protection Unit, Faculty of Medicine

Siriraj Hospital, Mahidol University, and Human Research

Ethics Committee of each 11 cancer center hospitals.

In addition, formal permission from the cancer center

administrators was sought aer. en, the researchers

approached prospective participants who met the inclusion

criteria and agreed to participate in the study. Aer

the participants signed the informed consent form to

indicate their willingness to take part in the study, data

collection commenced. e researchers collected data

from the PT2-3CC-PTs and retrieved the patients’ medical

Ngamprasert et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

271

Original Article

SMJ

history and CCI from existing medical records. e

EORTC QOQ-C30 and the EORTC QOQ-CX24 were

administered to explore patients’ QOL. en, the PDMIQ,

CCSMQ, and MOS-SSS were administered to collect data

regarding patient dimensional factors. e time required

to complete all questionnaires was approximately 30-45

minutes. If the participants had visual problems, their

caregiver or the researchers helped them complete the

questionnaires by giving them verbal assistance.

e researcher collected data from HC providers

who were key informants and agreed to participate in

the study aer the informed consent form was signed.

e SPCDQ and the PCDQ were administered. e time

to complete the questionnaires was approximately 30-45

minutes. All providers completed the questionnaires by

themselves.

Data analysis

SEM statistics were conducted to identify the causal

pathway of the PC dimensional factors and the patient

dimensional factors that aected QOL of PT2-3CC-PTs.

RESULTS

e study sample comprised 450 patients, three of

whom subsequently dropped out. One had a severe pain

problem, another went back to work and were unable to

participate in the study, and the last one was too tired

to join due to her old age. Finally, the total number of

participants was 447 PT2-3CC-PTs and 16 HC providers.

Patient Dimensional Factors on QOL

e mean age of PT2-3CC-PTs was 54.07 years old

(SD 11.72). More than a half (51.1%) were classied as

stage 2, and almost all (90.2%) received radiotherapy

with chemotherapy. Moreover, more than a half (60%)

had mild severity of comorbidity before cancer diagnosis

(Table 1). e patients received less SS and performed

good self-management. For instance, the results showed

that most of the patients did not receive overall dimension

of SS as they needed. Furthermore, the patients had low

ability to use HC resources and HC provider information

to perform SM and collaborate in decision making with

the HC provider (Table 2).

PC Dimensional Factors on QOL

e results regarding PC delivery structure in this

study are shown in Table 3. ere were a limited number

of PC nurses. As regards provider characteristics, nurses

were found to be the professionals providing overall PC

for CC-PTs. e age of nurses was older than 45 years,

and a few of the providers had experience in PC.

e PC delivery service was composed of two essential

structures, including provider competence and provider

support, as shown in Table 3. In this study, it was found

that the providers had less competence in disease and

treatment knowledge. Providers also had a variety of

forms of symptom and complications which impacted

patient communication. In multidisciplinary care, the

providers had the highest competence in PC. Providers used

simple language, had eective symptom and complication

impact communication, had eective patient support,

and referred the patients to the specialist when needed.

Comparing with providers in the multidisciplinary team,

the providers in the specialist care team had the lowest

competence in PC. e majority of the providers in this

study provided training, goal setting, and performance

feedback; however, they had limited ability to perform

coaching. Such results indicated that providers in the PC

delivery service system perceived themselves as having

low competence in PC, resulting in the ineectiveness

of PC services provided to patients.

Relationship between Patient Dimensional Factors

and PC Dimensional Factors and QOL

Structural equation model was performed to examine

the eects of the PC dimensional factors and the patient

dimensional factors on QOL of PT2-3CC-PTs by using

the program AMOS version 23. Table 5 presents the

hypothesized model (theoretical model) which did not t

the empirical data due to high inter-correlation, the poor

goodness of t coecient, and misspecication parameters.

us, the hypothesized model (Fig 1) was modied to

achieve a better t. In this step, the modications were

maintained until acceptable t indices were achieved.

Error covariance of self-management, provider support,

model of care, and quality of life were set free. One variable,

stage of disease, one path from the model of care to QOL,

and eight covariances were deleted. On the contrary,

two paths, one from CCI score to model of care and one

from the model of care to patient self-management, were

added. In the end, the modication model improved the

model of t better than the hypothesized model (Table 5).

e results regarding SEM of QOL in PT2-3CC-

PTs showed that the theoretical model explained less

than the modied model (Table 5). e modied model

(Fig 2) shows the factors that directly aected dierences

in QOL of PT2-3CC-PTs like age dierence (β = -0.161,

p < 0.001), patients’ SM dierence (β = -0.265, p < 0.001),

and palliative care delivery service dierence (β = 0.114,

p < 0.01), as well as factors that indirectly aected QOL

through patients’ SM like treatment dierence (β =

-0.031, p < 0.01), severity of comorbidity prior to cancer

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

272

TABLE 1. Demographic and Medical Characteristics of PT2-3CC-PTs (N = 447).

Characteristics Number Percent

Age (years) Mean = 54.07, SD = 11.72, Range = 24 - 86

≤ 45 107 23.9

> 45 340 76.1

Marital status

Single 21 4.7

Marriage 282 63.1

Widowed 87 19.5

Separated 57 12.8

Education

No formal education 36 8.1

Primary school 251 56.2

Lower than Bachelor’s degree 114 25.5

Undergraduate and graduate degree 46 10.3

Living with

Other family/ No one 21 4.7

Couple 109 24.4

2-generation family 136 30.4

≥ 3 generation (extended) family 181 40.5

Monthly Income (Baht)

no income 201 45.0

1-5,000 70 15.7

5,001-10,000 88 19.7

10,001 baht or more 88 19.7

Medical Payment Method

Universal healthcare coverage 265 59.3

Social insurance scheme 111 24.8

Government reimbursement 64 14.3

Others (self-payment, insurance, company benet, Private 7 1.6

Education Commission: OPEC)

Stage (FIGO)

2A 51 11.4

2B 222 49.7

3A 14 3.1

3B 160 35.8

Treatment

Radiotherapy 44 9.8

Radiotherapy & chemotherapy 403 90.2

Number of comorbidities

No comorbidity 140 31.3

1 comorbidity 185 41.4

2 comorbidities 88 19.7

≥ 3 comorbidities 34 7.6

CCI severity scores

0 140 31.30

1-2 (mild severity) 196 43.80

3-4 (moderate severity) 76 17.00

≥ 5 (high severity) 35 7.80

Ngamprasert et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

273

Original Article

SMJ

TABLE 2. Social Support and Self-management of PT2-3CC-PTs (N = 447).

Variables Possible range Actual range Median Mean SD

Social Support 16-75 26-83 68.00 65.92 11.01

Number of close friends 0-30 2.00 3.89 14.63

Number of close cousins 0-40 4.00 4.72 5.09

Material support 5-20 5-20 18.00 16.92 3.36

Emotional/informational support 6-30 6-30 23.00 22.75 5.18

Social relation support 3-15 3-15 12.00 11.79 2.61

Affective support 2-10 3-10 10.00 9.00 1.39

Self-management 19-76 26-76 56.00 56.69 9.01

Healthcare resources used 4-16 4-16 11.00 10.82 2.95

Ability caused by healthcare information 4-16 4-16 12.00 11.51 2.41

Collaborative care with healthcare providers 5-20 7-20 15.00 15.55 2.50

Shared decision making 3-12 3-12 9.00 8.84 1.77

Behavioral changes 3-12 4-12 10.00 9.97 1.55

TABLE 3. Characteristics of palliative care delivery service system.

Characteristics Total score Median IQR (Q1,Q3)

Palliative care delivery structure

Number of physicians 4.0 6.0 (4.0,10.0)

Number of onco-gynecologists 0.9 2.8 (0,2.8)

Number of onco-radiologists 4.1 2.3 (3.2,5.5)

Number of gynecologists 0.7 4.4 (0,4.4)

Number of registered nurses 5.9 5.9 (3.7,8.6)

Number of onco-nurses 1.8 2.8 (0.8,3.6)

Number of PC nurses 0.4 1.8 (0,1.8)

Work experience in clinic (yrs.) 9.8 9.2 (5.4,14.6)

Work experience in profession (yrs.) 22 16.1 (12.7,28.8)

Work experience in onco-specialization (yrs.) 2.4 14.7 (0.2,14.9)

Work experience in palliative care (yrs.) 1.8 7.3 (0,7.3)

Total service time/ one patient(min/patient) 132.8 111.5 (57.1,168.6)

Nurse service time/one patient(min/patient) 79.5 56.9 (29.8,86.7)

Screening time 3.8 3.4 (2.6,6.0)

Need/symptom assessment time 9.5 9.6 (4.5,14.1)

Need/symptom management time 13.4 15.6 (3.3,18.9)

SM resource information time 9.4 11.7 (4.5,16.2)

SM consultancy/coaching/skill training time 19.4 22.0 (6.9,28.9)

Referral to unit/service appropriate for

patient need time 9.1 7.5 (5.4,12.9)

Physician service time/one patient (min/patient)

(Monitor recurrence time) 54.2 56.3 (21.0,77.3)

Palliative care delivery service

Provider competence 100 88 15 (81-96)

Knowledge of disease 32 18 4 (15,19)

Use of simple language 12 12 3 (9,12)

Effective symptom and complication management

Effective communication 24 21 5 (18,23)

Patient relation building 12 11 3 (9,12)

Effective patient support 12 10 3 (9,12)

Referral to specialists 20 15 4 (15,19)

Provider support 80 71 12(60-72)

Coaching 40 36 8 (30,38)

Training 20 17 3 (15,18)

Goal setting 12 9 2 (9,11)

Performance feedback 8 7 1 (6,7)

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

274

TABLE 4. Quality of life scale scores* in PT2-3CC-PTs (N = 447).

Quality of life Possible range Actual range Median Mean SD

EORTC QLQ C30 scale score

General quality of life 2-14 2-14 11.00 11.20 2.06

Physical functioning 5-20 5-20 7.00 7.29 2.73

Role functioning 2-8 2-8 2.00 2.46 0.99

Emotional functioning 4-16 4-16 6.00 6.11 2.24

Cognitive functioning 2-8 2-8 3.00 2.90 0.90

Social functioning 2-8 2-8 2.00 2.55 1.01

EORCT C 30 symptoms

Fatigue 3-12 3-12 5.00 5.49 1.77

Nausea and vomiting 2-8 2-6 2.00 2.35 0.76

Pain 2-8 2-8 3.00 3.30 1.34

Dyspnea/Shortness of breath 1-4 1-4 1.00 1.38 0.58

Sleep disturbance 1-4 1-4 2.00 1.72 0.82

Lack of appetite 1-4 1-4 1.00 1.59 0.77

Constipation 1-4 1-4 1.00 1.52 0.68

Diarrhea 1-4 1-4 1.00 1.33 0.55

Financial problems 1-4 1-4 1.00 1.73 0.90

EORCT CX 24 symptoms

Sexual vaginal functioning** (n = 293) 4-16 4-16 5.00 4.73 4.12

Symptom experience 11-44 11-34 14.00 15.28 3.29

Body image 3-12 3-12 4.00 4.65 1.90

Neuropathy 1-4 1-4 2.00 1.76 0.79

Menopausal symptom 1-4 1-4 1.00 1.51 0.72

Sexual pleasure** (n = 293) 1-4 1-4 4.00 3.70 0.70

Sexual functioning** 1-4 1-4 4.00 3.73 0.61

Sexual anxiety 1-4 1-4 1.00 1.27 0.61

Lymphedema 1-4 1-4 1.00 1.17 0.45

* higher score = more symptoms/problems, ** higher score = less symptoms/problems

TABLE 5. Statistical tted index values of hypothesized model and modied model (N = 447).

Fitted Index χ2 χ2/df GFI CFI RMSEA Largest

Standardized

Residual

Hypothesized model 37.362 2.076 0.984 0.986 0.049 3.469

(df = 18, p= 0.005)

Modication model 20.220 0.843 0.990 1.000 0.000 1.804

(df=24, p = 0.684)

Abbreviations: df = degree of freedom GFI = Goodness of t index (≥0.95), CFI = Comparative t index =[d(Null model)- d (purposed

model)]/d (Null model) (≥0.95), RMSEA = Root mean square error of approximation= √( χ2 - df) / √[df(N - 1)] (≤ 0.05) Standardized

residual (≤ 2.58)

Ngamprasert et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

275

Original Article

SMJ

diagnosis dierence (β = 0.032, p < 0.01), and social

support dierence (β = -0.103, p < 0.001). e dierence

of provider competence had a minimal indirect eect

on QOL dierence through the model of care, provider

support, and patients’ SM (β = 0.095, p < 0.001). Model

of care dierence had no direct eect on patients’ QOL

but had the minimal indirect eect on the dierence of

QOL through patients’ SM (β = -0.033, p < 0.01).

DISCUSSION

The results showed that PT2-3CC-PTs who

completed response to cancer treatment had poor QOL

due to unsupportive health system. In particular, the

health system was not sensitive to sexual health needs,

resulting in a lack of engagement of patients, family,

and healthcare providers. For instance, the ndings

revealed no relationships between patients and family

Fig 1. Relationships between patient dimensional factors, palliative care dimensional factors, and QOL of PT2-3CC-PTs of the hypothesized

model

Fig 2. Relationships between patient dimensional factors, palliative care dimensional factors, and QOL of PT2-3CC-PTs of the modied

model

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

276

and providers who provided SM support in response

to the patients’ health problems. Hence, cooperation

among patients, family, and healthcare provider is an

important strategy to improve QOL of the patients as

shown in Fig 2.

is study also found that patients could not live

well because of functioning problems, general symptom

problems, sexual health problems, and economic problems,

as shown in Table 4. e results were correlated with

previous studies that PT2-3CC-PTs who had functioning

problems, general symptom problems, and sexual health

problems would have poor QOL.

3,7,9,19,22,31

e results in this study indicated that PC in ailand

was mainly developed by considering mainly physical

functioning of the patients. As such, the system overlooked

patients’ age, sexual health related to CC treatment, and

severity of comorbidity prior to CC diagnosis. ese

factors were associated with patients’ SM, resulting in

poor QOL of the majority of the patients. For example,

the presents study found no relationship between patients

and family and providers who provided SM support

in response to the patients’ health problems. Hence,

supportive health system is an important strategy to

improve QOL of the patients, as shown in Fig 2.

Support from family is the strength of ai society.

e strategy to encourage patients’ family to collaborate

in PT2-3CC-PTs’ HC is signicant for improvement of

QOL for the patients.

Recently, the strategy to provide

PC as early as possible by integrating PC services into

cancer treatment services is promoted to enhance patients’

QOL.

In this study, the results showed that the patients

could not live well because of their health problems,

especially sexual health problems, which, in turn, impacted

their emotional, role, and social functioning, as shown

in Table 4. e dierence of PC dimensional factors

associated with poor QOL in the patients in this study

is shown in Fig 2. Poor QOL of the patients implied that

although the patients had accessed to HC delivery service

(follow-up care), they did not receive the PC service

as needed. Moreover, PC delivery system in ailand

was not eective enough to encourage patients’ family

collaboration in PT2-3CC-PTs’ HC; thus, the patients

had poor QOL (Fig 2). e ndings were consistent

with previous studies which found that CC patients

undergoing complete care for cancer treatment could

not live well because HC providers mostly alleviated

only physical health problems

without considering

the dierences in age

18,22

, comorbidity severity prior to

cancer diagnosis

, and sexual health problems of the

patients.

33,34

As a result, the current service could not

enhance patients’ self-management, causing the patients

to continue suering from the disease and decreased

QOL.

CONCLUSION

In ailand, family is a key component of PC. SS by

family members who collaborate with HC delivery service

is a promising strategy to develop PC dimensional factors.

Another important strategy is to increase the number of

providers who understand sexual health care and PC to

support both women in the reproductive age and aging

women. PT2-3CC-PTs are women with dierent sexual

health problems who still have several roles to play in

daily life such as a wife, a mother, and a woman who have

their roles for themselves, family, and society. Enabling

PT2-3CC-PTs to manage care for themselves requires

the support system that directly serves their needs. In

addition, they also require the support from family and

HC providers who understand their HC needs, especially

sexual health problems that can dier in accordance

with their age, severity of comorbidity prior to cancer

diagnosis, and types of cancer treatment. erefore,

HC providers have an important role to support the

patients’ ability to SM by improving their ability to

manage their health problems, while simultaneously

encouraging the patients’ family to promote their SM. In

addition, competency of nurses who provide care for the

patients is an essential component of HC delivery service

systems that should be developed urgently. e persons in

charge of organizing HC delivery service systems should

consider nurses as an important bridge to link, transfer

knowledge, and develop skills in sexual health care to the

patients. Besides this, policy makers should be careful

when devising a policy on PC delivery services for PT2-

3CC-PTs. It is suggested that a longitudinal study should

be conducted to shed more light on the continuity of

relationships between PC dimensional factors and patient

dimensional factors. Moreover, further studies should be

carried out to conrm that the factors identied as being

related to QOL of PT2-3CC-PTs in this study including

patient dimensional factors such as age and SM and PC

dimensional factors such as provider support actually

have an inuence on QOL of PT2-3CC-PTs.

ACKNOWLEDGMENTS

is research was funded by the National Research

Council of ailand (NRCT), FY2017 esis Grant for

Doctoral Degree Students.

Ngamprasert et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

277

Original Article

SMJ

REFERENCES

1. Damodar G, Gopinath S, Vijayakumar S, Rao AY. Reasons for

Low Quality of Life in South Indian Cancer Patient Population:

A Prospective Observational Study. Indian J Pharm Sci 2014;76:

2-9.

2. Korfage IJ, Essink-Bot M-L, Mols F, van de Poll-Franse L,

Kruitwagen R, van Ballegooijen M. Health-Related Quality

of Life in Cervical Cancer Survivors: A Population-Based

Survey. J Radiat Oncol Biol Phys 2009;73:1501-9.

3. Osann K, Hsieh S, Nelson EL, Monk BJ, Chase D, Cella D, et al.

Factors associated with poor quality of life among cervical

cancer survivors: implications for clinical care and clinical

trials. Gynecol Oncol 2014;135:266-72.

4. Ashing-Giwa KT, Tejero JS, Kim J, Padilla GV, Kagawa-Singer

M, Tucker MB, et al. Cervical cancer survivorship in a population

based sample. Gynecol Oncol 2009;112:358-64.

5. Lim JW, Zebrack B. Dierent pathways in social support and

quality of life between Korean American and Korean breast

and gynecological cancer survivors. Qual Life Res 2008;17:679-

89.

6. Mantegna G, Petrillo M, Fuoco G, Venditti L, Terzano S,

Anchora LP, et al. Long-term prospective longitudinal evaluation

of emotional distress and quality of life in cervical cancer

patients who remained disease-free 2-years from diagnosis.

BMC Cancer 2013;13:127.

7. Jarroenwech C, Sindhu S, Ratinthorn A. e eect of empowerment

program using PEACE process on quality of life among cervical

cancer patients with sexual dysfunction aer treatment. Bangkok:

Mahidol University; 2015.

8. Beesley V, Janda M, Eakin E, Obermair A, Battistutta D.

Lymphedema aer gynecological cancer treatment. Cancer

2007;109:2607-14.

9. Foster C, Fenlon D. Recovery and self-management support

following primary cancer treatment. Br J Cancer 2011;105

(Suppl 1):S21-S8.

10. Herzog TJ, Wright JD. e impact of cervical cancer on quality

of life—e components and means for management. Gynecol

Oncol 2007;107:572-7.

11. Lammerink EAG, de Bock GH, Pras E, Reyners AKL, Mourits

MJE. Sexual functioning of cervical cancer survivors: A review

with a female perspective. Maturitas 2012;72:296-304.

12. Oskay ÜY, Beji NK, Bal MD, Yılmaz SD. Evaluation of sexual

function in patients with gynecologic cancer and evidence-

based nursing interventions. Sex Disabil 2011;29:33-41.

13. Azmawati MN, Najibah E, Hatta MD, Norfazilah A. Quality

of life by stage of cervical cancer among Malaysian patients.

Asian Pac J Cancer Prev 2014;15:5283-6.

14. Hui D, De La Cruz M, Mori M, Parsons HA, Kwon JH, Torres-

Vigil I, et al. Concepts and denitions for supportive care, best

supportive care, palliative care, and hospice care in the published

literature, dictionaries, and textbooks. Support Care Cancer

2013;21:659-85.

15. Johnston B, Rogerson L, Macijauskiene J, Blaževičienė A,

Cholewka P. An exploration of self-management support in

the context of palliative nursing: a modied concept analysis.

BMC Nursing 2014;13:1-10.

16. Yennurajalingam S, Kang JH, Hui D, Kang DH, Kim SH, Bruera

E. Clinical response to an outpatient palliative care consultation

in patients with advanced cancer and cancer pain. J. Pain

Symptom Manag 2012;44:340-50.

17. WHO. Cervical cancer screening in developing countries.

Report of a WHO consultation, 2002.

18. Lai BPY, Tang CS-k, Chung TKH. Age-specic correlates of

quality of life in Chinese women with cervical cancer. Support

Care Cancer 2009;17:271-8.

19. Ashing-Giwa KT, Lim J-W. Health-related quality of life

outcomes among cervical cancer survivors: Examining ethnic

and linguistic dierences. Cancer Epidemiol 2011;35:194-201.

20. Bulsara C, Styles I, Ward AM, Bulsara M. e psychometrics

of developing the patient empowerment scale. J Psychosoc

Oncol 2006;24:1-16.

21. Barnas E, Skret-Magierlo J, Skret A, Bidzinski M. e quality of

life of women treated for cervical cancer. Eur J Oncol Nurs

2012;16:59-63..

22. Nord C, Mykletun A, orsen L, Bjøro T, Fosså SD. Self-

reported health and use of health care services in long-term

cancer survivors. Int J Cancer 2005;114:307-16.

23. Stjernsward J, Foley KM, Ferris FD. Integrating palliative care

into national policies. J Pain Symptom Manage 2007;33:514-20.

24. Sumdaengrit B, Hanucharurnkul S, Dodd M, Wilailak S,

Vorapongsathorn T, Pongthavornkamol K. Symptom Experience

and Self-care among ai Women with Cervical Cancer. Pac

Rim Int J Nurs Res ail 2013;14:203-18.

25. Hair JF, Black WC, Babin BJ, and Anderson RE. Multivariate

Data Analysis (7th Edition). Pearson, New York, 2010.

26. Bierrenbach. Steps in applying Probability Proportional to

Size (PPS) and calculating Basic Probability Weights. 2008.p.1-5.

27. Rungruangsiripan M, Sitthimongkol Y, Maneesriwongul W,

Talley S, Vorapongsathorn T. Mediating Role of Illness

Representation Among Social Support, erapeutic Alliance,

Experience of Medication Side Eects, and Medication Adherence

in Persons With Schizophrenia. Arch Psychiatr Nurs 2011;25:269-83.

28. Charlson ME, Charlson RE, Peterson JC, Marinopoulos SS,

Briggs WM, Hollenberg JP. e Charlson comorbidity index is

adapted to predict costs of chronic disease in primary care

patients. J Clin Epidemiol 2008;61:1234-40.

29. Fayers P, Aaronson N, Bjordal K, Groenvold M, Curran D, A B.

EORTC QLQ-C30 Scoring Manual. Brussels, : EORTC

Publications; 2001.

30. Greimel E, Kuljanic Vlasic K, Waldenstrom A, Duric V,

Jensen P, Singer S, et al. e European Organization for Research

and Treatment of Cancer (EORTC) Quality-of-Life questionnaire

cervical cancer module: EORTC QLQ-CX24. Cancer 2006;107:

1812-22.

31. Pastrana T, Wenk R, De Lima L. Consensus-Based Palliative

Care Competencies for Undergraduate Nurses and Physicians:

A Demonstrative Process with Colombian Universities. J

Palliat Med 2016;19:76-82.

32. J. K, B. T. Family Support for Older Persons in ailand:Challenges

and Opportunities. Population Studies Center, University of

Michigan; 2017 March 2017. Contract No.: Report 17-879.

33. Juraskova I, Butow P, Robertson R, Sharpe L, McLeod C,

Hacker N. Post-treatment sexual adjustment following cervical

and endometrial cancer: a qualitative insight. Psycho-oncology

2003;12:267-79.

34. Khalil J, Bellefqih S, Sahli N, Af M, Elkacemi H, Elmajjaoui

S, et al. Impact of cervical cancer on quality of life: beyond

the short term (Results from a single institution): Quality

of life in long-term cervical cancer survivors: results from a

single institution. Gynecol Oncol Res Pract 2015;2:7.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

278

Anantachai Jaisaby, M.D.*, Paweena Phaliwong, M.D.*, Sinart Prommart, M.D.*, Buppa Smanchat, M.D.*,

Kornkarn Bhamarapravatana, Ph.D.**, Komsun Suwannarurk, M.D.***,****

*Department of Obstetrics and Gynecology, Bhumibol Adulyadej Hospital, Royal ai Air Force, **Department of Preclinical Sciences, ***Department

of Obstetrics and Gynecology, Faculty of Medicine, ammasat University, ****Chulabhorn International College of Medicine, ammasat University,

Pathum ani 12000, ailand.

The Accuracy of Cervical Length for Prediction of

Delivery in Term Pregnancy Patients Presenting

with Labor Pain

Corresponding author: Paweena Phaliwong

E-mail: p_phaliwong@yahoo.com

Received 10 October 2018 Revised 22 April 2019 Accepted 14 May 2019

ORCID ID: http://orcid.org/0000-0002-5083-8560

http://dx.doi.org/10.33192/Smj.2019.42

ABSTRACT

Objective: To investigate the relationship between cervical length (CL) at gestational age (GA) 37 to 41+6 weeks

and delivery within 7 days in term pregnancy patients who presented with labor pain.

Methods: Term pregnancy subjects who attended antenatal clinic and delivered at Bhumibol Adulyadej Hospital,

between September 2017 and June 2018 were recruited in this study. Participants who met the inclusion criteria

received transvaginal ultrasound to measure CL. Clinical and delivery outcomes were followed. Sensitivity,

specicity, negative predictive value (NPV) positive predictive value (PPV) and accuracy were calculated to assess

the relationship between CL and delivery within 7 days.

Results: A total of 106 pregnant women were included in the analysis. e mean age was 26 years. ree-quarters

of the cases were nulliparous. e average gestational age at delivery and newborn birth weight were 39 complete

weeks and 3,100 grams, respectively. Seventy-six cases successfully delivered within 7 days while 30 patients had

to wait for more than 7 days before delivery. e receiver operating characteristic (ROC) curve was used to assess

the optimal cuto of CL. e CL less than 25 mm. gave sensitivity and specicity at 69% and 73%, respectively and

gave accuracy of 70.75% in prediction of delivery within 7 days.

Conclusion: Transvaginal CL measurement was useful to predict the time of spontaneous delivery and help clinicians

to advise the patients about their delivery plans.

Keywords: Cervical length; term gestation; labor (Siriraj Med J 2019; 71: 278-283)

INTRODUCTION

Labor pain at term was one of the most reported

clinical presentations that brought term pregnant women

to the delivery room.

At our facility, if the patient was

not in an active phase for delivery

, she would be placed

in clinical observation for 4 hours and then re-evaluated

via pelvic examination. Pregnant women with no cervical

progression would then be discharged from the observation.

However, such a procedure did not allow the attending

physician to estimate time to delivery.

Cervical length (CL) is an ultrasonographic assessment

of the cervix. It has been used widely in contemporary

obstetrics. It was used to predict the risk of preterm

delivery

3,4

and in women before induction of labor to

Jaisaby et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

279

Original Article

SMJ

predict induction outcome.

5,6

Few studies have examined

cervical length at term and attempted to correlate a related

cervical length with duration of spontaneous labor.

The aim of the present study was to assess the

relationship between cervical length measurement by

transvaginal ultrasonography and time to delivery within

7 days in term pregnancy with presenting labor pain.

MATERIALS AND METHODS

Term pregnant women who had labor pain and

visited labor rooms at Bhumibol Adulyadej Hospital

from September 2017 to June 2018 were recruited. e

ethics of the study were approved by the Hospital Ethics

Committee, Bhumibol Adulyadej Hospital (IRB 52/60).

Informed consent was obtained prior to transvaginal

ultrasonographic examination. Labor pain was dened

as regular uterine contraction (≥ 4 times/20 min).

Inclusion criteria were dened as: singleton pregnancy,

term pregnant women with gestational age (GA) ranging

from 37 to 41+6 weeks, cephalic presentation, intact

membranes and certain estimated delivery dates.

Transvaginal ultrasonographic (TVUS) cervical length

measurements and pelvic examinations were performed

by obstetrics and gynecology residents who trained on

proper CL measurement technique at time of labor room

visit. CL measurements performing the TVUS, using a

LOGIQ V3 (General electric company, USA) machine

equipped with a 8 MHz transvaginal probe.

e participant was placed in the dorsal lithotomy

position. A vaginal probe at 8 MHz was used to perform

an ultrasonic reading aer the participant’s bladder was

emptied. Endocervical canal length was measured as the

distance from the internal and external os in a straight

line. If funneling was presented, the cervical length was

measured as the distance between the apex of funneling

and the external os parallel to the cervical canal. ree

images were collected in each patient and the image with

the shortest CL was chosen for analysis. All recruited

patients were monitored until delivery.

Demographics and obstetrics variable data was

collected including patient’s age, race, height, body

mass index (BMI), CL, gravidity, parity, onset of labor

pain, GA at CL measurement, date and time of CL

measurement, GA at birth, date and time of birth, time

from CL measurement to delivery, fetal weight and route

of delivery.

e data was processed using SPSS statistical soware

version 18 (IBM, Armonk, NY, USA). for analysis.

Continuous variables were presented as mean and standard

deviation (SD). e p-value of less than or equal 0.05 was

used for statistical signicance. Curves of cervical length

for predication of delivery within 7 days were generated

using receiver operating characteristic (ROC) curves.

Diagnostic accuracy of CL was subcategorized and used

for analysis of sensitivity, specicity, positive predictive

value (PPV), negative predictive value (NPV), positive

likely hood ratio (LR+), negative likelihood ratio (LR-)

and accuracy.

RESULTS

A total of 120 pregnant women were enrolled.

Fourteen cases were excluded due to patients choosing

to deliver at other facilities. One hundred and six cases

were interviewed and followed up until delivery. All

cases that met inclusion criteria were included in the

study.

Of the remaining participants, 76 successfully delivered

within 7 days, whilst the remaining 30 cases, delivered

outside the 7 days window. ere were no statistically

signicant dierences between groups in age, BMI and

parity as represented in Table 1.

e participants who delivered at more than 7 days

had longer CL, longer time to delivery aer visit and

higher fetal birth weight than women who delivered

within 7 days with ratios of: 29/21.8 mm, 272/54 hours

and 3,202/3,061 grams.

ROC curve of cervical length for predicting delivery

within 7 days was presented in Fig 1. e appropriate

cut o point was chosen from ROC curves that gave a

suitable value. CLs less than 25 mm gave the sensitivity

and specicity in prediction delivery within one week at

a percentage of 69 and 73, respectively. e other cut-o

points are also represented in Table 2.

DISCUSSION

e main outcome of this study was the correlation

between CL and delivery date within 7 days prediction

of term pregnancy cases that presented at the hospital

with labor pain. Optimal CL cut-o was suggested at

less than 25 mm which gave appropriate sensitivity and

specicity (69% and 73%, respectively). All cases with

CL less than 15 mm had delivered their newborns within

7 days.

Bayramoglu et al

reported from France that

sensitivity and PPV of spontaneous labor within 7 days

by using CL cut o point at 24.5-29.5 mm were 75-82.9

and 54.8-95.5%, respectively. His work was conducted in

vertex and singleton term pregnancy. Half of participants

were nulliparous and CL cut o point decreased with the

higher GA (29.5 mm at GA 37 weeks, 27.5 mm at GA 38

weeks, 25.5 mm at GA 39 weeks and 24.5 mm at GA 40

weeks). Comparison to the present study at cut o point

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

280

TABLE 1. Demographic data of the study population.

TABLE 2. Diagnostic accuracy of CL to predict delivery within 7 days.

Within More P-value

Age (years) 26.43±6.29 25.60±6.16 0.538

BMI (kg/m

) 26.39±4.63 27.00±3.44 0.510

CL (mm) 21.8±7.4 29.0±6.3 <0.01

Time (hours) 53.92±42.26 271.99±91.49 <0.01

GA (days)

Visit 274.08±6.35 268.93±6.82 <0.01

Delivery 276.42±6.26 280.40±6.43 0.04

FBW (gm) 3061.84±393.22 3202.23±249.56 0.03

Nulliparous 51(76.1%) 16(23.9%) 0.18

Multiparous 25(64.1%) 14(35.9%)

Data were shown as mean ± SD (standard deviation), Within: delivery within 7 days from pelvic examination, More: delivery more than

7 days from pelvic examination

Abbreviations: BMI: body mass index, CL: cervical length, Time: time from examination to delivery, GA: gestational age, Visit: gestational

age at examination, Delivery: gestational age at delivery, FBW: fetal birth weight.

CL (mm)

< 15 < 20 < 25 < 30

Sensitivity 19.7 40.7 69.7 85.5

Specicity 100 96.6 73.3 36.6

PPV 100 96.8 86.8 77.3

NPV 32.9 39.1 48.8 50.0

LR+ 12.2 2.6 1.3

LR- 0.8 0.6 0.4 0.4

Accuracy 42.45 56.60 70.75 71.7

Data were shown as %,

Abbreviations: CL: cervical length, PPV: positive predictive value, NPV: negative predictive value, LR+: positive likelihood ratio,

LR-: negative likelihood ratio

Jaisaby et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

281

Original Article

SMJ

Fig 1. ROC curve of CL to predict delivery within 7 days.

Abbreviations: ROC = receiver operating characteristic, CL= cervical length (mm)

Fig 2. Subject selection and exclusion.

Participants: Term pregnancy who had labor pain, within 7 days: delivery within 7 days from pelvic examination, More: delivery more than

7 days from pelvic examination, Drop out: Term pregnancy who had delivery at other hospitals.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

282

at 25 mm, sensitivity and PPV of Bayramoglu were higher

than the present study (75.5/69.5,90.9/73%, respectively).

e discrepancy may be due to dierences in patient

population. e population in Bayramoglu’s study was

majority French, whilst this study was conducted on a

majority ai population (with 63% nulliparous cases).

Tolaymat et al

performed transvaginal ultrasonography

to measure CL in 120 pregnant women with singleton

and gestations at 37 to 40 weeks to assess the relationship

between CL and spontaneous labor within 7 days. Tolaymat

suggested CL less than 25 mm that gave sensitivity and

PPV of 77.5 and 50.8%, respectively. Our study suggested

a cut-o point at the same value as Tolaymat’s work (less

than 25 mm). e present study had less sensitivity than

the Tolaymat’s work at of 69.7/77.5 %, while PPV of the

present study was higher than Tolaymat’s work at of

86.8/50, respectively. e dierence was due to dierences

in patients between the two studies. ree quarters of

subjects in Tolaymat’s study were of Hispanic race and

half of cases were nulliparous women. While this study

was conducted in ai women and all participants had

labor pain that increased chances of imminent delivery,

compared to pregnant women who had not yet presented

with labor pain. e overall prevalence of spontaneous

labor within 7 days in Tolaymat’s work was low (32.8%)

while the present study was 71.6% due to subjects enrolled

in this study having presented labor pain prior to the

research registration.

In the year 2010, Miura et al reported the CL

prediction of labor in Japanese pregnant women. Parity

and gestational age of Miura’s work were similar to the

present study. His work recruited the cases who had no

labor pain like Bayramoglu’s and Tolaymat’s work.

9,12

Miura’s work concluded that CL measurement could

predict the spontaneous labor within 7 days. His work

supported our study.

Systematic review study by Saccoon et al.

assessing

the accuracy of TVU CL in the prediction of spontaneous

onset of labor in singleton gestations revealed the accuracy

of CL for prediction of spontaneous labor within 7

days. Pooled sensitivities and speciﬁcities yielded the

conclusion that the higher CL had better sensitivity;

the lower CL had better speciﬁcity. Saccoon’s work

consisted of European cases (90%) who had the same

inclusion criteria with this study. All cases in his work

had no clinical presentation of labor pain. e present

study recruited pregnant women with the same criteria

who had clinical labor pain. However, the Saccoon’s

work still supported the present study that CL was an

important factor to predict spontaneous delivery.

Mukherji et al

conducted the serial ultrasound for

CL measurement in pregnant women at GA 36-40 weeks.

Mukherji’s population consisted of Indian nulliparous

pregnant women. His literature concluded that a single

CL could not predict spontaneous labor. CL cut of point

at GA 38 weeks from transabdominal and transvaginal

ultrasonography in Mukherji’s study were 35 mm and

31 mm, respectively can predict post-dated pregnancy.

Mukherji conducted his investigation in pregnant women

who had no labor pain. In the present study, we studied

women who had labor pain. Our study showed that

single CL measurement in any gestational age of term

pregnancy can predict delivery.

Most CL studied in prediction of spontaneous onset

of labor were done in term pregnancy with no labor

pain.

9-12

is study was the study that evaluated term

pregnancy with labor pain and a wide range of gestational

age (37–41+6 weeks). is study suggested a CL value

that predicted delivery within 7 days for patients 37+

week gestation with acceptable robustness.

Pregnant population using obstetric service at

Bhumibol Adulyadej hospital could be divided into

two subgroups, normally local residents and people who

work in the area but originated upcountry. e results

from our investigation can be used to advise patients

for their delivery plan. ose who are discharged from

labor rooms were normally concerned about how long

would it be before their time to deliver. Women with

short CL should be adviced to prepare to stay near the

hospital or notify their care givers to stand by for future

visit to a labor room. A long CL can be advised that the

labor might take longer than one week.

Expecting mothers sometimes choose to deliver

their ospring in their hometown so as to benet from

family care and support. A few patients had le our

unit aer their initial reports of labor pain to deliver

elsewhere. Our CL cut-o allowed a good prediction of if

the baby would be due within 7 days or not. Armed with

this certainly, patients can choose if they would stay in

Bangkok to deliver or allow them enough time to travel

to their hometown destination for the birth.

ACKNOWLEDGMENTS

e study was funding supported by Bhumibol

Adulyadej Hospital, Royal ai Air Force, ailand in year

2016. I would like to express my sincere appreciation to

Flt.Lt. Amphol Saisiriwachakul for his statistical assistance.

Conict of Interests: None

Jaisaby et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

283

Original Article

SMJ

REFERENCES

1. ACOG Committee Opinion. Deﬁnition of term pregnancy.

Committee opinion Number 579. Obstet Gynecol 2013;122:

1139-40.

2. orp JM Jr, Laughon SK. Clinical Aspects of Normal and

Abnormal Labor. In: Creasy RK, Resnik R, editors. Creasy and

Resnik’s Maternal-Fetal Medicine: Principles and Practice,

ed. Philadelphia: Saunders Elsevier; 2014.p.673-706.

3. Lim K, Butt K, Crane JM. No. 257-Ultrasonographic Cervical

Length Assessment in Predicting Preterm Birth in Singleton

Pregnancies. J Obstet Gynaecol Can 2018;40: e151-64.

4. Khalifeh A, Berghella V. Universal cervical length screening

in singleton gestations without a previous preterm birth:

ten reasons why it should be implemented. Am J Obstet Gynecol

2016;214:603.e1-5.

5. Verhoeven CJ, Opmeer BC, Oei SG, Latour V, van der Post

JA, Mol BW. Transvaginal sonographic assessment of cervical

length and wedging for predicting outcome of labor induction

at term: a systematic review and meta-analysis. Ultrasound

Obstet Gynecol 2013;42:500-8.

6. Hatﬁeld AS, Sanchez-Ramos L, Kaunitz AM. Sonographic cervical

assessment to predict the success of labor induction: a systematic

review with meta-analysis. Am J Obstet Gynecol 2007;197:

186-92.

7. Saccone G, Simonetti B, Berghella V. Transvaginal ultrasound

cervical length for prediction of spontaneous labour at term:

a systematic review and meta-analysis. BJOG 2016;123:16-22.

8. e American College of Obstetricians and Gynecologists.

American Institute of Ultrasound in Medicine and Society for

Maternal-Fetal Medicine. Committee opinion no. 611: method

for estimating due date. Obstet Gynecol 2014;124:836-7.

9. Tolaymat LL, Gonzalez-Quintero VH, Sanchez-Ramos L,

Kaunitz A, Wludyka P, O’Sullivan MJ, et al. Cervical length

and the risk of spontaneous labor at term. J Perinatol 2007;27:

749-53.

10. Meijer-Hoogeveen M, Van Holsbeke C, Van Der Tweel I,

Stoutenbeek P, Visser GH. Sonographic longitudinal cervical

length measurements in nulliparous women at term: prediction

of spontaneous onset of labor. Ultrasound Obstet Gynecol

2008;32:652-6.

11. Miura H, Ogawa M, Hirano H, Tanaka T. Time-related changes

of cervical length as a predictor of labor onset within one week

using transvaginal ultrasonography. Acta Obstet Gynecol

Scand 2010;89:757-61.

12. Bayramoglu O, Arslan M, Yazici FG, Erdem A, Erdem M,

Bayramoglu K, et al. Prediction of spontaneous onset of labor

at term: the role of cervical length measurement and funneling

of internal cervical os detected by transvaginal ultrasonography.

Am J Perinatol 2005;22:35.

13. Mukherji J, Bhadra A, Ghosh SK, Hazra A, Anant M, Bhattacharya

SK, et al. Cervical length measurement in nulliparous women

at term by ultrasound & its relationship to spontaneous onset

of labour. Indian J Med Res 2017;146:498-504.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

284

Siriwan Piyapittayanan, M.D.*, Natthawut Jarunnarumol, M.D.*, Panai Laohaprasitiporn, M.D.**, Suthon

Noiwatana, B.Sc.*, Orasa Chawalparit, M.D.*

*Department of Radiology, **Department of Orthopedic Surgery, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.

Diagnostic Performance of Short MR-Neurography

Protocol for Brachial Plexus Injuries

Corresponding author: Orasa Chawalparit

E-mail: oak_art@yahoo.com

Received 18 September 2018 Revised 7 March 2019 Accepted 14 March 2019

ORCID ID: http://orcid.org/0000-0001-7073-3750

http://dx.doi.org/10.33192/Smj.2019.43

ABSTRACT

Objective: To study the diagnostic performance of MR neurography (MRN) for brachial plexus injuries and to

optimize the protocol using clinical contexts as the reference standard.

Methods: ere were 21 patients with brachial plexus injury who were scheduled for conventional myelography. A

brachial plexus MRN including T2-weighted image-high resolution (T2WI/HR), mDIXON and diusion weighted

image was performed prior to a conventional myelography on the same day. e results of the conventional

myelography and the MR imaging were recorded and compared, with the clinical contexts as the reference standard.

e sensitivities, specicities, accuracies, false positive and false negative were calculated and compared.

Results: e accuracy, sensitivity, specicity, false positive and false negative of the conventional myelography were

69.52%, 73.61%, 60.61%, 19.70% and 48.72%, respectively. e diagnostic performance of T2WI/HR were 72.00%,

78.26%, 58.06%, 19.40% and 45.45% for T2WI/HR, respectively which were comparable to those of conventional

myelography. e accuracy, sensitivity, specicity, false positive and false negative of the combination of T2WI/HR

and mDIXON were 78.00%, 97.10%, 35.48%, 22.99% and 15.38%, respectively which yielded the highest accuracy.

Conclusion: MRN with the combination of T2WI/HR and mDIXON was superior to conventional myelography

for the evaluation of brachial plexus injuries due to its shorter processing time, the lack of a need for contrast

medium administration, its noninvasive nature, and the provision of information about both preganglionic and

postganglionic injuries.

Keywords: MR-neurography; high-resolution MRI; brachial plexus injury; nerve root avulsion (Siriraj Med J 2019;

71: 284-289)

Piyapittayanan et al.

INTRODUCTION

Traumatic brachial plexus injuries in adults mostly

occur in young people aged between 20 and 30 years.

Preganglionic lesions result from the avulsion, or the

tearing o, of the insertion of the nerve root from the

spinal cord, while more-distal nerve ruptures within the

brachial plexus are dened as postganglionic lesions. e

management and prognosis of brachial plexus injuries

depend on the degree of damage and the site of the injuries,

with a poor prognosis for preganglionic injuries. Many

imaging modalities are available to not only dierentiate

between preganglionic and postganglionic lesions (such

as standard myelography, CT myelography, conventional

MRI, MR myelography and MR neurography), but also

determine the severity of the injuries. Conventional

myelography and CT myelography have long been used

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

285

Original Article

SMJ

for the assessment of preganglionic root injuries. However,

myelography is an invasive procedure, exposes patients

to radiation, and carries the risk of adverse reactions

to the iodinated contrast medium. Because of its non-

invasive nature and ability to evaluate both the proximal

and distal parts of the brachial plexus, MR neurography

(MRN) has recently been introduced. MRN comprises

a set of evolving protocol techniques to evaluate many

abnormalities of the brachial plexus.

To achieve the

objectives of the MRN, good and clear anatomical or

structural information on the entire brachial plexus needs

multiple planes and pulse sequences. e high-quality of

MR imaging, provided by its high resolution, good fat

suppression, and 3D imaging (short inversion-recovery

sequence or mDIXON), results in excellent images of

the peripheral brachial plexus structure. Furthermore,

diusion-weighted neurography can be used to evaluate

postganglionic brachial plexus lesions. However, to

evaluate preganglionic brachial plexus injuries, a high-

resolution heavily T2-weighted pulse sequence oers

comparable information to CT myelography.

e purposes of this study were to optimize the

protocol of the brachial plexus MRN used by our institute

for brachial plexus injuries, and to study the diagnostic

performance of the protocol, with clinical contexts used

as the reference standard.

MATERIALS AND METHODS

Patient selection

is prospective study was performed between

February 2015 and August 2016. It was approved by the

ethics committee of the institute before patients were

enrolled (Si 427/2015). A total of 21 patients tted the

inclusion criteria, which including age over 18 years,

diagnosed with a brachial plexus injury, and scheduled for

conventional myelography. Written informed consents

were obtained from all patients before enrolment in the

MRI study.

MR imaging

All patients underwent an MRI just prior to conventional

myelography. All MRI examinations were performed

on a 3.0-T MR unit (Ingenia, Philips Medical Systems,

Best, the Netherlands), with a 16-element, head and

neck matrix coil. We designed three pulse sequences of

MRN brachial plexus protocol for the setting of brachial

plexus injury and covering C4-T1 levels. e 3 MR pulse

sequences acquired were axial T2-weighted image-high

resolution (repetition time = 3000; echo time = 140;

matrix, 180 x 178; eld of view, 180 mm x 180 mm; turbo

spin echo factor, 38; Sense factor, 1.7; slice thickness,

2.0 mm, no gap; number of signal average, 2); coronal

T2-weighted image-mDIXON (repetition time = 2721;

echo time = 90; matrix, 444 x 217; eld of view, 400 mm

x 252 mm; Sense factor, 2; slice thickness, 2.0 mm, gap,

0.2 mm; number of signal average, 2); and axial diusion

weighted image (repetition time = 9000; inversion time

= 250; echo time = 67; matrix, 112 x 112; eld of view,

300 mm x 300 mm; Sense factor, 2.5; slice thickness,

3.0 mm, no gap; number of signal average, 6; b factor =

0 and 800 s/mm

; directions of gradient sampling, 16).

e scan time took 13 minutes for T2-weighted image-

high resolution, 5 minutes for mDIXON and 8 minutes

for diusion weighted image; the total scan time was

therefore 26 minutes.

Conventional myelography

All patients underwent conventional myelography.

Under a sterile technique, the subarachnoid space at the

L3-L4 intervertebral level was punctured by a spinal needle.

en, 10 ml of water-soluble, non-ionic contrast medium

(iopamiro) was injected via the spinal needle. e cervical

myelography was performed with uoroscopic guidance

for the AP, lateral and both oblique views. Aer that,

patients had to rest with their head elevated to observe

for any immediate complications. It took about 90-120

minutes to perform the conventional myelography.

e results of the conventional myelography and MR

imaging were recorded as a normal or abnormal nerve

from C5-T1 levels, bilaterally. With the conventional

myelography, abnormal nerve roots were dened as the

loss of normal conguration (i.e. blunting) of nerve roots

and/or demonstrable pseudomeningocele (Fig 1). As for

the MR imaging, abnormal nerves were dened as the

loss of normal nerve alignments, distributions and/or

architectures in either the preganglionic or postganglionic

portions (Fig 1, Fig 2, Fig 3). e preganglionic nerve root

was interpreted from T2-weighted image-high resolution,

and the postganglionic nerve root was observed by using

T2-weighted image-mDIXON and diusion weighted

image. e abnormality of the nerve roots was interpreted

and compared with the normal contralateral side in both

the conventional myelography and MR imaging.

Two neuroradiologists separately reviewed the

MRN study and the conventional myelography, and they

were blind to the clinical data. Disagreements between

the readers were resolved by consensus.

Statistical analysis

e results of conventional myelography and each

sequence of MR imaging were correlated, with the clinical

contexts being used as the gold standard. e accuracies,

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

286

Fig 1. A 28-year-old man with history of brachial plexus injury. (A) Conventional

myelography reveals no demonstration of normal le C5-T1 preganglionic nerve

roots with pseudomeningocele at le C8 nerve root. (B) T2-weighted image-high

resolution, no detectable normal ventral and dorsal le C5 nerve roots, representing

nerve root avulsion.

Fig 2. Coronal MIP image of mDIXON of a 20-year-old man

reveals abnormal contour with irregularity of le C5 and C6

roots and total avulsion of le C7 and C8 roots with associated

pseudomeningoceles at le C7 and C8 roots.

Fig 3. Coronal MIP image of diusion weighted image of a 28-year-old man reveals abnormal contour with total irregularity of le C5-T1

nerves (B) as compared with the contralateral normal appearance (A)

Piyapittayanan et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

287

Original Article

SMJ

sensitivities, specicities, and the false positive and false

negative values were calculated. All data analyses were

performed using SPSS soware version 18.0 (SPSS Inc.,

USA).

RESULTS

e 21 patients with brachial plexus injury in this

study were investigated 2 to 13 months aer injury (mean:

5.1 months). e mean age of the patients was 27.1 years,

with an age range of 16 to 39 years; 19 were male, and

2 were female. A total of 210 nerves of the brachial plexus

(bilateral C5-T1) were interpreted for their individual

modalities/MR sequences. In the case of T2-weighted

image-high resolution, one patient was excluded due to

motion artifact during imaging, leaving 200 nerve roots.

e number of nerve abnormalities evaluated for each

nerve level and for the dierent modalities are at Table 1.

Table 2 summarizes the diagnostic performance

(specically, the accuracy, sensitivity, specicity, false

positive and false negative values) of each diagnostic tool

(conventional myelography, T2-weighted image-high

resolution, mDIXON, diusion weighted image, and the

combination of T2-weighted image-high resolution and

mDIXON and T2-weighted image-high resolution and

diusion weighted image). e diagnostic performances of

conventional myelography and T2-weighted image-high

resolution were similar. e combination of T2-weighted

image-high resolution and mDIXON had the highest

accuracy for the diagnosis of brachial plexus injury.

A comparison of the critical performance factors

of the conventional myelography and the MRI study

is at Table 3. e benets of the MRI study over the

conventional myelography were its shorter processing

time, lack of radiation exposure, lack of the need for

contrast medium administration, and the ability to evaluate

both pre- and postganglionic lesions. However, the cost

of using an MRI study was much higher than that for

conventional myelography.

TABLE 1. Number of nerve abnormalities, by nerve level, for dierent modalities.

TABLE 2. Diagnostic performance of imaging modalities in brachial plexus injury compared with clinical contexts.

Diagnostic tool (total patients) Number of nerve abnormalities

C5 C6 C7 C8 T1

Clinical (21) 17 19 15 11 10

Myelography (21) 5 17 18 16 10

T2WI/HR (20) 9 15 17 16 10

mDIXON (21) 17 21 20 18 9

DWI (21) 14 19 21 20 10

Abbreviations: T2WI/HR = T2-weighted image-high resolution, DWI = diusion weighted image

Diagnostic tool Accuracy Sensitivity Specicity False positive False negative

MYELOGRAHY 69.52% 73.61% 60.61% 19.70% 48.72%

T2WI/HR 72.00% 78.26% 58.06% 19.40% 45.45%

mDIXON 76.19% 91.67% 42.42% 22.35% 30.00%

DWI 67.62% 84.72% 30.30% 27.38% 52.38%

T2WI/HR + mDIXON 78.00% 97.10% 35.48% 22.99% 15.38%

T2WI/HR + DWI 71.00% 91.30% 25.81% 26.74% 42.86%

Abbreviations: T2WI/HR = T2-weighted image-high resolution, DWI = diusion weighted image

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

288

TABLE 3. Comparison of the critical performance factors of conventional myelography and MRI study.

Factor Conventional myelography MRN study

Time [Minutes] 90-120 26

Cost [Baht] 3300 17 600

Contrast medium Yes No

Radiation exposure Yes No

Evaluation Preganglionic injury Both pre- and postganglionic injury

DISCUSSION

Most ai patients with brachial plexus injuries

are male and of a younger age because the injuries

commonly occur in motor vehicle accidents, as has been

reported in multiple studies.

4-6

Brachial plexus injuries

are classied into three categories: preganglionic lesions,

postganglionic lesions, and a combination of both.

Multiple root injuries are also seen in most patients.

Adequate surgical technique requires the dierentiation

of preganglionic lesions from postganglionic lesions.

is is because the surgical procedure for preganglionic

lesions consists of neurotization (nerve transfer), whereas

postganglionic lesions may be treated by neurolysis or

repair with nerve graing.

8,9

e accurate diagnosis of

brachial plexus injuries before surgery remains challenging.

Many methods are used to evaluate the injury pattern,

including physical examination, a nerve conduction

study, electromyography, conventional myelography, CT

myelography, and MRI. However, none of those methods

is considered the reference standard.

Some studies have

used the results of operative ndings or have combined

them with intraoperative electrodiagnostic studies to

establish a reference standard.

4,8,11

Some studies have

also used CT myelography as the gold standard for the

imaging of nerve root avulsion.

7,9

However, in experienced

surgeons at our institute use physical examination to

diagnose brachial plexus injuries in combined modality

with a nerve conduction study or an electromyogram test

in some cases. Sometimes, surgeons cannot identify the

entire course of the nerves, probably due to brosis, so

making a denite diagnosis by using operative ndings

is not possible. erefore, we used the clinical contexts

as the reference standard in our study.

For the evaluation of preganglionic root injuries,

conventional myelography is the diagnostic tool currently

used by our institute. However, it is an invasive procedure,

and it takes considerably longer to perform than an MRI

study. D. Somashekar et al. found that a high-resolution

MRI had a similar sensitivity to CT myelography when

evaluating neonatal brachial plexus palsy.

According to

our results, T2-weighted image-high resolution was also

comparable to conventional myelography in terms of its

accuracy, sensitivity, specicity, and false positive and

false negative values. In addition, T2-weighted image-

high resolution is noninvasive and needs a noticeably

shorter performance time (approximately 13 minutes)

than conventional myelography (approximately 90-120

minutes). Furthermore, the patients are not exposed to

the contrast medium, have no risk of iodine contrast

allergy, and do not suer from any side eects or radiation

exposure. erefore, we suggest that T2-weighted image-

high resolution replace conventional myelography for

the diagnosis of brachial plexus injuries. However, in our

study, both conventional myelography and T2-weighted

image-high resolution had a diagnostic accuracy of

preganglionic nerve root injury of only 69.52% and

72.00%, respectively. e inaccurate interpretations are

probably due to partial nerve root avulsions, intradural

broses, scars, or postganglionic lesions.

Conventional myelography and T2-weighted image-

high resolution demonstrate preganglionic injuries,

whereas mDIXON and diusion weighted image detect

postganglionic injuries. In our study, mDIXON had a

slightly higher diagnostic yield than diusion weighted

image in the diagnosis of postganglionic brachial plexus

injuries. In addition, mDIXON can provide accurate,

indirect signs of root avulsion injuries, namely, the regional

denervation of muscle changes which are seen as muscle

atrophy, or a high T2 signal change, consistent with the

article by A. Chhabra et al.

erefore, mDIXON may

have more benets than diusion weighted image in the

detection of postganglionic lesions. However, it is also

feasible to apply diusion tensor imaging by improving

the spatial resolution in order to obtain better information

Piyapittayanan et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

289

Original Article

SMJ

on nerve abnormalities than diusion weighted image,

including microstructural changes and ber tractography.

By contrast, mDIXON only provides information on

the structural changes of nerves. e combination of

T2-weighted image-high resolution with mDIXON or

diusion weighted image provided better results for the

detection of brachial plexus injuries than mDIXON or

diusion weighted image alone because a combination

can detect lesions in both pre- and postganglionic injuries.

Given that the combination of T2-weighted image-high

resolution and mDIXON had the highest diagnostic

yield, we recommend it be used as the diagnostic tool

for brachial plexus injuries in preference to conventional

myelography.

e present study used the clinical contexts as the

reference standard. e low specicities and high false-

positive values in the results were probably due to the

following:

1. At our institute, expert hand orthopedists use

the clinical contexts as the gold standard, and only total

avulsion tears are considered as abnormal ndings. In

other words, partial tears are excluded. However, there

is a view that the clinical contexts might be inaccurate.

Some believe that surgery might be more accurate, but in

practice, orthopedists cannot explore the brachial plexus

for each root level in all patients. is is because some

patients need no surgery, and if surgery is performed, the

brachial plexus still cannot be well identied, not only

due to the anatomical defect but also brosis. us, there

is still doubt as to what should be used for the denition

of the gold standard.

2. ere might be overinterpretation by the radiologists

as the study dened abnormal ndings in the myelography

and MRI as a loss of normal nerve root alignment,

distribution and/or architecture relative to the normal

contralateral side, which may have included a partial

tear at the site of the nerve injuries.

LIMITATIONS

e reference standard in our study only used a

physical examination, which might be an inaccurate

reference. Using a combination of physical examination,

surgical ndings, and electromyogram as the reference

standard might improve the diagnostic yield.

In addition, the reference standard used in our study

could not dierentiate preganglionic from postganglionic

lesions.

CONCLUSION

We recommend that the combination of T2-weighted

image-high resolution and mDIXON be adopted as

the diagnostic tool for the diagnosis of brachial plexus

injuries in preference to conventional myelography. is

is because of its faster processing time, the lack of the need

for contrast medium administration, and its usefulness

in evaluating both preganglionic and postganglionic

injuries.

REFERENCES

1. Silbermann-Homan O, Teboul F. Post-traumatic brachial

plexus MRI in practice. Diagn IntervImaging2013;94:925-43.

2. Mallouhi A, Marik W, Prayer D, Kainberger F, Bodner G,

Kasprian G. 3T MR tomography of the brachial plexus: structural

and microstructural evaluation. Eur J Radiol2012;81:2231-45.

3. Somashekar D, Yang LJ, Ibrahim M, Parmar HA. High-

resolution MRI evaluation of neonatal brachial plexus palsy:

A promising alternative to traditional CT myelography. AJNR

Am J Neuroradiol2014;35:1209-13.

4. Balakrishnan G, Kadadi BK. Clinical examination versus routine

and paraspinal electromyographic studies in predicting the site

of lesion in brachial plexus injury. J Hand Surg Am2004;29:140-3.

5. Moran SL, Steinmann SP, Shin AY. Adult brachial plexus

injuries: mechanism, patterns of injury, and physical diagnosis.

Hand Clin2005;21:13-24.

6. Carvalho GA, Nikkhah G, Matthies C, Penkert G, Samii M.

Diagnosis of root avulsions in traumatic brachial plexus injuries:

value of computerized tomography myelography and magnetic

resonance imaging. J Neurosurg1997;86:69-76.

7. Yoshikawa T, Hayashi N, Yamamoto S, Tajiri Y, Yoshioka N,

Masumoto T, et al. Brachial plexus injury: clinical manifestations,

conventional imaging ndings, and the latest imaging techniques.

Radiographics2006;26 Suppl 1:S133-43.

8. Walker AT, Chaloupka JC, de Lotbiniere AC, Wolfe SW,

Goldman R, Kier EL. Detection of nerve rootlet avulsion on

CT myelography in patients with birth palsy and brachial

plexus injury aer trauma. AJR Am J Roentgenol1996;167:

1283-7.

9. Giure JL, Kakar S, Bishop AT, Spinner RJ, Shin AY. Current

concepts of the treatment of adult brachial plexus injuries. J

Hand Surg Am2010;35:678-88.

10. Caporrino FA, Moreira L, Moraes VY, Belloti JC, Gomes dos

Santos JB, Faloppa F. Brachial plexus injuries: diagnosis

performance and reliability of everyday tools. Hand Surg2014;19:7-11.

11. Abul-Kasim K, Backman C, Bjorkman A, Dahlin LB. Advanced

radiological work-up as an adjunct to decision in early

reconstructive surgery in brachial plexus injuries. JBrachialPlex

Peripher Nerve Inj2010;5:14.

12. Chhabra A, await GK, Soldatos T, akkar RS, Del Grande F,

Chalian M, et al. High-resolution 3T MR neurography of the

brachial plexus and its branches, with emphasis on 3D imaging.

AJNR Am J Neuroradiol2013;34:486-97.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

290

Ruttanaporn Kongkar, R.N., Ph.D. (Candidate)*, Wanpen Pinyopasakul, R.N., Ph.D.**, Kanaungnit Pongthavornkamol,

R.N., Ph.D.**, Piyapat Dajpratham, M.D.***, Pisamai Orathai, R.N., Ph.D.****

*Joint Program between Faculty of Nursing and Ramathibodi School of Nursing, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok

10400, **Department of Medical Nursing, Faculty of Nursing, Mahidol University, Bangkok 10700, ***Department of Rehabilitation Medicine, Faculty

of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ****Ramathibodi School of Nursing, Faculty of Medicine Ramathibodi Hospital,

Mahidol University, Bangkok 10400, ailand.

The Determinants of Quality of Life in Thai Family

Caregivers of Stroke Survivors

Corresponding author: Wanpen Pinyopasakul

E-mail: wanpen.pin@mahidol.ac.th

Received 9 November 2018 Revised 19 April 2019 Accepted 14 May 2019

ORCID ID: http://orcid.org/0000-0001-9318-6809

http://dx.doi.org/10.33192/Smj.2019.44

ABSTRACT

Objective: Stroke survivors suer from multiple health problems, leading to physical and psychological impairments

or disabilities. Largely, family caregivers encounter stressful situations when providing stroke care at home, however,

little is known about their quality of life (QOL) and its determinants. is study aimed to identify the determinants

of QOL among ai family caregivers of stroke survivors.

Methods: e sample consisted of stroke caregivers from two tertiary hospitals in the central region of ailand.

Data collection using a set of questionnaires was performed at the out-patient department of each hospital. Data

analysis included descriptive statistics and path analysis to examine the hypothesized relationship between the

study variables and QOL.

Results: A total of 300stroke caregiverswere recruited. eir age ranged from 20-84years old, and mostly were

female (76%).Determinants of QOL among family caregivers were caregivers’ age, caregivers’ income and care

burden. Care burden was the strongest predictor which explained 64 % of the total variance (p<.001).

Conclusion: Sociodemographic factors impact on QOL among ai family caregivers of stroke survivors. Caregivers’

burden remains a serious issue, especially for those at an advanced age with low income who provide the care for

severely dependent stroke survivors. erefore, nurses and health care team should be aware of these determinants

and develop a family intervention program to support them so as to improve their QOL.

Keywords: Family caregiver; stroke survivor; quality of life; determinants (Siriraj Med J 2019; 71: 290-296)

Kongkar et al.

INTRODUCTION

Stroke is a serious global health problem. Each year,

more than 6.5 million adults die from stroke and another

5 million are le permanently disabled.

Family caregivers

provide most of the home-based care for stroke patients.

Previous studies in other countries suggest that the

provision of stroke care can be a signicant stressor in the

lives of family caregivers, negatively aecting their quality

of life (QOL). e denition of QOL as proposed by the

World Health Organization

(an individuals’ perception

of their physical health, psychological well-being, level

of independence, and social relationships in the context

of their sociocultural environment) was used to guide

this study. Previous cross-national literature involving

family caregivers of stroke survivors reported a negative

relationship between measures of QOL and perceived

caregiver burden.

3,4,5

Some but not all studies have also

reported a relationship between dependency for activities

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

291

Original Article

SMJ

of daily living (ADLs), age, income, social support and

decreased QOL scores.

3,4,6

Cultural beliefs and values

may account for some of the discrepant ndings across

studies.

Currently, little data are available regarding

QOL in the context of ai family caregivers of stroke

survivors, providing little for health care providers and

policymakers to develop strategies to develop family-

centered support.

Similar to an American study with caregivers of

stroke survivors

, we used the Stress Process Model (SPM)

to frame this study. is model is composed of four

domains, including background and context, stressors,

mediators, and outcomes. Within a ai context, we

considered the stressor to be the degree of dependency

for ADLs, potential mediators to be age, income, coping

style and social support, and the measured outcome was

the family caregiver’s QOL score. e primary aim of

this study was to examine family caregivers’ QOL within

a ai context and its relationship with factors that have

been previously reported to inuence QOL in family

caregivers from other countries who provide daily care

to their family member who has survived a stroke.

MATERIALS AND METHODS

A cross-sectional study design was used to examine

the variables of interest and their potential impact on

family caregivers’ QOL. Approval for this study was

obtained from the Institution Review Board for the

protection of human subjects from Mahidol University

(Si 251/2558). e inclusion criteria for family caregivers

were adults aged 18 years and older; self-identied as a

primary caregiver that lived with the survivor and provided

unpaid continuous care for a family member that had

experienced a stroke at least 6 months prior to the study;

and those able to speak, read, and understand ai. e

calculated sample size, using the method proposed by

Hair and colleagues

, to meet our primary objective was

estimated as requiring 300 participants.

Data were collected from family caregivers at the

out-patient department (OPD) of two tertiary care

hospitals located in an urban area in the central region

of ailand. e rst hospital is a tertiary hospital under

the Ministry of Education whereas the second hospital is

a tertiary hospital under the Ministry of Public Health.

Both hospitals are selected because they are teaching

hospitals, referral centers with certied stroke centers

that could represent characteristics of stroke survivors

and family caregivers throughout the country. A single

interview was completed in a private area of the OPD in

each hospital, either before or aer the patient and family

caregiver were seen in follow-up by their physician. Aer

obtaining informed consent participants were asked

questions from a sociodemographic form and asked

each item from a series of standardized questionnaires.

Sociodemographic data included age, marital status, and

perceived income suciency. e Barthel Activities of

Daily Living Index (BAI)

ai version

was used to

assess the ADL dependency of the stroke survivors. It is

composed of 10 items, with the total score ranging from 0

to 20. Lower scores indicate high dependency. e Burden

Interview (BI) developed by Zarit

and translated into ai

by Mapi Research Trust

was used to measure caregiver

burden. Total BI scores range from 0 to 88, with higher

scores indicative of greater perceived burden. Coping

strategies used by the family caregiver were identied

using the Jalowiec Coping Scale (JCS),

which had been

previously translated into ai.

e JCS is divided into

three subscales: confrontive coping (constructive problem

solving; 13 items), emotive coping (expressing emotions as

a stress reliever; 9 items), and palliative coping (activities

to make one feel better; 14 items). Total coping scores

range from 36 to 180; higher scores indicative of more

frequent use of that coping strategy. e Social Support

Questionnaire (SSQ)

was translated and modied for

cultural relevance

and was used to assess perceived social

support from three potential sources: family members

(5 items), relatives and friends (5 items), and healthcare

providers (5 items). Total scores range from 0 to 60,

with higher scores indicative of greater social support.

QOL was measured by the World Health Organization

Quality of Life Instrument BREF (WHOQOL-BREF)

ai version.

e rst two items rate general health

satisfaction and overall QOL, while the other 24 items

gather information about physical health (7 items),

psychological health (6 items), social relationships (3

items), and the environment (8 items). e total QOL

scores range from 26 to 130; higher scores indicative of

greater perceived QOL. Reliability was tested for each

of the instruments used in the study; with Cronbach’s

alpha coecients calculated for each (BAI = 0.89; BI =

0.94; JCS = 0.84; SSQ = 0.94; and WHOQOL-BREF =

0.88).

Data analysis

e data in the present study were analyzed using

descriptive statistics to delineate the sociodemographic

characteristics of the family caregivers and the stroke

survivors. Student t-tests were used to examine for gender

and age group dierences in the study variables. is was

followed by calculating Pearson’s correlation coecients

to examine the relationships between each variable of

interest and the QOL score. Finally, path analysis [using

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

292

the Linear Structural Relationship (LISREL) 8.72 Program

(Student version, Scientic Soware International, Chicago,

IL, USA)] was used to develop a best t model. e

indicators that were used to establish model t, included:

1) the non-signicant value of chi-square (χ

); 2) the

ratio of chi-square (χ

) / degree of freedom (df) less than

2; 3) goodness of t (GFI) greater than 0.9; 4) adjusted

goodness of t (AGFI) greater than 0.9; 5) comparative

t index (CFI) greater than 0.95; 6) root mean square

error of approximation (RMSEA), and standard root

mean square residual (SRMR) less than 0.05.

Of note,

the study variables in this study were not normally

distributed. erefore, the robust maximum likelihood

(RML) method using an asymptotic covariance matrix

was used for input10 and the Satorra-Bentler scaled chi-

square was used to indicate the value of the chi-square

(χ

) for measuring the model t with the empirical data.

RESULTS

e ages of the participants reected a broad range

from 20 - 84 years old (M = 52.32, SD = 13.59). When

examining the number of participants in each age group,

the result showed that 67.7% were adults (20 - 59 years

old) and 32.3% were older persons (60 years old and

older). Age group dierences between adults and older

persons were not signicant in the scores associated

with coping, caregiver burden and QOL. Similar to the

majority of other studies regarding caregivers, most of the

participants in this study were female (76%) and married

(69.7%). Two-thirds of the caregivers considered they

had sucient income and also indicated that they had

secondary caregivers that helped them (mainly other

family members). More than half of the participants

(56%) reported physical health problems (musculoskeletal

pain, stress and anxiety, and insomnia were the most

frequent concerns). Table 1 summarizes the data from

the standardized questionnaires. Significant gender

dierences were observed in the scores obtained in the

measures of coping, caregiver burden and QOL. Almost

half (49%) of the stroke survivors had experienced their

stroke between 6 to 12 months prior to the study.

e modied model t well with the data at χ

= 0.43,

df = 11, χ

/df = 0.03, p = 1.00, GFI = 0.99, AGFI = 0.98

(Fig 1). e contributions of the variables to the scores

on the WHOQOL-BREF, in terms of direct, indirect,

and total eects, are displayed in Table 2. Specically,

the model illustrates that the caregivers’ age and income

had a positive direct eect on their WHOQOL-BREF

QOL scores, and the caregivers’ age had a signicant

indirect eect on the scores through social support.

Caregiver burden had a negative direct eect on the

WHOQOL-BREF QOL scores and an indirect eect

through coping. Social support was more powerful than

coping as a mediator of WHOQOL-BREF QOL scores

among the stroke caregivers. ADL dependency scores

did not have a signicant direct or indirect eect on the

caregivers’ WHOQOL-BREF QOL scores (Table 2). is

best t model explained 64% of the variance in QOL,

10% of the variance in coping, and 4% of the variance

in social support.

TABLE 1. Overall and gender dierences in characteristics of stroke caregivers (n = 300).

Overall Male (n = 72) Female (n = 228) P value*

Variable Mean SD Mean SD Mean SD

Caregiver burden 44.65 17.59 39.81 1.84 46.18 1.12 0.007

Coping 92.60 13.74 87.84 1.17 94.10 0.87 0.001

Confrontive 42.36 7.99 28.61 0.64 32.10 0.36

Emotive 17.22 3.98 21.62 0.51 19.69 0.43

Palliative 33.01 5.72 39.54 0.87 40.37 0.43

Social support 32.45 12.21 31.79 1.28 32.66 0.83 0.597

Family 11.79 6.11 12.09 0.74 11.69 0.40

Relative 10.49 5.83 9.29 0.63 10.86 0.39

Health care provider 10.13 4.33 10.40 0.46 10.10 0.29

Overall QOL 75.88 14.44 79.81 1.69 74.64 0.94 0.008

Physical health 21.22 4.69 22.79 0.54 20.73 0.30

Psychological health 18.47 3.73 19.26 0.37 18.22 0.25

Social relationship 7.64 2.01 8.00 0.26 7.52 0.12

Environment 22.99 4.40 23.79 0.52 22.73 0.28

Note: * the results of the t-tests comparing male and female caregivers

Kongkar et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

293

Original Article

SMJ

TABLE 2. Direct eect, indirect eect and total eect of study variables.

Variable & impact The modied model

DE IE TE

BETA

Social support

➝

QOL .28*** - .28***

Coping

➝

QOL .10* - .10*

GAMMA

Age

➝

Social support -.18*** - -.18***

Age

➝

QOL -.15*** -.05 -.20***

Income

➝

Social support .10 - .10

Income

➝

QOL .14*** .03 .17***

ADL

➝

Coping .00 .00 .00

ADL

➝

QOL .05 - .05

Burden

➝

Coping .31*** - .31***

Burden

➝

QOL -.64*** .03 -.61***

Note: ***p < .001, *p < .05 Abbreviations: DE = Direct Eect, IE = Indirect Eect, TE = Total Eect

DISCUSSION

In this study, most of the ndings were congruent

with the Stress Process Model (age, income, care burden).

However, ADL dependency did not make a signicant

contribution to QOL scores. is is in contrast to some

previous research

19,20

but consistent with other studies.

5,21

Although more than 50% of caregivers reporting

that they provided most of the hygiene care, mobility

and emotional support, and transportation for follow-up

medical, this was not related to the caregivers perceived

QOL. is is consistent with a previous study that used

the same theoretical model and reported that loss of

functional capacity in the stroke survivor decreased

leisure activities for the caregiver, however this decrease

did not aect scores on the QOL measure.

Potential

contributors to this dierence include the duration of

Fig 1. Determinants of quality of life in ai family caregivers of stroke survivors.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

294

caregiving (all caregivers had been providing care for

more than 6 months and 50% for more than 12 months),

thus perhaps they had adapted to their caregiving role

while caregiving became more structured into routines.

In addition, aer this time period, there may also have

been improvements in ADLs that provided positive

reinforcement for caregivers.

In addition, most of the

participants (64%) had secondary caregivers to help

them in providing care for their family member.

In response to the questions from the BI, some

participants expressed that they were grateful to have

a chance to repay their parent/spouse for past support.

is sense of lial responsibility and pride in having

that responsibility is consistent with ai spiritual and

cultural beliefs and values and has been noted by other

ai researchers.

Indeed, older adults living in care

homes in ailand feel stigmatized, reecting a belief

that others must see them as being poor with no family.

In the original article regarding development of the BI,

Zarit and colleagues

cautioned that culture is important

in the concept of caregiver burden. In a recent literature

review, caregivers derived meaning in their caregiving

from multiple sources (their personal sacrice, a moral

or religious obligation, others’ expectations, social norms

and a subjective choice based on love, hope, and a sense

of reciprocity).

us, comparing scores obtained from

cross-cultural studies must be interpreted with caution,

considering interactions among individual, cultural and

societal contexts. For example, BI scores reported from

studies involving family caregivers of stroke survivors

ranged from 29.6 to 34.9 in two Brazilian studies

5,25

47.4 in a Turkish study

26,27

, 21.6 in another ai study

and 34.1 in a Japanese study.

In an Indian study 58%

of the caregivers had BI scores over 75.20. e mean

score (44.7) observed in this study was higher than some

other studies and may reect an increased prevalence

of physical health problems reported by caregivers and

perhaps a more accurate estimate of caregiver burden,

as suggested by other ai researchers.

29,30

An American

study calculated a cuto score on the BI of 24 - 26 as

being predicative of increased risk of depression in

caregivers.

However, in contrast to the extensive and

well-established long-term care system in North America,

most Asian countries rely heavily on family members

to provide care until the family member’s death.

As with other studies

33,34

, we found nancial security

had a signicant positive direct eect on the caregivers’

QOL. is consistent nding has led to eorts to quantify

the cost associated with informal care by families and

to consider providing employee benets or government

subsidies to support family caregiving, especially for

women who leave employment to care for parents who

need care.

35,36,37

In relation to an age issue, the nding in

this study revealed that about one third of the participants

aged 60 years old and older. Although some previous

study

found that the more advanced age, the more mature

and experienced people have in care providing skills.

However, consistent with previous stroke literature

3,4

this study revealed that advanced age had a negative

direct eect on QOL. Older people tend to have health

decline and therefore require support to cope with care

burden for other family members who are ill. Finally, the

importance of social support and adaptive coping was

armed in our study. As reviewed in previous studies,

multiple family support interventions have been developed

and tested, with positive impact on QOL measures.

37,39, 40

Limitations

is study may have limited generalizability due to

a potential bias selection of participants. Based on the

selection criteria, the participants were family caregivers

who were able to bring their stroke relatives to follow

up at the OPD. is means the ADL of their stroke

relatives might have been improved. However, there

may be a number of stroke survivors that have been

bedridden and difficult to bring to follow up at the

OPD. In these cases, the ADL dependency may make a

signicant contribution to the QOL scores. erefore,

further study should investigate more on those caring for

stroke survivors who are bedridden at home. Qualitative

studies that can yield rich text regarding factors aecting

QOL of stroke caregivers would also make an important

contribution to further understanding in this issue.

CONCLUSION

This study identified a number of potentially

modiable determinants (socioeconomic of caregiver,

perception of care burden, and social support) that could

be used to develop and test interventions to improve

the QOL among family caregivers of stroke survivors in

ailand. However, our data provide evidence that such

development would require consideration of gender.

Qualitative studies on the lived experience of ai male and

female caregivers of stroke survivors could provide rich

information about their biggest challenges, what resources

would be most helpful, and what priority area(s) would

improve their QOL. ese data could provide further

details to optimize potential the development and testing

gender-sensitive and/or gender-specic interventions.

The topics addressed in such an intervention could

focus on improving caregivers’ physical health status,

providing psychological support, and creating routines

Kongkar et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

295

Original Article

SMJ

that allow them for more predictability and scheduled

time for social activity and for self, leading to grater

QOL.

ACKNOWLEDGMENTS

is research is part of a doctoral dissertation at

Faculty of Graduate Studies, Mahidol University, ailand.

e principal investigator is grateful to the Commission

of Higher Education, Ministry of Education and Faculty

of Nursing, Mahidol University for providing funding

support for this doctoral study.

REFERENCES

1. Feigin VL, Mensah GA, Norrving B, Murray CJ, Roth GA.

Atlas of the global burden of stroke (1990-2013): e GBD

2013 Study. Neuroepidemiology 2015;45:230-6.

2. World Health Organization. Study protocol for the World

Health Organization project to develop a Quality of Life

assessment instrument (WHOQOL). Qual Life Res 1993;2:

153-9.

3. Tsai YH, Lou MF, Feng TH, Chu TL, Chen YJ, Liu HE. Mediating

eects of burden on quality of life for caregivers of rst-time

stroke patients discharged from the hospital within one year.

BMC Neurology 2018;18:50.

4. Morimoto T, Schreiner AS, Asano H. Caregiver burden and

health-related quality of life among Japanese stroke caregivers.

Age Ageing 2003;32:218-23.

5. Caro CC, Costa JD, Da Cruz DMC. Burden and quality of

life of family caregivers of stroke patients. Occuper Health

Care 2018;32:154-71.

6. Caro CC, Mendes PV, Costa JD, Nock LJ, Cruz DM. Independence

and cognition post-stroke and its relationship to burden and

quality of life of family caregivers. Top Stroke Rehabil 2017;24:

194-9.

7. Zarit SH, Reever KE, Bach-Peterson J. Relatives of the impaired

elderly: correlates of feelings of burden. Gerontologist 1980;20:

649-55.

8. Kniepmann K. Family caregiving for husbands with stroke:

an occupational perspective on leisure in the stress process.

OTJR (orofare N J) 2014;34:131-40.

9. Pearlin LI, Mullan JT, Semple SJ, Ska MM. Caregiving and

the stress process: an overview of concepts and their measures.

Gerontologist 1990;30:583-94.

10. Hair, JF, Black, WC, Babin, BJ, Anderson, RE, & Tatham, RL.

Multivariate data analysis (7th ed), New Jersy: Prentice Hall,

2010. p. 642-65.

11. Mahoney FI, Barthel DW. Functional evaluation: the barthel

index. Md State Med J 1965; 14:61-5.

12. Jitapunkul S, Kamolratanakul P, Ebrahim S. e meaning of

activities of daily living in a ai elderly population: development

of a new index. Age Ageing 1994;23(2):97-101.

13. Warunee M, Yupapin SN, Supreeda M, Pisamai O, Northouse

LL. Factors inuencing quality of life among family caregivers

of patients with advanced cancer: a causal model. Pacic Rim

Int J of Nurs Res 2013;17:304-16.

14. Schaefer K. Using the Jalowiec Coping Scale: concerns and

questions. J Cardiovasc Nurs 1988;2:9-10.

15. Suthanyakorn, N. Relationship between family coping behavior

and condence in dependent care of stroke patients. Unpublished

master’s thesis, Mahidol University, Bangkok, ailand; 1988.

p. 58-68.

16. Schaefer C, Coyne JC, Lazarus RS. e health-related functions

of social support. J Behav Med 1981;4:381-406.

17. Kaveevivitchai, J. Relationship between selected factors, uncertainty,

social support and adaptation of breast cancer patients receiving

chemotherapy. Unpublished master’s thesis, Mahidol University,

Bangkok, ailand; 1993.p.40-7.

18. Chiamwongpaet S, Vanichseni S, Sirinak C, Buavirat A.

WHOQOL-BREF-ai. Ministry of Public Health, ailand;

2004. p.1-6.

19. Liu LW, McDaniel SA. Family caregiving for immigrant seniors

living with heart disease and stroke: Chinese Canadian perspective.

Health Care Women Int 2015;36:1327-45.

20. Menon B, Salini P, Habeeba K, Conjeevaram J, Munisusmitha

K. Female caregivers and stroke severity determines caregiver

stress in stroke patients. Ann Indian Acad Neurol 2017;20:

418-24.

21. Takemasa S, Nakagoshi R, Murakami M, Uesugi M, Inoue Y,

Koeda H, et al. Factors aecting quality of life of family caregivers

of home-based care stroke patients using day care services.

Rigakuryoho Kagaku 2012;27:61-6.

22. Rachpukdee S, Howteerakul N, Suwannapong N, Tang-Aroonsin

S. Quality of life of stroke survivors: a 3-month follow-up

study. J Stroke Cerebrovasc Dis 2013;22:e70-8.

23. Tosangwarn S, Clisset P, Blake H. Predictors of depressive

symptoms in older adults living in care homes in ailand.

Arch Psychiatr Nurs 2018;32:51-56.

24. Limpawattana P, Intarasattakul N, Chindaprasirt J, Tiamkao

S. Perceived burden of ai caregivers for older adults aer

stroke. Clinical Gerontologist 2015;38:19-31.

25. Zhang J, Lee DTF. Feature Article: Meaning in stroke family

caregiving: A literature review. Geriatr Nurs 2017;38:48-56.

26. Pereira R, Santos E, Fhon J, Marques S, Rodrigues R. Burden

on caregivers of elderly victims of cerebrovascular accident.

Revista Da Escola De Enfermagem Da USP 2013; 47:185-92.

27. Tosun Z, Temel M. Burden of caregiving for stroke patients

and the role of social support among family members: an

assessment through home visits. Int J Caring Sci 2017;10:

1696-704.

28. Arai Y, Zarit S, Sugiura M, Washio M. Patterns of outcome

of caregiving for the impaired elderly: a longitudinal study in

rural Japan. Aging Ment Health 2002;6:39-46.

29. Limpanichkul Y, Magilvy K. Managing caregiving at home:

Thai caregivers living in the United States. J Cult Divers

2004;11:18-24.

30. Muangpaisan W, Praditsuwan R, Assanasen J, Srinonprasert

V, Assantachai P, Intalapaporn S, et al. Caregiver burden and

needs of dementia caregivers in ailand: a cross-sectional

study. J Med Assoc ai 2010;93:601-7.

31. Schreiner A, Morimoto T, Arai Y, Zarit S. Assessing family

caregiver’s mental health using a statistically derived cut-o

score for the Zarit Burden Interview. Aging Ment Health

2006;10:107-11.

32. Jullamate P, de Azeredo Z, Paul C, Subgranon R. ai stroke

patient caregivers: who they are and what they need. Cerebrovasc

Dis 2006;21:128-33.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

296

33. Santos NM, Tavares DM. Correlation between quality of life

and morbidity of the caregivers of elderly stroke patients.

Revista da Escola de Enfermagem da U S P 2012; 46:960-6.

34. Mant J, Carter J, Wade DT, Winner S. Family support for

stroke: a randomized controlled trial. Lancet 2000;356:808-13.

35. Bouldin ED, Shaull L, Andresen EM, Edwards VJ, McGuire

LC. Financial and health barriers and caregiving-related diculties

among rural and urban Caregivers. J Rural Health 2018;34:

263-74.

36. Zhu W, Jiang Y. A meta-analytic study of predictors for informal

caregiver burden in patients with stroke. J Stroke Cerebrovasc

Dis 2018;27:3636-46.

37. Aun NSM, Mohd RH. Informal caregiving: empowering social

support programs by employers. Akedemika 2016;86:3-9.

38. Parisi JM, Rebok GW, Carlson MC, Fried LP, Seeman TE, Tan

EJ, et al. can the wisdom of aging be activated and make a

dierence societally? Educ Geron 2009;35:867-79.

39. Cameron JI, Naglie G, Green TL, Gignac MA, Bayley M,

Huijbregts M, et al. A feasibility and pilot randomized controlled

trial of the “Timing it Right Stroke Family Support Program”.

Clin Rehabil 2015;29:1129-40.

40. Cheng HY, Chair SY, Chau JP-C. Review: e eectiveness of

psychosocial interventions for stroke family caregivers and

stroke survivors: A systematic review and meta-analysis. Patient

Educ Couns 2014;95:30-44.

Kongkar et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

297

Original Article

SMJ

Narumol Silpa-archa, M.D., Pichanee Chaweekulrat, M.D., Natchaya Junsuwan, M.D., Chanisada Wongpraparut, M.D.

Department of Dermatology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.

Clinical Characteristics and Prognosis of Morphea

(Localized Scleroderma) in Adults:

a Retrospective Study

Corresponding author: Chanisada Wongpraparut

E-mail: chanisada@hotmail.com

Received 1 December 2017 Revised 22 February 2018 Accepted 17 May 2019

ORCID ID: http://orcid.org/0000-0002-9014-3229

http://dx.doi.org/10.33192/Smj.2019.45

ABSTRACT

Objective: To study the clinical characteristics and prognosis of morphea in ai adult patients.

Methods: e medical records of 81 morphea adult patients who visited Siriraj Hospital, ailand, between 2006 and

2015, were retrospectively reviewed. e demographics, clinical features, treatments and outcomes were analyzed.

Clinical improvement was categorized as excellent (>80%), partial (1%–80%), and no response.

Results: Circumscribed morphea was the most common subtype (34, 42%) of the 81 patients, followed by linear

morphea (21, 26%), generalized morphea (14, 17%), en coup de sabre (11, 14%) and Parry–Romberg syndrome

(1, 1%). Systemic treatment and ultraviolet A (UVA)1 phototherapy had satisfactory outcomes in 83% and 79%

of cases, respectively, providing a partial to excellent response. For overall remission, one year aer the treatment,

30% of patients achieved a partial to excellent response. Aer 2 years, this proportion rose to 50%. e median time

to clinical response was 24 months.

Conclusion: Morphea is a dicult-to-treat dermatosis, with the majority of the patients having a partial clinical

response and a high recurrent rate. Combination of treatment might be a worthy option.

Keywords: Morphea; prognosis; phototherapy; systemic; treatment (Siriraj Med J 2019; 71: 297-301)

INTRODUCTION

Morphea, or localized scleroderma, is a rare,

chronic, autoimmune disease manifested by dermis

and/or subcutaneous tissue sclerosis.

e incidence of

morphea is approximately 2.7 cases/100,000 people.

Morphea is divided into ve subtypes: (1) circumscribed

morphea (including a supercial and deep variant);

(2) linear morphea (including a limb/trunk variant

and a head variant); (3) generalized morphea; (4) the

pansclerotic subtype; and (5) the mixed subtype.

e

dierent clinical manifestations of morphea have led to the

development of dierent classications and progressions

of the disease.

1,3

erefore, this study aimed to elucidate

the clinical characteristics, treatment types, results aer

treatment, and prognosis of adult patients with morphea.

MATERIALS AND METHODS

Subjects

is study was approved by the Institutional Review

Board, Faculty of Medicine Siriraj Hospital (Si 690/2015).

e medical records of patients age ≥ 18 years diagnosed

with morphea at the Department of Dermatology, Faculty

of Medicine Siriraj Hospital, Mahidol University, between

January 2006 and December 2015 were retrospectively

reviewed. A total of 81 adult morphea patients, evidenced

by clinical and/or pathological ndings, were included.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

298

Demographic data, clinical manifestations, lesion sites,

symptoms, treatments and clinical responses at nal follow-

up were collected. A telephone survey was conducted to

follow up the relapse status aer the last consultation.

e clinical responses were subjectively categorized

as excellent (>80%), partial (1%-80%), and no response.

Serial photographs before and after treatment were

evaluated for clinical response by two dermatologists.

Additionally, the eects of the UVA1 dosage on the

clinical responses and recurrence rates were gathered.

UVA1 phototherapy was categorized as low-dose (20-

40 J/cm

), medium-dose (>40-80 J/cm

) and high-dose

(>80-120 J/cm

). UVA1 phototherapy was administered

three to ve times a week for 30 sessions.

Statistical analysis

Descriptive statistics, including frequency counts

and tables, were used for demographic data, clinical

manifestations, clinical responses, treatments, and

disease recurrence. A Chi-square test or, if there were

<5 responses in a cell, a Fisher exact test was used to

examine the differences in the proportions and the

association strength between the clinical response and

disease recurrence, according to the treatment types and

UVA1-phototherapy doses. A p value of less than .05 was

set as the cuto for statistical signicance. All analyses

were performed using IBM SPSS Statistics for Windows,

version 22.0 (IBM Corp., Armonk, N.Y., USA).

RESULTS

A total of 81 patients, aged 18-83 years and with a

median age of 34 years, were included in the analysis.

ere were 64 (79%) females, and 17 (21%) males, making

the female to male ratio 3.8:1. e median time between

the initial disease manifestation and diagnosis was one

month. e most common subtype of morphea was

circumscribed morphea (34, 42%), followed by linear

morphea (21, 26%), generalized morphea (14, 17%), en

coup de sabre (11, 14%), and Parry-Romberg syndrome

(1, 1%).

Morphea lesion was mostly found in the lower

extremities, followed by the trunk, head and neck, and

upper extremities. In circumscribed morphea, lesions

were mostly found in the trunk. However, in the linear

and generalized groups, lesions were mostly found in the

lower extremities. Skin discoloration (79%), induration

(43%) and atrophy (39%) were the top three, most common

clinical symptoms of all of the types of morphea. None

of the patients progressed to systemic sclerosis during

the follow-up period.

e combination therapy was the most oen used

treatment for generalized morphea and linear morphea

(Table 1). Topical corticosteroids (41%) were the most

frequently prescribed topical treatment, followed by

combined topical corticosteroids with a calcineurin

inhibitor (24%), and then by a calcineurin inhibitor

alone (11%). In the case of systemic treatment, colchicine

(23%), corticosteroid (19%), and methotrexate (14%) were

the top three prescribed medicines. Colchicine was the

most commonly prescribed drug for systemic treatment

in the circumscribed and en coup de sabre groups. As

for linear and generalized morphea, oral prednisolone

(23.8%) and methotrexate (35.7%) were the most used

drugs, respectively.

Fiy-nine percent of patients achieved a partial

clinical response, another 18% had no response, while

9% had an excellent response. Generalized morphea

(Fig 1) displayed the most favorable response rate at

79%, followed by circumscribed morphea (70%), linear

morphea (67%), and en coup de sabre (46%). e linear

group had the shortest recurrence time, at 8 months,

followed by en coup de sabre (10 months) and generalized

(31 months). Circumscribed morphea had the longest

recurrence time, at 47 months.

UVA1 phototherapy was administered in 24 (30%)

cases. Eighteen (75%) patients received a medium dose

(>40-80 J/cm

) regimen of UVA1 (Table 1). For all

types of morphea, the clinical responses did not dier

signicantly with dierent doses of UVA1 (p=0.490;

Table 2). UVA1 phototherapy combined with systemic

treatment (oral prednisolone, methotrexate, colchicine

or chloroquine) provided a signicantly better response

than UVA1 alone (p= 0.044). e recurrence rate did

not vary signicantly (p=0.674) for morphea patients

treated with just UVA1 or with UVA1 plus systemic

treatment. Dierent doses of UVA1 also had no eect

on the recurrence rate (p=0.850) (Table 2).

e Kaplan–Meier curve demonstrated the probability

of a partial to excellent clinical response for the morphea

patients overall. One year aer the treatment, 30% of

patients achieved a partial to excellent response. Aer

2 years, this proportion rose to 50%. e median time

to clinical response was 24 months.

DISCUSSION

is retrospective study emphasizes the clinical

features and prognoses of dierent types of morphea. We

found circumscribed morphea to be the most common

type, which was similar to results from the Korea and

Netherlands.

4,5

However, a study from the United States

found that generalized morphea was the most common

subtype among adults.

Silpa-archa et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

299

Original Article

SMJ

TABLE 1. Number and type of treatments by morphea subtypes.

Type of morphea, n (%)

Total, n (%)

Localized Linear Generalized

En coup de sabre

(n=80)

(n=34) (n=21) (n=14) (n=11)

Demographics

Mean age (range) 45 (18-83) 29 (18-52) 35 (21-65) 31 (18-49) 37 (18-83)

Male 6 (18) 5 (29) 1 (7) 5 (46) 17 (21)

Female 28 (82) 16 (76) 13 (93) 6 (55) 64 (79)

Mean age onset (range) 37 (4-81) 19 (2-52) 27 (9-58) 22 (8-37) 28 (18-83)

Mean duration of disease (months) (range)

98 (12-240) 124 (2-276) 93 (6-240) 118 (12-216) 107 (2-81)

Location

Head and neck 9 (24) 2 (7) 2 (6) 11 (100) 25 (23)

Trunk 13 (35) 3 (11) 9 (27) 0 25 (23)

Buttock 1 (3) 2 (7) 0 0 3 (3)

Upper extremities 6 (16) 9 (34) 10 (31) 0 25 (23)

Lower extremities 8 (22) 11 (41) 11 (33) 0 30 (27)

Intertriginous area 0 0 1 (3) 0 1 (1)

Symptoms

Tightness 1 (1) 6 (13) 0 0 7 (4)

Pain 1 (1) 2 (4) 0 0 3 (2)

Skin color change 29 (43) 16 (33) 12 (43) 6 (32) 63 (38)

Erythema/edema 5 (8) 6 (13) 5 (18) 1 (5) 17 (10)

Deformity 1 (1) 6 (13) 0 2 (11) 9 (6)

Induration 17 (25) 10 (20) 6 (21) 1 (5) 34 (21)

Atrophy 14 (21) 2 (4) 5 (18) 9 (47) 31 (19)

Type of treatment

Topical treatment

Corticosteroids 13 (38) 10 (48) 8 (57) 2 (18) 33 (41)

Calcineurin inhibitor 5 (15) 3 (14) – 1 (9) 9 (11)

Corticosteroids plus calcineurin inhibitor

10 (29) 4 (19) 2 (14) 3 (27) 19 (24)

Others

†

3 (9) 2 (6) – 2 (18) 7 (9)

Systemic treatment

Oral prednisolone 6 (18) 5 (24) 4 (29) 1 (9) 15 (19)

Methotrexate 3 (9) 3 (14) 5 (36) – 11 (14)

Colchicine 9 (27) 3 (14) 4 (29) 2 (18) 18 (23)

Others

†

11 (32) 9 (43) 8 (58) 4 (36) 32 (40)

Phototherapy (UVA1)

‡

Low dose 1 (3) 1 (5) – – 2 (2)

Medium dose 4 (12) 5 (24) 7 (50) 2 (18) 18 (28)

High dose – 2 (10) 1 (7) 1 (9) 4 (5)

Other treatments

Intralesional corticosteroids 1 (3) – – 1 (9) 2 (3)

Surgery for reconstruction – – – 1 (9) 1 (1)

Clinical response

Excellent (>80%) 2 (6) 1 (5) 4 (29) – 7 (9)

Partial (≤80%) 22 (64) 13 (62) 7 (50) 5 (46) 47 (59)

No response 5 (15) 4 (19) 1 (7) 4 (36) 14 (18)

Loss to follow-up 5 (15) 3 (14) 2 (14) 2 (18) 12 (15)

Recurrence of disease

Recurrence (months) 7 (29) 6 (43) 3 (27) 3 (60) 19 (24)

Time to recurrence (months) – 47 (4, 93) 8 (2, 26) 31 (19, 32) 10 (4, 15) 17 (2, 93)

median (min, max)

† Other systemic treatments included chloroquine, hydroxychloroquine, isotretinoin, pentoxifylline, d-penicillamine and indomethacin

‡ UVA1 phototherapy was categorized as low dose (20–40 J/cm

), medium dose (>40–80 J/cm

) and high dose (>80–120 J/cm

)

Abbreviation: UVA1= ultraviolet A1

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

300

TABLE 2. Clinical response and disease recurrence from UVA1 phototherapy.

Clinical response, n (%) Disease

Excellent Partial No

P-value recurrence P-value

(>80%) (≤80%) response

n (%)

Treatment

UVA1 plus systemic treatment 3 (33) 6 (67) – 0.044* 5 (56) 0.674

UVA1 1 (7) 11 (73) 3 (20) 6 (50)

Dose of UVA1

†

Low dose – 2 (100) – 0.490 1 (50) 0.850

Medium dose 3 (17) 12 (66) 3 (17) 8 (53)

High dose 1 (25) 3 (75) – 2 (50)

† UVA1 phototherapy was categorized as low dose (20–40 J/cm

), medium dose (>40–80 J/cm

) and high dose (>80–120 J/cm

)

* p-value ≤ 0.05 considered statistically signicant

Abbreviation: UVA1 = ultraviolet A1

Fig 1. Excellent response of generalized morphea on the breast. A) Baseline, indurated plaque on the breast. B) Aer treatment with oral

prednisolone (30 mg/day) for 6 months, combined with methotrexate 7.5-10 mg/weeks for 12 months and medium dose UVA1 for 30

sessions. e morphea lesion le with postinammatory hyperpigmentation.

Silpa-archa et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

301

Original Article

SMJ

In the present study, clinical responses, including

partial to excellent responses, were detected in 59% of

the treated morphea cases. e overall level of clinical

improvement was similar to the aforementioned Korean

study, which showed a 63% improvement.

Based on

the Kaplan–Meier curve, approximately 50% of treated

morphea cases obtained partial to excellent clinical

responses within 2 years, regardless of the treatment

employed. By comparison, Mertens et al., evaluated the

clinical responses aer treatment with methotrexate

alone; and 62% of patients show a partial to excellent

response aer 2 years.

UVA1 phototherapy is a commonly-used treatment

for morphea.

Medium- and high-dose UVA1 phototherapy

have proven to provide better outcomes for morphea

than low-dose UVA1.

Our study showed that 80% of

patients had a partial to excellent response to UVA1

(medium dose), similar to the ecacy results found in

studies by Vasquez et al. (60%), Su et al. (82.9%) and

Andres et al. (82%).

10-12

Recurrence is common aer morphea treatment.

Mertens et al., reported a disease recurrence rate of 17%

for adult-onset localized morphea, and 27% for pediatric-

onset morphea.

Linear morphea was the most frequent

type.

Methotrexate combined with systemic corticosteroids

showed a recurrence rate of 30%, with a 6-month mean

relapse time aer methotrexate discontinuation.

Moreover,

Vasquez et al., reported a recurrence of 46% for active

morphea aer successful UVA1 phototherapy.

UVA1

phototherapy combined with immunosuppressives may

inhibit episodes of recurrence.

Similar to another study,

the present study found a recurrence rate of 50% aer

successful treatment with UVA1. Our study supports the

view that UVA1 combined with a systemic treatment

for morphea provides signicantly more benets than

UVA1 alone.

is study is a retrospective chart review and there

was a lack of standardized criteria to evaluate the treatment

responses; so the evaluation of the eectiveness of the

treatments could therefore be subjective. Lastly, there

may be a recall bias from the telephone interviews. We

recommend a further study should be performed using

validated assessments of morphea activity to reduce any bias

in collecting data, such as the ultrasound measurements

of skin thickness.

CONCLUSION

Morphea is a dicult-to-treat dermatosis, with

most patients having a partial clinical response and a

high recurrence rate. Overall, the median time to achieve

a partial to excellent clinical improvement was 2 years.

UVA1 phototherapy combined with systemic treatment

was found to be a worthy option for patients.

ACKNOWLEDGMENTS

We acknowledge Mr. Suthipol Udompanthurak for

his statistical advice. is research received no specic

grant from any funding agency in the public, commercial,

or not-for-prot sectors.

Conicts of interest: None

REFERENCES

1. Mertens JS, Seyger MMB, Thurlings RM, Radstake T, de

Jong E. Morphea and Eosinophilic Fasciitis: An Update. Am

J Clin Dermatol 2017;18:491-512.

2. Peterson LS, Nelson AM, Su WP, Mason T, O’Fallon WM,

Gabriel SE. e epidemiology of morphea (localized scleroderma)

in Olmsted County 1960-1993. J Rheumatol 1997;24:73-80.

3. Sartori–Valinotti JC, Tollefson MM, Reed AM. Updates

on morphea: role of vascular injury and advances in treatment.

Autoimmune Dis 2013;2013:467808.

4. Noh JW, Kim J, Kim JW. Localized scleroderma: a clinical

study at a single center in Korea. Int J Rheum Dis 2013;16:437-

41.

5. Mertens JS,Seyger MM,Kievit W,Hoppenreijs EP,Jansen TL,van

de Kerkhof PC,et al. Disease recurrence in localized scleroderma:

a retrospective analysis of 344 patients with paediatric- or

adult-onset disease. Br J Dermatol 2015;172:722-28.

6. Johnson W, Jacobe H. Morphea in adults and children cohort

II: patients with morphea experience delay in diagnosis and

large variation in treatment. J Am Acad Dermatol 2012;67:881-

89.

7. Mertens JS,van den Reek JM,Kievit W,van de Kerkhof

PC,urlings RM,Radstake TR,et al. Drug Survival and Predictors

of Drug Survival for Methotrexate Treatment in a Retrospective

Cohort of Adult Patients with Localized Scleroderma. Acta

Derm Venereol 2016;96:943-47.

8. Petersen MJ, Hansen C, Craig S. Ultraviolet A irradiation

stimulates collagenase production in cultured human broblasts.

J Invest Dermatol 1992;99:440-44.

9. Sator PG, Radakovic S, Schulmeister K, Honigsmann H,

Tanew A. Medium-dose is more eective than low-dose ultraviolet

A1 phototherapy for localized scleroderma as shown by 20-

MHz ultrasound assessment. J Am Acad Dermatol 2009;60:

786-91.

10. Vasquez R, Jabbar A, Khan F, Buethe D, Ahn C, Jacobe H.

Recurrence of morphea aer successful ultraviolet A1 phototherapy:

A cohort study. J Am Acad Dermatol 2014;70:481-88.

11. Su O,Onsun N,Onay HK,Erdemoglu Y,Ozkaya DB,Cebeci

F,et al. Eectiveness of medium-dose ultraviolet A1 phototherapy

in localized scleroderma. Int J Dermatol 2011;50:1006-13.

12. Andres C, Kollmar A, Mempel M, Hein R, Ring J, Eberlein B.

Successful ultraviolet A1 phototherapy in the treatment of

localized scleroderma: a retrospective and prospective study.

Br J Dermatol 2010;162:445-47.

13. Cox D, G OR, Collins S, Byrne A, Irvine A, Watson R. Juvenile

localised scleroderma: a retrospective review of response to

systemic treatment. Ir J Med Sci 2008;177:343-46.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

302

Suksri Chotikavanich, M.D., Panida Kosrirukvongs, M.D., Jatuporn Duangpatra, M.D., Pinnita Prabhasawat,

M.D., Wipawee Booranapong, M.D., Chareenun Chirapapaisan, M.D., Sabong Srivannaboon, M.D.

Department of Ophthalmology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.

A Two-Year Outcome of Intrastromal Corneal Ring

Segment Implantation in Keratoconus:

Initial Report in Thai Patients

Corresponding author: Suksri Chotikavanich

E-mail: suksri_c@yahoo.com

Received 7 February 2018 Revised 20 March 2018 Accepted 22 February 2019

ORCID ID: http://orcid.org/0000-0002-5973-643X

http://dx.doi.org/10.33192/Smj.2019.46

ABSTRACT

Objective: To determine two-year outcome of intrastromal corneal ring segment (ICRS) implantation in keratoconus

in ai patients

Methods: A retrospective review of medical records of the patients underwent Ferrara-type ICRS implantation

(single and two segments) at Siriraj Hospital between November 2013 and December 2017 was conducted. Clinical

outcomes were assessed at 1 month, 6 months, 1 year, and 2 years postoperatively.

Results: Of 9 eyes in 8 patients, the mean age of the patients was 24.6 ± 7.5 years. e mean follow-up time was

32.2 ± 9.4 months. Overall, the median visual acuity was signicantly improved postoperatively (p value = 0.007).

At 2 years, the uncorrected visual acuity (UCVA) improved from 1.00 logMAR to 0.56 logMAR, and the corrected

distance visual acuity (CDVA) improved from 0.76 logMAR to 0.10 logMAR. Correspondingly, the median spherical

equivalent refraction was signicantly improved postoperatively from -7.38 D to -3.13 D (p < 0.001). Moreover, the

median anterior corneal topographic data signicantly changed between visits (p < 0.02). e Kmax decreased from

52.65 D to 46.65 D and the Kmean decreased from 48.10 D to 45.40 D at 2 years. Postoperative adverse eects were

glare and halos (3 eyes), visually insignicant small white corneal deposits around the segments (2 eyes), extrusion

of a ring segment needed removal with reversible to baseline vision (1 eye).

Conclusion: is initial report in ai patients showed that the ICRS implantation in keratoconus could improve

the visual, refractive, and topographic parameters with stability at 2 years. However, appropriate case selection and

surgical technique should be considered.

Keywords: Intrastromal corneal ring segment; keratoconus; ailand (Siriraj Med J 2019; 71: 302-309)

INTRODUCTION

Keratoconus is a progressive ectatic disease resulting in

severe visual loss. ere is a wide spectrum of treatments

for keratoconus including non-surgical options of glasses

and contact lenses and invasive surgery of keratoplasty.

For the reason to improve spectacle corrected vision

or contact lens tolerance and postpone keratoplasty,

minimally invasive surgery of keratoconus by intrastromal

ring segment (ICRS) implantation was introduced by

Colin et al in 2000

and has gained wide acceptance

in the previous decade. e explanation to correct the

myopia was that adding tissue to the periphery of the

cornea with small arc-like implants resulted in attening

the central cornea.

2,3

Chotikavanich et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

303

Original Article

SMJ

Previous studies evaluated the eectiveness of the

surgery in their populations.

3-7

To our knowledge, and

based on our review of the English-language literature

(Ovid, PubMed, ProQuest, Google Scholar, ScienceDirect,

and Scopus databases), no previous study has reported on

the outcome of the ICRS implantation in keratoconus in

ai patients. We conducted a retrospective record study

to demonstrate the clinical outcome and the stability of

outcome of the surgery at 2-year follow up.

MATERIALS AND METHODS

Patients

e protocol for this study was approved by the

Committee for the Protection of Human Participants

in Research at the Faculty of Medicine Siriraj Hospital,

Mahidol University, Bangkok, ailand (Si 787/2558).

e study was a longitudinal retrospective analysis of

consecutive patients with keratoconus in which ICSR

implantation was performed at Siriraj Hospital from

November 2013 to December 2017. e inclusion criteria

for ICRS implantation in this study were progressive

intolerance to contact lenses and a clear cornea with

minimum thinnest corneal thickness of over 300 µm.

e exclusion criteria were a signicant apical corneal

opacity and scarring, corneal hydrops, endothelial count

less than 2000 cells/mm

, history of glaucoma, patients

with intense atopy, autoimmune disease or systemic

connective tissue disease. All surgeries were performed

by four surgeons (P.K., P.P., S.C., and C.C.).

Preoperative examination

Before the surgery, all patients had a complete

ophthalmologic examination including uncorrected

(UCVA) and best-corrected (CDVA) distance visual acuity,

manifest refraction, slit-lamp biomicroscopy, fundoscopy,

and Pentacam corneal topography analysis (Oculus

Optikgerate GmbH, Wetzlar, Germany). e following

topographic data were evaluated: the thinnest corneal

thickness, anterior keratometry, and corneal asphericity

(Q value) at 8 mm. e grading of the keratoconus was

classied according to Amsler-Krumeich classication

based on preoperative anterior keratometry; grade I:

Kmean ≤ 48 D; grade II: Kmean > 48D to 53 D; grade

III: Kmean: > 53 D to 55D; and grade IV: Kmean > 55D.

Surgical technique

e polymethyl methacrylate Ferrara-type ICRS

(AJL Ophthalmic, S.A., Spain), which has a triangular

cross section (at basis width = 0.6 mm) that induces a

prismatic eect on the cornea, was used in all eyes. e

segment has variable thicknesses (150 to 300

μm) and

arc lengths (90°, 120°, 140°, 160°, and 210°). e protocol

used for ICRS selection was based on the manufacturer-

dened nomogram.

e surgery was performed under local anesthesia.

The ICRS tunnel was created manually. The center

of the visual axis on the cornea was marked for a

reference point for centration by asking the patient

to xate on the corneal light reex of the microscope

light. en, a 5 mm., optical zone was marked and a small

radial incision was made at the most curved meridian by

a diamond knife for the ICRS to be implanted. e depth

of this perpendicular incision was 80% of the corneal

thickness. Preparation of the intrastromal pocket for the

ICRS was performed at the side of the incision from its

base by the spreader and the 270-degree semicircular

(clockwise and counterclockwise) dissectors maintaining

a uniform depth. Aer channel creation, the segment

was inserted using the modied McPherson forceps with

the at side of the ring downward.

Postoperatively, levooxacin eye drops were prescribed

four times daily for 2 weeks and topical steroid eye drops

four times daily for 2 weeks with tapering of the dose

over the following 1 month. Non-preservative articial

tear was also prescribed.

Postoperative assessment

Postoperative follow-up visits were at 1 day; 1, 3, 6

months, and then every year following ICRS implantation.

We assessed only the clinical outcomes at 1 month, 6

months, 1 year, and 2 years postoperatively to evaluate

eectiveness and stability of the surgery. At each of the

follow-up visits, the record of UDVA, CDVA, manifest

refraction, slit-lamp biomicroscopy, and corneal topography

were reviewed. Visual acuity was measured using the

Snellen rating and transformed into the logarithm of

the minimum angle of resolution (LogMAR)

Statistical analysis

Data analysis was performed using PASW Statistics

(SPSS) 18.0 (SPSS Inc., Chicago, IL, USA). A non-parametric

Friedman test was used to compare preoperative and

postoperative data at all visits. If the dierence was

statistically signicant, a post-hoc paired comparison

was implemented using the nonparametric two related

sample comparison method with adjusting Bonferroni-

corrected alpha level. A p value of < 0.05 was considered

signicant throughout.

RESULTS

Clinical data

Nine eyes in 8 patients were included in the study,

and 5 patients were male and 3 patients were female.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

304

One patient received ICRS implantation in both eyes.

Four eyes were right eye. e mean age of the patients was

24.6 ± 7.5 years. Considering age group of the patients,

6 eyes were ≤ 30 years and 3 eyes were > 30 years. About

grading of keratoconus; 4 eyes were grade I, 2 eyes were

grade II, 2 eyes were grade III, and 1 eye was grade IV.

e mean thinnest corneal thickness was 442 ± 45.5 µm.

Eight eyes had two segments of ICRS implantation

(segment thicknesses of 150 μm and 200 μm, arc lengths

of 140°and 160°), whereas 1 eye had single segment

(segment thicknesses of 200 μm, arc lengths of 210°).

The mean follow-up time of all patients was 32.2 ±

9.4 months. Representative slit-lamp pictures aer the

surgery were shown in Fig 1. Postoperative symptom of

tolerable glare and halos occurred in 3 eyes. Concerning

information about scotopic pupil size, centering or depth

of the ICRS was not enough to analyze the possible

associated factors of those glare and halos. Small white

intrastromal deposits which accumulated in the lamellar

channel around the segments were found in 2 eyes. ere

was an extrusion of a ring segment at the incision site in

1 eye at a few weeks aer 1 month postoperatively, and

both segments in the eye were removed because of the

awareness of surgically induced astigmatism. However,

4 months later, her visual acuity and refraction returned

back to the baseline before surgery. As the result, eight

eyes remained for the data analysis at 6 months, 1 year,

and 2 years.

Visual acuity

Visual acuity before and aer ICRS implantation

at 1 month, 3 months, 6 months, 1 year, and 2 years

were presented in Table 1 and Fig 2. Overall, the median

UCVA signicantly improved aer surgery (p = 0.007).

Aer post-hoc paired comparison, the median UCVA

still signicantly improved at 2 years postoperatively

(p < 0.02). Moreover, the median CDVA signicantly

improved aer surgery (p = 0.001). Aer a post-hoc paired

comparison, the median CDVA signicantly improved

aer 1 year postoperatively (p < 0.02). At the last follow

up, the median CDVA improved from 0.76 logMAR to

0.10 logMAR.

Manifest Refraction

Manifest refraction before and aer ICRS implantation

at 1 month, 3 months, 6 months, 1 year, and 2 years

were also presented in Table 1 and Fig 2. Overall, the

median spherical equivalent of high myopia signicantly

decreased aer surgery (p < 0.001). Aer post-hoc paired

comparison, the median spherical equivalent signicantly

decreased at 6 months postoperatively and was still

signicantly decreased at 2 years postoperatively (p <

0.02). Postoperatively, the median spherical equivalent

improved from -7.38 D to -3.13 D. e median refractive

astigmatism also decreased aer the surgery from -8.75

D to -2.25 D, but did not reach statistical signicance.

Corneal topography

Anterior corneal topographic data before and aer

ICRS implantation at 1 month, 3 months, 6 months, 1

year and 2 years were presented in Table 2 and Fig 2.

Representative anterior corneal topographic ndings

before and aer surgery were shown in Fig 3. Overall,

the median Kmax, Kmean, and keratometric astigmatism

signicantly decreased postoperatively (p < 0.02). At

the last visit, Kmax decreased from 52.65 D to 46.65

D, Kmean decreased from 48.10 D to 45.40 D, and

keratometric astigmatism decreased from 6.70 D to

1.40 D. Aer post-hoc paired comparison, Kmax and

keratometric astigmatism significantly decreased at

6 months postoperatively and remained signicantly

decreased at 2 years postoperatively (p < 0.05). Aer

post-hoc paired comparison, Kmean also signicantly

decreased at 2 years postoperatively (p < 0.03). e median

Q value of the anterior cornea changed signicantly

toward prolateness aer surgery (p = 0.01). At the last

visit, Q value changed from -0.83 to -0.04. Aer post-

hoc paired comparison, Q value changed signicantly

toward prolateness aer 1 month postoperatively (p =

0.01).

DISCUSSION

To investigate the outcome of ICRS implantation

in keratoconus in ai patients, a two-year retrospective

review in 9 eyes was conducted. e result showed that

the median visual acuity signicantly improved aer

surgery (p ≤ 0.007). At the last follow up, the median

UCVA improved form 1.00 logMAR to 0.56 logMAR,

and the CDVA improved from 0.76 logMAR to 0.10

logMAR, which represented approximately 0.40 logMAR

change or 4 lines gain of the vison and considered to

be clinically signicant. Among publications with the

same Ferrara type ICRS and similar follow-up time of

6 months to 2 years, the improvement of CDVA in this

study was comparable with others which reported the

mean improvement ranged from 0.14 logMAR to 0.40

logMAR.

3-7

Consistent with the visual outcome, the refractive

outcome was signicantly improved postoperatively in

terms of the spherical equivalent (p < 0.001). At the last

follow up, the median spherical equivalent improved from

-7.38 D to -3.13 D. e change of 4.25 D in this study

Chotikavanich et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

305

Original Article

SMJ

TABLE 1. Visual acuity and manifest refraction before and aer intrastromal corneal ring implantation.

Examination

Pre-operative Post-operative

Parameter 1 month 6 months 1 year 2 years P value

UCVA (logMAR)

Median 1.00 0.50 0.50 0.70 0.56 0.007

*,d

Interquartile range 0.60, 1.30 0.40, 1.00 0.30, 0.90 0.20, 0.90 0.20, 0.90

CDVA (logMAR)

Median 0.76 0.30 0.20 0.10 0.10 0.001

*,c,d

Interquartile range 0.30, 0.80 0.10, 0.70 0.10, 0.70 0.00, 0.56 0.00, 0.48

Spherical equivalent

Median -7.38 -6.38 -3.13 -5.13 -3.13 <0.001

*,b,c,d

Interquartile range -14.13,-4.88 -15.63,-1..75 -10.00, 1.00 -9.13, 0.75 -7.63, 0.50

Refractive astigmatism

Median -8.75 -3.50 -2.75 -1.50 -2.25 0.052

Interquartile range -11.00, -7.50 -7.75, -2.00 -4.00, -2.25 -3.50, -1.25 -4.75, -1.25

*statistically signicant by non-parametric Friedman test,

statistically signicant with a post-hoc paired comparison using Dunn-Bonferroni

test between preoperative examination and postoperative examination at 6 months (p value = 0.013),

statistically signicant with a post-hoc

paired comparison using Dunn-Bonferroni test between preoperative examination and postoperative examination at 1 year (p value ≤0.01),

statistically signicant with a post-hoc paired comparison using Dunn-Bonferroni test between preoperative examination and postoperative

examination at 2 years (p value <0.02)

TABLE 2. Anterior corneal topographic data before and aer intrastromal corneal ring implantation.

Examination

Pre-operative Post-operative

Parameter 1 month 6 months 1 year 2 years P value

Kmax

Median 52.65 47.20 46.80 46.50 46.65 0.005

*,b,d

Interquartile range 50.93, 59.20 44.20, 53.53 42.45, 52.60 44.28, 52.45 44.05, 52.20

Kmean

Median 48.10 45.45 45.75 45.55 45.40 0.014

*,d

Interquartile range 47.78, 56.03 42.10, 51.55 43.50, 50.60 43.90, 50.40 43.83, 50.55

Kastigatism

Median 6.70 2.35 2.20 1.45 1.40 0.005

*,b,c,d

Interquartile range 5.25, 8.30 1.58, 7.20 0.48, 3.75 0.73, 2.83 0.55, 3.23

Q value

Median -0.83 0.41 0.02 -0.03 -0.04 0.01

*,a,b

Interquartile range -1.56, -0.46 -0.96, 0.71 -0.80, 0.55 -0.80, 0.49 -0.81, 0.51

* statistically signicant by non-parametric Friedman test,

statistically signicant with a post-hoc paired comparison using Dunn-Bonferroni

test between preoperative examination and postoperative examination at 1 month (p value=0.01),

statistically signicant with a post-hoc

paired comparison using Dunn-Bonferroni test between preoperative examination and postoperative examination at 6 months (p value<0.05),

statistically signicant with a post-hoc paired comparison using Dunn-Bonferroni test between preoperative examination and postoperative

examination at 1 year (p value<0.05),

statistically signicant with a post-hoc paired comparison using Dunn-Bonferroni test between

preoperative examination and postoperative examination at 2 years (p value<0.03)

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

306

Fig 1. Representative pictures of postoperative intrastromal corneal ring segment implantation. (A) Successful two-segment typed surgery.

(B) Small white deposits around the rings (arrows). (C) A complicated case with one segment extrusion (arrow).

Fig 2. Comparison of (A) visual acuity, (B) refraction, and (C) keratometry between visits.

Chotikavanich et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

307

Original Article

SMJ

Fig 3. Representative topographic ndings (A) before and (B) 2 years aer intrastromal corneal ring implantation. e curvature maps

showed attening eect on the central cornea aer the surgery. e parameters of the anterior cornea appeared on the le column were

recorded.

was comparable with previous studies which reported

the change ranged from 3.36 D to 5.8 D.

3-7

Because anterior corneal shape is the most important

parameter to determine corneal deformation aer ICRS

implantation

, anterior corneal topographic outcomes were

evaluated. ere was signicant change of keratometry

between visits. At the last visit, the median decrease of

Kmax was 6.00 D and the median decrease of Kmean

was 2.70 D. e results were comparable with a previous

report by Ferrara et al

that showed the Kmax change of

5.19 D and the Kmean change of 4.64 D postoperatively.

Moreover, other reports of the Ferrara type with 6 months

to 2 years follow up found the changes of the Kmean

between 0.09 D to 3.82 D.

5,6,10

Considering the other

commomly used Intacs type, the changes of the Kmean

were 1.94 D to 7.87 D.

10-16

Besides corneal attening and improving visual acuity,

another goal of the treatment in keratoconus is to improve

quality of vision. Corneal asphericity was reported to

increase in keratoconus and was proportional to the severity

(grading) of the disease.

In this study, aer the surgery,

the median anterior Q value showed signicant change

between follow-up visits (p = 0.01). e preoperative Q

value was -0.83, which demonstrated excessive prolateness

of the cornea. en cornea changed towards an oblate

cornea for a while at early postoperative 1 month (Q value

0.41). However, aer 6 month postoperatively, the Q

value changed toward minimal prolateness of the cornea

which was from -0.03 to -0.04 at the last visit. Similar to

this study, Torquetti et al reported that the Ferrara ICRS

signicantly reduced the Q value from -0.85 to -0.32 at

their 16-month follow up.

To illustrate, the anterior

corneal asphericity reduced aer ICRS implantation to

be more similar to the normal physiologic cornea in

which the most commonly accepted value in a young

adult population is approximately -0.23 ± 0.08.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

308

Considering comparison of the outcomes between

visits, we noted that the major parameters including

Kmax, keratometric astigmatism, and spherical equivalent

refraction were at their signicant outcome or relatively

highest eect at 6 months (p < 0.05), then the outcomes

were still stable with no regression at 2 years (p < 0.03).

is stability was consistent with the previous publications

with the long-term follow up.

19,20

Torquetti et al showed the

good stability of the outcome with the Kmax signicantly

decreased of 4.41 D at 5 years and later with no regression

at 10 years, consistently, and the Kmean with signicantly

decreased of 3.13 D at 5 years with no regression at 10

years aer surgery.

In contrast, Vega-Estrada et al

reported that the K reduction was observed only at 6

month aer sugery, then aer 6 months there was a

regression of Kmax for 3.31 D at 1 year and 3.14 D at

5 years, and a regression of Kmean for 3.14 D at 1 year

and 3.36 D at 5 years.

Postoperative adverse eects of glare and halos

occurred in 3 eyes. ese symptoms were also reported in

other studies.

22,23

However, none of them requested that

their implants to be removed. Small white intrastromal

corneal deposits around the segments were found in 2

eyes. is was reported to be frequently found (about 70%)

and not resulted in alteration of visual performance.

et al elucidated that the mechanical and physiologic

stresses introduced by the implantation lead to the

accumulation of lipid deposits in the extracellular matrix

of the cornea.

One case of extruded ICRS, which needed removal,

was reported. However, her visual acuity and refractive

status could return to preoperative baseline in months.

is reversible status was also reported in other studies.

26-28

In fact, this is an advantage of the surgery that the rings

are removable if complicated or poorly tolerated.

Ferrara

et al suggested that this complication could be minimized

aer mastering the surgical technique, especially in the

deep incision and the well-constructed intrastromal

tunnel.

is study has some mentionable limitations. One

was the retrospective study design. A second limitation

was the small sample size. Prevalence of keratoconus

varies in dierent parts of the world, but may be relatively

rare in ailand (no prevalence study in ai people).

Moreover, only some of the patients were eligible for

the benet of ICRS. e strength of this study is that,

ICRS implantation has been one of the internationally

approved treatment options for keratoconus and its

outcomes were rstly reported in ai patients. Further

study could include larger and wider range of patients

with keratoconus such as various preoperative grading

and progression. Moreover, the combined treatment

of ICRS implantation with the other novel treatment

of corneal collagen crosslinking for keratoconus could

be another aspect to study.

In conclusion, this initial report in ai patients with

keratoconus showed that the ICRS implantation signicantly

improved the visual, refractive, and topographic outcomes

during the 2-year follow up. e evidence seemed to

show that with the appropriate case selection and surgical

technique, the ICRS implantation was considered to be

an eective and safe procedure in most cases.

ACKNOWLEDGMENTS

We are grateful to Julaporn Pooliam, M.Sc., from the

Oce for Research and Development, for her assistance

with the statistical analyses, and to Mathuwan Srikong

from the Medical Education Technology Center, Faculty

of Medicine Siriraj Hospital, Mahidol University, for

preparing the gures.

REFERENCES

1. Colin J, Cochener B, Savary G, Malet F. Correcting keratoconus

with intracorneal rings. J Cataract Refract Surg 2000;26:1117-

22.

2. Barraquer JI. Modication of refraction by means of intracorneal

inclusions. Int Ophthalmol Clin 1966;6:53-78.

3. Gharaibeh AM, Muhsen SM, AbuKhader IB, Ababneh OH,

Abu-Ameerh MA, Albdour MD. KeraRing intrastromal corneal

ring segments for correction of keratoconus. Cornea 2012;31:

115-20.

4. Siganos D, Ferrara P, Chatzinikolas K, Bessis N, Papastergiou G. F

errara intrastromal corneal rings for the correction of keratoconus.

J Cataract Refract Surg 2002;28:1947-51.

5. Ferrara G, Torquetti L, Ferrara P, Merayo-Lloves J. Intrastromal

corneal ring segments: visual outcomes from a large case series.

Clin Exp Ophthalmol 2012;40:433-39.

6. Ameerh MA, Hamad GI, Ababneh OH, Gharaibeh AM, Refai

RM, Bdour MD. Ferrara ring segments implantation for treating

keratoconus. Int J Ophthalmol 2012;5:586-90.

7. Ferrara P, Torquetti L. Clinical outcomes aer implantation

of a new intrastromal corneal ring with a 210-degree arc length.

J Cataract Refract Surg 2009;35:1604-8.

8. Krumeich JH, Daniel J, Knulle A. Live-epikeratophakia for

keratoconus. J Cataract Refract Surg 1998;24:456-63.

9. Kling S, Marcos S. Finite-element modeling of intrastromal ring

segment implantation into a hyperelastic cornea. Invest

Ophthalmol Vis Sci 2013;54:881-89.

10. Haddad W, Fadlallah A, Dirani A, El Rami H, Fahd D, Khanafer

D, et al. Comparison of 2 types of intrastromal corneal ring

segments for keratoconus. J Cataract Refract Surg 2012;38:

1214-21.

11. Niknam S, Shamshiri M, Shahrzad SS, Alipour A, Rajabi MB,

Rajabi MT. Treatment of moderate to severe keratoconus with

6-mm Intacs SK. Int J Ophthalmol 2012;5:513-16.

12. Sansanayudh W, Bahar I, Kumar NL, Shehadeh-Mashour R,

Ritenour R, Singal N, et al. Intrastromal corneal ring segment

Chotikavanich et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

309

Original Article

SMJ

SK implantation for moderate to severe keratoconus. J Cataract

Refract Surg 2010;36:110-3.

13. Zare MA, Hashemi H, Salari MR. Intracorneal ring segment

implantation for the management of keratoconus: safety and

ecacy. J Cataract Refract Surg 2007;33:1886-91.

14. Colin J, Malet FJ. Intacs for the correction of keratoconus:

two-year follow-up. J Cataract Refract Surg 2007;33:69-74.

15. Hellstedt T, Makela J, Uusitalo R, Emre S. Treating keratoconus

with intacs corneal ring segments. J Refract Surg 2005;21:236-

46.

16. Kotb AM, Hantera M. Ecacy and safety of Intacs SK in

moderate to severe keratoconus. Middle East Afr J Ophthalmol

2013;20:46-50.

17. Safarzadeh M, Nasiri N. Anterior segment characteristics in

normal and keratoconus eyes evaluated with a combined

Scheimpug/Placido corneal imaging device. J Curr Ophthalmol

2016;28:106-11.

18. Torquetti L, Ferrara P. Corneal asphericity changes after

implantation of intrastromal corneal ring segments in keratoconus.

J Emmetropia 2010;1:178-81.

19. Fernandez-Vega Cueto L, Lisa C, Madrid-Costa D, Merayo-

Lloves J, Alfonso JF. Long-Term Follow-Up of Intrastromal

Corneal Ring Segments in Paracentral Keratoconus with

Coincident Corneal Keratometric, Comatic, and Refractive

Axes: Stability of the Procedure. J Ophthalmol 2017;2017:4058026.

20. Torquetti L, Ferrara G, Almeida F, Cunha L, Araujo LP, Machado

A, et al. Intrastromal corneal ring segments implantation

in patients with keratoconus: 10-year follow-up. J Refract Surg

2014;30:22-26.

21. Vega-Estrada A, Alio JL, Plaza-Puche AB. Keratoconus progression

aer intrastromal corneal ring segment implantation in young

patients: Five-year follow-up. J Cataract Refract Surg 2015;41:

1145-52.

22. Levinger S, Pokroy R. Keratoconus managed with intacs: one-

year results. Arch Ophthalmol 2005;123:1308-14.

23. Colin J. European clinical evaluation: use of Intacs for the

treatment of keratoconus. J Cataract Refract Surg 2006;32:

747-55.

24. Ruckhofer J, Twa MD, Schanzlin DJ. Clinical characteristics

of lamellar channel deposits aer implantation of intacs. J

Cataract Refract Surg 2000;26:1473-79.

25. Ly LT, McCulley JP, Verity SM, Cavanagh HD, Bowman RW,

Petroll WM. Evaluation of intrastromal lipid deposits aer

intacs implantation using in vivo confocal microscopy. Eye

Contact Lens 2006;32:211-5.

26. Asbell PA, Ucakhan OO, Abbott RL, Assil KA, Burris TE,

Durrie DS, et al. Intrastomal corneal ring segments: reversibility

of refractive eect. J Refract Surg 2001;17:25-31.

27. Chan SM, Khan HN. Reversibility and exchangeability of

intrastromal corneal ring segments. J Cataract Refract Surg

2002;28:676-81.

28. Alio JL, Artola A, Ruiz-Moreno JM, Hassanein A, Galal A,

Awadalla MA. Changes in keratoconic corneas aer intracorneal

ring segment explantation and reimplantation. Ophthalmology

2004;111:747-51.

29. Torquetti L, Ferrara G, Almeida F, Cunha L, Ferrara P, Merayo-

Lloves J. Clinical outcomes aer intrastromal corneal ring

segments reoperation in keratoconus patients. Int J Ophthalmol

2013;6:796-800.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

310

Jarurin Pitanupong, M.D., Katti Sathaporn, M.D.

Department of Psychiatry, Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkhla 90110, ailand.

The Prevalence and Factors Associated with

Mistreatment Perception among Thai Medical

Students in a Southern Medical School

Corresponding author: Jarurin Pitanupong

E-mail: pjarurin@medicine.psu.ac.th

Received 16 November 2018 Revised 17 May 2019 Accepted 31 May 2019

ORCID ID: http://orcid.org/0000-0001-9312-9775

http://dx.doi.org/10.33192/Smj.2019.47

ABSTRACT

Objective: To assess the prevalence of mistreatment perception among medical students as well as to identify the

types of mistreatment and their associated factors.

Methods: is cross-sectional study surveyed all of the 4

to 6

year medical students at the Faculty of Medicine,

Prince of Songkla University, from January to April 2017. ree questionnaires were employed: 1) Demographic

data 2) Mistreatment perception, and 3) the Patient Health Questionnaire (PHQ)-9 ai version. e data were

analyzed using descriptive statistics. e results were presented as frequency, percentage, average and standard

deviation. e factors associated with mistreatment perception were analyzed by means of the chi-square test and

logistic regression.

Results: Two hundred and ninety-eight medical students (55.0%) completed the questionnaires, 66.1% of them

were female. eir mean age was 22.5+1.1 years. As to the medical students’ perception, the majority (63.4%)

reported experiencing at least one incidence of mistreatment by attending physicians (53.7%), residents (36.2%)

and nurses (16.4%) within the previous year. e majority of mistreatment types were verbal criticism (59.7%) and

discriminative behavior (51.4%). e mistreated medical students reported consequences such as experiencing

unpleasant feelings (41.3%) and burnout (35.6%). According to the PHQ-9 ai version ndings, 11.1% of all of our

students had depression; however, depression did not correlate with the mistreatment perception. e signicant

factor that correlated with mistreatment perception was the academic year.

Conclusion: More than a half of the surveyed medical students perceived being mistreated and reported experiencing

unpleasant feelings as a consequence.

Keywords: Perception; mistreatment; prevalence; medical students (Siriraj Med J 2019; 71: 310-317)

Pitanupong et al.

INTRODUCTION

Medical student mistreatment has been recognized

for decades; it was initially described in 1982 by Henry

Silver, who highlighted its similarities to child abuse.

Examples of such mistreatment include inappropriate

physical contact, verbal abuse, sexual harassment and

power abuse.

In 1990, the rst document regarding

incidence, severity, and signicance of medical student

mistreatment were conducted, by a major medical school.

At same time while enrolled in medical school, 46.4%

of all medical students had been abused, and 80.6% of

seniors reported being abused by the senior year.

e

medical student mistreatment is known to adversely impact

students both personally and professionally. Similarly,

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

311

Original Article

SMJ

burnout has been shown to negatively impact students

;

69.1% of those abused reported that at least one of the

episodes they experienced was of “major importance

and very upsetting” and 49.6% of them indicated that

the most serious episode of mistreatment aected them

adversely for a month or more, whereas 16.2% of them

said that it would “always aect them”.

3-6

Concerning

discrimination, some studies reported that females were

signicantly more likely than males to report gender

discrimination, exclusion from informal settings and

discomfort from sexual humor. Typically, males low in

masculinity and females low in femininity were most

likely to report abuse.

Furthermore, the previous studies

have reported that abused medical students develop

increased levels of cynicism about medicine; a lack of

condence in performing certain skills; a low opinion

of the physician profession, an increase desire to drop

out of school; feelings of depression, suicide, anxiety,

burnout, anger, and hostility; a drinking problem;

4-6,7-13

a lower career satisfaction;

5,6,12

and become less likely to

pursue careers in academic medicine.

In 2011, the Association of American Medical

Colleges (AAMC) found that most medical students

(57.4% - 88.7%) in the United States were aware of

mistreatment at their school.

One study reported at least

one incident of medical student mistreatment by faculty

(64.0%) and residents (75.5%).

e mistreatment rates

were the lowest for family medicine and the highest for

obstetrics-gynecology and surgery specialties. However,

the perception of mistreatment in dierent departments

varied signicantly by gender and race.

e purpose of this study was to assess the prevalence

of mistreatment perception among medical students, as

well as identify various types of mistreatment and the

association between student perception and teaching

behavior. e ensuing data of this study will provide

preliminary useful evidence that will hopefully encourage

reforms in medical education in order to prevent and

manage mistreatment in medical schools appropriately.

Denition

Medical student mistreatment consists of inappropriate

physical contact (hitting, pinching, and throwing objects at

the student), verbal criticism or humiliation, discriminative

behavior, and power abuse or being forced to work outside

of one’s duty, which the medical students perceived to

be signicant causes of stress.

Major departments include internal medicine,

surgery, obstetrics and gynecology, orthopedic surgery

and physical medicine, and pediatrics.

Minor departments include anesthesiology, emergency,

otolaryngology head and neck surgery, ophthalmology,

radiology, family medicine and preventive medicine,

and psychiatry.

MATERIALS AND METHODS

is cross-sectional study was approved by e

Ethics Committee of the Faculty of Medicine, Prince

of Songkla University (REC: 60-472-03-4).

e study surveyed medical students from the 4

year that studying in the Faculty of Medicine, Prince of

Songkla University, Hat Yai Hospital Medical Education

Center, and Yala Hospital Medical Education Center at

the end (January to April) of the 2017 academic year.

e inclusion criterion was being a medical student who

could complete the questionnaire.

A medical students were contacted by the research

assistant in class, and were provided with the rationale

along with an overview of the research. Aer the self-

reporting questionnaires and information sheet were

distributed, medical student, they took 5-10 minutes to

consider whether to participate in the study or not. Aer

that, the research assistant handed out documentation

that assured the volunteers that their identities would be

protected. Adhering to a policy of strict condentiality,

the signatures of the participants were not required, and

all of the participants retained the right to withdraw

from the research at any time. e participants were

permitted to nish and return the questionnaires later

by 2 options; submitting them in the box at the front of

the classroom, or drop them in the box located at the

Psychiatry Department. us, participant condentiality

was protected.

Instruments

e questionnaire comprised 3 parts:

1) e demographic characteristics questionnaire

collected data regarding age, year of medical school,

religion, cumulative GPA, hometown, income, and

underlying disease.

2) e mistreatment perception and its consequences

questionnaire was edited by 5 psychiatrists. It employed

6 questions like, “Since the beginning of your clinical

rotations, how many times have you been mistreated in

the past year” e response categories were “never”, “once

or twice”, “a few times”, “several times”, and “numerous

times”. For the analysis, we classied mistreatment as never,

infrequent, or recurrent. Mistreatment was categorized

as infrequent if it occurred “once or twice” or “a few

times” and recurrent if it occurred “several times” or

“numerous times”.

3) e patient health questionnaire-9 ai version

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

312

(PHQ-9), consisted of 9 questions and employed a 4-point

rating Likert Scale to evaluate depression. e scores for

each question were between 0 and 4. e cut-o score

was > 9. e sensitivity and specicity were 84.0% and

77.0%, respectively. e positive and negative predictive

value were 21.0% and 99.0%, respectively. e internal

consistency reliability was assessed with a Cronbach’s

alpha coecient of 0.79-0.87. When the score was higher

than 9, it indicated depression.

Statistical analysis

e results are presented as percentage, frequency,

mean, and standard deviation. e factors associated

with mistreatment perception were analyzed using the

chi-square test for the univariate analysis and presented

using odds ratios.

RESULTS

Demographic data

e total number of medical students who completed

the questionnaires was 298; the response rate was 55.0%.

e sex proportion and demographic data of the responders

were not dierent from demographic data of the whole

population. Of the participants, 197 were female (66.1%)

(Table 1). e mean age was 22.5±1.1 years, whereas the

mean cumulative grade point average (GPA) was 3.3±0.3,

and the median income (IQR) was 8,000 (7,000-10,000)

Baht per month.

Perception of mistreatment

In the past year, the prevalence of mistreatment

perception among medical students was 63.4%, or 189 of

298 respondents. e gender proportion of mistreatment

perception was 58.4% of males and 66.0% of females.

According to the medical students’ perception, the majority

reported experiencing mistreatment by attending physicians,

residents and nurses (53.7%, 36.2% and 16.4% respectively)

(Fig 1). Most of the departments reporting mistreatment

were major departments. Meanwhile, the mistreatment

perception in the minor departments was of a lower rate.

e most frequently reported types of mistreatment were

verbal criticism (59.7%) and discriminative behavior

(51.4%) (Fig 2). e mistreated medical students reported

experiencing consequences such as unpleasant feelings

(41.3%), boredom and burnout (35.6%), anxiety (28.2%)

and anger (20.1%) (Table 2).

e medical student’s perception of positive teaching

behaviors were the giving of encouragement, being

praised, and giving advice on how to improve, whereas

the unpleasant teaching behaviors were blaming or

criticizing, devaluation, looking down or insult, and

ignorance (Fig 3).

e internal consistency reliability of our study was

0.62 for perception of mistreatment.

Depression screening

According to the PHQ-9 ai version results, 11.1%

of our medical students had PHQ-9 > 9, that indicated

depression (Table 1). e internal consistency reliability

of our study was 0.87 for depression screening.

Factors associated with perception of mistreatment

e variables whose p-values from the univariate

analysis were lower than 0.2 were included in the initial

model; only one variable, academic year, met this criterion

(p=0.002) (Table 3). An analysis using odds ratios revealed

that, the 5

year medical students had a 2.5 times greater

risk for mistreatment perception when compared with

year medical students (95% CI = 1.5, 4.3) (Table 4).

Moreover, we found that the 5

year medical students

had a 2.8 times greater risk for mistreatment perception

compared with 6

year medical students (95% CI = 1.2,

6.5). However, the level of depression did not associate

with the medical students’ mistreatment perception.

DISCUSSION

is study revealed that the prevalence of mistreatment

perception among medical students in southern ailand

within 1 year was of the same rate as the data reported

from other medical schools. We discovered that the

prevalence of medical student mistreatment perception

in our population was 63.4% whereas the prevalence of

mistreatment among medical students in other countries

has been reported at 64.0% by faculty and 75.5% by

residents, respectively.

e most commonly reported

types of mistreatment were verbal criticism (59.7%) and

discriminative behavior (51.4%). Our ndings concurred

with those reported by a 2011 study of 24 dierence

medical schools in the United States of America.

Even

though these data may not provide any new information

as it regards the worldwide body of knowledge, they

constitute the rst useful concrete evidence from ailand

on this topic.

Concerning the consequences of mistreatment among

medical students, a previous study identied a relationship

between verbal mistreatment and condence in clinical

abilities.

Moreover, recurrent mistreatment has been

associated with a high rate of burnout (57.0%)

whereas in

this study, the mistreated medical students reported

experiencing unpleasant feelings (41.3%) and burnout

Pitanupong et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

313

Original Article

SMJ

TABLE 1. Demographic characteristics (n=298).

Demographic characteristics Number (%)

Sex

Male 101 (33.9)

Female 197 (66.1)

Academic year

year 157 (52.7)

year 113 (37.9)

year 28 (9.4)

Religion

Buddhism 265 (88.9)

Islam 16 (5.4)

Christianity 7 (2.3)

Other 4 (1.3)

Unreported 6 (2.0)

Home province

Songkhla 115 (38.6)

Other 178 (59.7)

Unreported 5 (1.7)

Underlying disease

No 250 (83.9)

Yes 46 (15.4)

Medical illness e.g. allergic rhinitis 31 (10.4)

Psychiatric illness e.g. dysthymia 15 (5.0)

Unreported 2 (0.7)

Depression by PHQ-9

No depression 238 (79.9)

Depression 33 (11.1)

Unreported 27 (9.1)

Depression categorized by academic year

year 20 (12.7)

year 11 (9.7)

year 2 (7.1)

Depression categorized by gender

Male 9 (27.3)

Female 24 (72.7)

Abbreviation: PHQ-9 = Patient Health Questionnaire-9

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

314

TABLE 2. Consequence of being mistreated within 1 year (n=298).

TABLE 3. Crude association between general characteristics and mistreatment perception.

Type Number (%)

Unpleasant feeling but tolerable experience 123 (41.3)

Boredom and burnout 106 (35.6)

Stress and anxiety 84 (28.2)

Anger 60 (20.1)

Sadness and depression 59 (19.8)

Fight or perseverance 34 (11.4)

Escape/ try to terminate study program 33 (11.1)

Low self-esteem and has suicide ideation 11 (3.7)

Need for psychological therapy 5 (1.7)

Felt shame 6 (2.0)

General characteristics

Experience of mistreatment Chi

Yes (n=189) No (n=109) P-value

Sex 0.247

Male 59 (31.2) 42 (38.5)

Female 130 (68.8) 67 (61.5)

Academic year 0.002

year 88 (46.6) 69 (63.3)

year 86 (45.5) 27 (24.8)

year 15 (7.9) 13 (11.9)

Cumulative GPA 0.528

<3.00 33 (23.6) 23 (30.7)

3.01-3.50 70 (50.0) 34 (45.3)

3.51-4.00 37 (26.4) 18 (24.0)

Home province 0.637

Songkhla 71 (38.0) 44 (41.5)

Other 116 (62.0) 62 (58.5)

Underlying disease 0.447

No 156 (83.0) 94 (87.0)

Yes 32 (17.0) 14 (13.0)

Depression (PHQ-9) 0.611

No 151 (86.8) 87 (89.7)

Yes 23 (13.2) 10 (10.3)

Note: ere were missing values for some variables

Pitanupong et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

315

Original Article

SMJ

TABLE 4. Factors related to mistreatment perception.

Factor

Crude OR P-value

(95% CI) Wald’s test

Academic year

year Reference

year 2.5 (1.5, 4.3) < 0.001

year 0.9 (0.4, 2.0) 0.808

Abbreviation: OR = odds ratio, CI = condence interval

53.7

36.2

16.4

5.0

3.7

0.3

100

Attending

physician

Resident Nurse Friend Senior medical

student

Other

(%)

Person associated with mistreatment

Fig 1. Person associated with mistreatment (can be more than one type)

12.1

0.7

39.9

19.8

30.9

20.5

14.1

4.0

100

Infrequent Recurrent Infrequent Recurrent Infrequent Recurrent Infrequent Recurrent

Physical Verbal Discriminative Power abuse

(%)

Type of mistreatment

Fig 2. Type of mistreatment (can be more than one type)

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

316

(35.6%) as the two most common consequences

of mistreatment. This indicates that mistreatment

impacts medical students negatively both personally

and professionally. Regarding the factor associated with

mistreatment perception, this study identied an association

between academic year and risk of mistreatment perception.

Since no previous study has reported such correlation,

future in-depth analyses should be undertaken in order

to explain this phenomenon.

Moreover, this study identied positive and unpleasant

teaching behaviors on the part of the instructing medical

personnel that may be useful for reforming our medical

educational system. ey suggested that the teaching

behaviors they appreciated were receiving encouragement,

praise, and recommendations or advice whereas the

unpleasant teaching behaviors were blaming, criticizing,

devaluation and ignorance. erefore, we need to ask

ourselves the question, “What is the teacher’s role?”. is

study might have reected the current teacher role in the

ai medical education. Consequently, the reformation

of our medical education should be further enhanced.

In other words, well-designed curriculum reforms or

the adoption of sound concepts regarding the role and

conduct of teachers that reect the highest philosophy

and values need to be considered by faculty.

As it is well-known, the processes of teaching for

eective learning such as team-based learning, idea

sharing and interaction need safe environments. Such

environments require the teacher to understand the

students, and build strong rapport with them by eective

listening and communication. Furthermore, additional

necessary skills of teachers include being intentional

about constructive interpersonal interaction by oering

positive reinforcement and feedback.

An example of an appropriate teacher role is found

in preclinical education, where the medical students

are educated via the problem-based learning (PBL)

process, which requires the teacher to act as a coach.

Additionally, even in the clinical-teaching process, our

study, revealed that the medical students want the teacher

to act like a coach. ey suggested that the teacher should

be encouraging and oer positive feedback. erefore,

our values and philosophy of teaching should be about

the creation of a climate of trust and respect, being open

and responsive, and evaluating the needs of medical

students.

Limitations

is study had a cross-sectional design and employed

self-reporting for individual perception assessment.

45.6

46.6

38.6

21.8

0 10 20 30 40 50 60 70 80 90 100

Encouragement: never mind, try again next

time

Praise: good, great, excellent

Recommendation or advice for improvement

Open to inquiry: do you have any question?

Constructive

71.5

37.2

6.0

2.7

0 10 20 30 40 50 60 70 80 90 100

Blame or criticism

Devaluation, looking down or insulting

Being ignored

Being offensive

Unpleasant

Fig 3 Medical students’ perception of positive/constructive and unpleasant teaching behavior (can be more than one type)

Pitanupong et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

317

Original Article

SMJ

Moreover, its response rate was 55.0%, which might

have led to information bias. In addition, the sample

size was limited to only medical students in the Faculty

of Medicine, Prince of Songkla University. us, it is

too soon to generalize our ndings to the nation-wide

setting.

Implications and future recommendations

Further studies should employ a more quantitative

method and cover more medical schools within ailand.

In so saying, a multi-center study is recommended.

CONCLUSION

More than a half of our medical students perceived

having been mistreated and reported experiencing unpleasant

feelings. A new philosophy and values regarding teaching

and the role of the teacher in the learning process should

be adopted in education.

ACKNOWLEDGMENTS

is research was fully funded by the Faculty of

Medicine, Prince of Songkla University, ailand. e

authors are very grateful for the data analysis conducted

by Mrs. Nisan Werachattawan and Ms. Kruewan

Jongborwanwiwat. Moreover, we would like show our

appreciation to every medical students who participated

in this study.

REFERENCES

1. Silver HK. Medical students and medical school. JAMA

1982;247:309-10.

2. Bursch B, Fried JM, Wimmer PF, Cook IA, Baillie S, Zackson H,

et al. Relationship between medical student perceptions of

mistreatment and mistreatment sensitivity. Med Teach

2013;35:998-1002.

3. Silver HK, Glicken AD. Medical student abuse. Incidence,

severity and signicance. JAMA 1990;263:527-32.

4. Cook AF, Arora VM, Rasinski KA, Curlin FA, Yoon JD. e

prevalence of medical student mistreatment and its association

with burnout. Acad Med 2014;89:749-54.

5. Frank E, Carrera JS, Stratton T, Bickel J, Nora LM. Experiences

of belittlement and harassment and their correlates among

medical students in the United States: longitudinal survey.

BMJ 2006;333:682.

6. Stratton TD, McLaughlin MA, Witte FM, Fosson SE, Nora

LM. Does students’ exposure to gender discrimination and

sexual harassment in medical school aect specialty choice

and residency program selection? Acad Med 2005;80:400-8.

7. Richman JA, Flaherty JA, Rospenda KM, Christensen ML.

Mental health consequences and correlates of reported medical

student abuse. JAMA 1992;267:692-4.

8. Heru A, Gagne G, Strong D. Medical student mistreatment results

in symptoms of posttraumatic stress. Acad Psychiatry 2009;33:

302-6.

9. Dahlin ME, Runeson B. Burnout and psychiatric morbility

among medical students entering clinical training: a three year

prospective questionnaire and interview-based study. BMC

Med Educ 2007;7:6.

10. Schuchert MK. The relationship between verbal abuse of

medical students and their condence in their clinical abilities.

Acad Med 1998;73:907-9.

11. Frank E, Brogan D, Schiman M. Prevalence and correlates

of harassment among US women physicians. Arch Intern Med

1998;158:352-8.

12. Sheehan KH, Sheehan DV, White K, Leibwitz A, Baldwin

DC. A pilot study of medical student abuse: student perceptions of

mistreatment and misconduct in medical school. JAMA

1990;263:533-7.

13. Elnicki DM, Curry RH, Fagan M, Friedman E, Jacobson E,

Lous T, et al. Medical students’ perspective on and responses

to abuse during the internal medicine clerkship. Teach Learn

Med 2002;14:92-7.

14. Haviland MG, Yamagata H, Werner LS, Zhang K, Dial TH,

Sonne JL. Student mistreatment in medical School and planning

a career in academic medicine. Teach Learn Med 2011;23:

231-7.

15. Richardson DA, Becker M, Frank RR, Sokol RJ. Assessing

medical students’ perceptions of mistreatment in their second

and third years. Acad Med 1997;72:728-30.

16. Lotrakul M, Sumrithe S, Saipanish R. Reliability and validity

of the ai version of the PHQ-9. BMC Psychiatry. 2008;8:46.

17. Schuchert MK. The relationship between verbal abuse of

medical students and their condence in their clinical abilities.

Acad Med 1998;73:907-9.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

318

INTRODUCTION

Myopic strabismus xus, also known as heavy eye

syndrome, is an acquired progressive esotropia and

hypotropia associated with high axial myopia. Typically,

this condition tends to develop esotropia with restricted

abduction which is oen accompanied by hypotropia and

limited abduction. Various theories about the etiology

of this clinical entity have been postulated including the

increased weight of the eyeball or forward movement of the

center of globe; that is ‘heavy eye syndrome’

, myopathic

paralysis of lateral rectus (LR) muscle by pressure from

the lateral orbital wall

, and the mechanical restriction

limiting motility due to contact between posterior elongated

globe and orbital wall.

3,4

Recently, many investigators

Pittaya Phamonvaechavan, M.D., Piangporn Saksiriwutto, M.D.

Department of Ophthalmology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.

Surgical Treatment of Myopic Strabismus Fixus by

Loop Myopexy Augmented with Scleral Fixation:

a Case Report

Corresponding author: Pittaya Phamonvaechavan

E-mail: pittaya2002@gmail.com

Received 6 February 2019 Revised 24 April 2019 Accepted 14 May 2019

ORCID ID: http://orcid.org/0000-0002-7514-8224

http://dx.doi.org/10.33192/Smj.2019.48

ABSTRACT

Objective: To present the clinical ndings, surgical procedure and long-term outcome of a ai patient with

myopic strabismus xus.

Case presentation: e patient presented with recurrent progressive esotropia and hypotropia. Two strabismus

surgeries were performed 20 years ago. Magnetic resonance imaging (MRI) ndings indicated the etiology which

was secondary to inferiorly deviated lateral rectus (LR) muscle and medially deviated superior rectus (SR) muscle.

e patient had undergone loop myopexy of lateral rectus and superior rectus muscles augmented with scleral

xation and satisfactory result was achieved up to six years aer the operation.

Conclusion: e patient with myopic strabismus xus should be evaluated by MRI to conrm the etiology of this

disorder. Marked esotropia and hypotropia can be safely aligned by simple loop myopexy augmented with scleral

xation.

Keywords: Myopic strabismus xus; heavy eye syndrome; high myopia; loop myopexy (Siriraj Med J 2019; 71: 318-321)

Phamonvaechavan et al.

reported MRI ndings of displacement of both LR and

superior rectus (SR) muscles pulley system producing this

condition.

5-7

Surgical correction of this restrictive motility

disorder is dicult. Multiple surgical procedures have

been described; for example large bilateral medial rectus

(MR) recession and recession-resection procedure, but

results were unsatisfactory.

Yokoyama et al

proposed

the loop myopexy technique of SR and LR muscle as

an alternative treatment for this disorder. is report

demonstrates the clinical and radiological ndings as

well as outcome of surgical treatment with loop myopexy

augmented with scleral xation of a patient with myopic

strabismus xus.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

319

Case Report

SMJ

CASE REPORT

A 61-year-old man with a 30-year history of esotropia

and high myopia presented with increasing esotropia and

hypotropia. Twenty years ago, the patient underwent

MR muscle recession 9 mm, LR muscle resection 10 mm,

superior oblique tenectomy on the right eye and MR

recession 6 mm on the le eye. His right eye gradually

increased inward and downward at 6 months aer surgery,

and thereaer the patient was lost to follow up. On

examination, best-corrected visual acuity was hand motion

OD, 6/24 OS with le eye xation preference. Both eyes

revealed unremarkable anterior segment examinations

except cataract and myopic degeneration on dilated

fundus examination. Posterior segment on the right eye

could not be evaluated due to extreme adduction and

depression position of the right eye. In primary position,

the right cornea was shown only 2 mm of superolateral

part as Fig 1a. Ocular movement on the right eye was

severely limited to abduct (-4) and elevate (-4) beyond

the midline while only mildly limited to abduct (-trace)

on the le eye as shown in Fig 1b. On coronal view of

MRI orbit displayed a nasally deviated SR muscle and an

inferiorly deviated LR muscle on the right eye as shown

in Fig 2. Orthoptic examination revealed an esotropia

of 95 prism diopters (PD) and a right hypotropia of 25

PD with the le eye xing.

At the time of operation, forced duction testing

showed marked restriction of MR muscle, moderate

restriction of inferior rectus (IR) muscle on the right

eye and mild restriction of MR muscle on the le eye.

Also, a nasally displaced course of SR muscle and an

inferiorly displaced course of LR muscle on the right eye

were conrmed. On the right eye, MR muscle insertion

was found 7 mm from limbus whereas LR and SR muscle

insertions were attached at the normal position. Strabismus

surgery performed for the right eye consisted of looping

of SR and LR muscles together and anchoring to the

sclera 15 mm behind the limbus using a 5/0 polyester

and MR muscle was re-recessed 12 mm from the limbus.

On the rst postoperative month, the patient had

satisfactory alignment with a 12 PD exotropia and 8 PD

right hypotropia as shown in Fig 3. Motility markedly

improved in abduction, but limited in adduction (-2)

on the right eye. Although visual acuity on the right eye

improved to 6/24 aer strabismus surgery, the patient

still had signicant cataract on both eyes. Axial length

measurements conrmed aer strabismus surgery were

31.35 mm OD and 31.94 mm OS. Hence, 4 months

later cataract surgery was performed on both eyes and

vision reached 6/7.5 OU. Fig 4 disclosed the rst year

postoperative alignment. e patient had no double

vision, no stereopsis on Fly test and alternate suppression

on Worth-4-dot testing. ese outcomes were stable 6

years postoperatively.

Fig 1. (A) Preoperative appearance in primary position, le eye xating, (B) Preoperative photograph of ocular motility

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

320

Fig 2. MRI of the orbit before surgery: bilateral displacement of superior recti nasally and lateral recti inferiorly and the oval distortion of

the globe

Fig 3. Postoperative photograph of this case, 1 month aer surgery

Fig 4. Postoperative photograph of this case, 1 year aer surgery

DISCUSSION

Several reports have demonstrated pathogenesis of

restricted strabismus in myopic strabismus xus including

displacement of LR and SR muscle pulley system

the mechanical limitation of rectus muscles due to the

contact between the posterior elongated globe and the

orbital apex.

Many surgical procedures to correct the

alignment in this condition have been reported

; for

example, simple recession-resection procedures, and

disinsertion or large recession of MR muscle. ese

procedures showed eective results in the early stages,

but later esotropia and hypotropia gradually returned.

Likewise, our patient had esotropia and hypotropia again

aer previous large MR recession and LR resection.

Recent ndings showed shiing of LR muscle inferiorly

and SR muscle nasally caused by a posterior prolapse of

an elongated eye beyond the muscle cone due to superior,

posterior and lateral protrusion of posterior eyeball.

6,11

e deviant paths of the SR and LR muscles in our case

were conrmed with both MRI preoperatively and at the

time of surgery. erefore, the goal of surgical correction

was to reestablish the physiological muscle plane by

binding the LR and SR muscles to each other and with

or without suturing to sclera at equator. Recent studies

Phamonvaechavan et al.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

321

Case Report

SMJ

have described the eectiveness of this surgery including

simple loop myopexy with or without anchoring to

sclera, muscle belly union of half tendon of the SR and

LR muscles combined with MR recession.

8,12-14

Outcomes

showed improved alignment and motility.

Loop myopexy of the SR and LR muscles was

performed to normalize their course in our case. A

suture looped around the two muscles was secured to

the sclera to stabilize the muscle path and to avoid risk

of suture migration. Also, the use of a nonabsorbable

suture supports the stability of suture placement over

time. According to pathogenesis of this disorder, this

technique is more eective than the conventional recess-

resect procedures because it moves the deviant path of

extraocular muscles to the desired position to straighten

the eye in primary position. Also, this technique minimizes

the risk of anterior segment ischemia since both LR

and SR muscles were not disinserted. However, the

risk of scleral perforation may be encountered in this

procedure due to scleral suture. Aer extraocular muscle

surgery, the patient underwent cataract extraction with

intraocular lens implantation in both eyes and vision

reached 6/7.5 in both eyes. Both ocular alignment and

visual results were favorable and stable throughout 6 years

of follow-up.

In summary, patients with high myopia, progressive

esotropia and hypotropia should be assessed by MRI

preoperatively to conrm the cause of this condition.

Loop myopexy augmented with scleral xation was

sucient to correct both large horizontal and vertical

deviation and maintain ocular alignment for a longer

period. Hence, this surgical method is recommended in

myopic strabismus xus.

REFERENCES

1. Basghaw J. e “heavy eye” phenomenon: a preliminary report.

Br J Ophthalmol 1966;23:73-8.

2. Bagolini B, Tamburrelli C, Dickmann A, Colosimo C. Convergent

strabismus xus in high myopic patients. Doc Ophthalmol

1990;74:309-20.

3. Demer JL, von Noorden JK. High myopia as an unusual cause

of restrictive motility disturbance. Surv Ophthalmol 1989;33:

281-4.

4. Webb H, Lee J. Acquired distance esotropia associated with

myopia. Strabismus 2004;12:149-55.

5. Aoki Y, Nishida Y, Hayashi O, Nakamura J, Oda S, Yamada

S, et al. Magnetic resonance imaging measurements of extraocular

muscle path shi and posterior eyeball prolapse from the muscle

cone in acquired esotropia with high myopia. Am J Ophthalmol

2003;136:482-9.

6. Krzizok TH, Schroeder BU. Measurement of recti eye muscle

paths by magnetic resonance imaging in highly myopic and

normal subjects. Invest Ophthalmol Vis Sci 1999;40:2554-60.

7. Krzizok TH, Kaufmann H, Traupe H. New approach in

strabismus surgery in high myopia. Br J Ophthalmol 1997;81:

625-30.

8. Rowe FJ, Noonan CP. Surgical treatment for progressive

esotropia in the setting of high myopia. JAAPOS 2006;10:

596-97.

9. Yokoyama T, Ataka S, Tabuchi H, Shiraki K, Miki T. Treatment of

progressive esotropia caused by high myopia – a new surgical

procedure based on its pathogenesis. In: de Faber JT, editor.

Transactions: 27

Meeting, European Strabismological

Association, Florence, Italy, 2001. Lisse (Netherlands): Swets

& Zeitlinger; 2002.p.145-8.

10. Hayashi T, Iwashige H, Maruo T. Clinical features and surgery

for acquired progressive esotropia associated with severe

myopia. Acta Ophthalmol Scand. 1999;77:66-71.

11. Yokoyama T, Tabuchi H, Ataka S, Shiraki K, Miki T, Mochizuki

K. e mechanism of development in progressive esotropia

with high myopia. In: de Faber JT, editor. Transactions: 26

Meeting, European Strabismological Association, Barcelona,

Spain, 2000. Lisse (Netherlands): Swets & Zeitlinger; 2001.

p.218-21.

12. Wong I, Leo S, Khoo B. Loop myopexy for treatment of myopic

strabismus xus. JAAPOS 2005;9:589-91.

13. Ahadzadeghan I, Akbari MR, Ameri A, Anvari F, Jafari AK,

Rajabi MT. Muscle belly union for treatment of myopic

strabismus xus. Strabismus 2009;17:57-62.

14. Durnian JM, Maddula S, Marsh IB. Treatment of “heavy eye

syndrome” using simple loop myopexy. JAAPOS 2010;14:

39-41.

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

322

Pseudohyperkalemia Caused by EDTA

Contamination: a Not Uncommon Pre-analytical

Error

Laiwejpithaya et al.

To the Editor,

Pseudohyperkalemia from potassium ethylene-

diaminetetraacetic acid (EDTA) contamination has been

reported for decades. e mechanism of this anticoagulant

is chelation of calcium, which is a divalent cation that is

essential for the proper function of the clotting cascade.

EDTA contamination is most oen caused by inappropriate

order of blood draw, and it can be recognized by the

apparent presence of spurious coexisting hyperkalemia

and hypocalcemia.

We report the case of a 93-year-old cachexic ai

male who was diagnosed as having diabetes mellitus in

a hyperosmolar hyperglycemic state. His laboratory test

results showed markedly hyperkalemia and hypocalcemia.

Electrocardiogram (ECG) showed sinus rhythm without

tall peak T wave or widening of the QRS complex. e

discordance between the ECG result and the reported

hyperkalemia prompted a second blood draw to rule

out pseudohyperkalemia in the rst sample (Table 1).

ereaer, 20 ml of 10% calcium gluconate was pushed

intravenously to treat the patient’s hypocalcemia.

At our central laboratory and due to the abrupt

change in potassium level, total calcium was performed in

the second specimen which showed normocalcemia. No

signicant hemolysis was observed in either the rst or

second specimen. e patient had no known underlying

medical conditions that could have been the cause of

pseudohyperkalemia, such as thrombocytosis or marked

leukocytosis. From these evidences together with the

coexisting spurious hyperkalemia and hypocalcemia in the

rst sample, the most likely cause of pseudohyperkalemia

was judged to be potassium EDTA contamination.

e mechanisms of EDTA contamination can be

described, as follows: syringe needle contamination due

to the lling of blood into the EDTA sample tube before

other tubes, pouring the blood from EDTA sample tube

into other tubes

1-3

, and mistakenly switching the cap

of EDTA sample tubes with others. e person who

took the rst blood specimen claimed that there was

no transfer of blood or switching of tube caps from one

tube to another, but could not conrm that the proper

order of blood draw had been followed. According to

European Federation of Clinical Chemistry and Laboratory

Medicine guidelines, the order of blood draw should be,

as follows: blood culture tubes, coagulation tubes, serum

tubes, heparin tubes, EDTA tubes, glycolytic inhibitor

tubes, and then other tubes.

In our case, the rst blood

draw included lithium heparin tube, K2EDTA tube,

and sodium uoride tube. ere was, therefore, a high

probability that blood was collected in the K2EDTA

tube rst, followed by the lithium heparin tube, which

would have resulted in K2EDTA contamination in the

lithium heparin tube.

TABLE 1. Patient blood glucose and electrolyte results.

Test Unit 1

sample 2

sample Normal range

Blood glucose mg/dl 1,031 869 74-99

Sodium mmol/L 150 153 136-145

Potassium mmol/L 9.9 4.2 3.4-4.5

Chloride mmol/L 109 115 98-107

Bicarbonate mmol/L 22.0 19.0 22-29

Total calcium mg/dl 3.5 9.08 8.2-9.6

Volume 71, No.4: 2019 Siriraj Medical Journal

www.smj.si.mahidol.ac.th

323

Letter to the Editor

SMJ

Corresponding author: Sathima Laiwejpithaya

E-mail: sathima.lai@mahidol.edu

Received 14 January 2019 Revised 31 May 2019 Accepted 12 June 2019

ORCID ID: http://orcid.org/0000-0002-5838-6722

http://dx.doi.org/10.33192/Smj.2019.49

The failure of our laboratory staff to recognize

coexisting distinct hyperkalemia and hypocalcemia in

the rst sample result caused misinterpretation and

inappropriate calcium replacement in our patient.

Fortunately, the patient showed no sign of hypercalcemia

aerwards, and the follow-up calcium level was within

normal range.

Recognition of coexisting hyperkalemia and

hypocalcemia in the same sample strongly suggests

an EDTA contaminated sample. A campaign designed

to remind and reinforce appropriate blood collection

procedures would help to decrease the frequency, but not

totally eliminate EDTA contamination.

1,5

Furthermore, it

only takes a very small amount of EDTA contamination to

inuence a pseudohyperkalemia and pseudohypocalcemia

result, and the presence of EDTA can be dicult to

identify.

1-3

ree previous studies reported that a lack of

routine EDTA measurement as a screening protocol would

result in 36-42% of pseudohyperkalemic samples going

undetected and being reported as true hyperkalemia.

1,2,5

us, a cost-eective strategy would be to implement a

routine EDTA analysis protocol.

1-3

In conclusion, lack of awareness about EDTA

contamination and clue for how to identify it can lead

to misinterpretation and inappropriate treatment. We

would like to highlight the need for ongoing reinforcement

of the importance of blood draw protocols, and the need

for implementation of EDTA assay for routine EDTA

screening.

ACKNOWLEDGMENTS

Research author Sudarat Piyophirapong was supported

by a Chalermphrakiat Grant, Faculty of Medicine Siriraj

Hospital, Mahidol University. e authors gratefully thank

Dr. Kosit Sribhen, Department of Clinical Pathology,

Faculty of Medicine Siriraj Hospital, Mahidol University

for his expert advice.

REFERENCES

1. Sharratt CL, Gilbert CJ, Cornes MC, Ford C, Gama R. EDTA

sample contamination is common and oen undetected, putting

patients at unnecessary risk of harm. Int J Clin Pract 2009;63:

1259-62.

2. Cornes MC, Ford C, Gama R. Spurious hyperkalaemia due

to EDTA contamination: common and not always easy to

identify. Ann Clin Biochem 2008;45:601-3.

3. Chadwick K, Whitehead SJ, Ford C, Gama R. kEDTA Sample

Contamination: A Reappraisal. J Appl Lab Med 2019;3:925-35.

4. Cornes M, van Dongen-Lases E, Grankvist K, Ibarz M, Kristensen

G, Lippi G, et al. Order of blood draw: Opinion Paper by the

European Federation for Clinical Chemistry and Laboratory

Medicine (EFLM) Working Group for the Preanalytical Phase

(WG-PRE). Clin Chem Lab Med 2017;55:27-31.

5. Bouzid K, Bartkiz A, Bouzainne A, Cherif S, Ramdhani S, Zairi

A, et al. How to reduce EDTA contamination in laboratory

specimens: a Tunisian experience. Clin Chem Lab Med 2015;53:

e9-e12.

Sathima Laiwejpithaya, M.D., Sudarat Piyophirapong, M.D.

Department of Clinical Pathology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand