The Cleft Palate–Craniofacial Journal 00(00) pp.

000–000 Month 0000

Ó Copyright 2016 American Cleft Palate–Craniofacial Association

ORIGINAL ARTICLE

Palatal Fistula Risk after Primary Palatoplasty: A Retrospective Comparison

of a Humanitarian Organization and Tertiary Hospitals
Kimberly M. Daniels, M.S., Emily Yang Yu, B.S., Rebecca G. Maine, M.D., M.P.H., Yin Heng, M.D.S., Li Yang, D.D.S.,
Ph.D., Bing Shi, D.D.S., Ph.D., D. Scott Corlew, M.D., M.P.H., William Y. Hoffman, M.D., F.A.A.P., George A.
Gregory, M.D.

Background: Humanitarian surgical organizations provide palatoplasties for patients without

access to surgical care. Few organizations have evaluated the outcomes of these trips. This
study evaluates the palatal fistula rate in patients from two cohorts in rural China and one in the
United States.
Methods: This study compared the odds of fistula formation among three cohorts whose
palates were repaired between 2005 and 2009. One cohort included 97 Chinese patients operated
on by teams from the United States and Canada under the auspices of Resurge International.
They were compared to cohorts at Huaxi Stomatology Hospital and the University of California
San Francisco (UCSF). Age, fistula presence, and Veau class were compared among cohorts
using Chi-square tests. Logistic regression was used to analyze predictors of fistula formation.
Results: The fistula risk was 35.1% in patients treated by humanitarian teams, 12.8% at Huaxi
University Hospital and 2.5% at UCSF (P ,0.001). Age and Veau class were associated with
fistula formation (Age P ¼ 0.0015; Veau P ,0.001). ReSurge and Huaxi patients had 20.2 and 5.6
times the odds of developing a fistula, respectively, compared to UCSF patients (P ,0.01, both).
A multivariable model controlling for surgical group, age, and gender showed an association
between Veau class and the odds of fistula formation.
Conclusions: Chinese children undergoing palatoplasty by international teams had higher
odds of palatal fistula than children treated by Chinese surgeons in established institutions and
children treated in the United States. More research is required to identify factors affecting
complication rates in low-resource environments.

KEY WORDS: fistula, palatoplasty

A cleft palate occurs in 1:700 to 1:500 live births,
depending on ethnicity (Derijcke et al., 1996; Mossey, 2004;
Maine et al., 2012; Centers for Disease Control and
Prevention, 2013). A cleft palate can affect speech, hearing,
eating, and appearance (Maine et al., 2012; Thanawiratta-
Kimberly Daniels is Research Data Manager, Department of Plastic
and Oral Surgery, Boston Children’s Hospital, Boston, MA. Emily Yu
and Dr. Maine are senior residents, Department of Surgery, University
of California San Francisco, San Francisco, CA. Dr. Heng is
SpeechTherapist, Cleft Department, West China Hospital of Stoma-
tology, Sichuan University; Dr. Yang is Associate Professor, Deputy
Director, Cleft Department, West China Hospital of Stomatology,
Sichuan University; Dr. Shi is Professor and Vice Dean, West China
Hospital of Stomatology, Sichuan University, Sichuan Province,
China. Drs. Corlew and Hoffman are Professors of Surgery, Division
of Plastic and Reconstructive Surgery, University of California San
Francisco; Dr. Gregory is Professor, Departments of Anesthesia and
Pediatrics, University of California San Francisco, San Francisco, CA
This study was funded by Fulbright. No products, devices or drugs
were used in this research.
Poster presentations: London Launch of the Lancet Commission on
Global Surgery at the Royal Society of Medicine, London, UK April
27–28, 2015. This abstract was also published by The Lancet on April
27, 2015.
Acknowledgements: We thank Libby Wilson, M.D., for her help
confirming the fistula diagnosis in the ReSurge patients.
nanit et al., 2012; Deshpande et al., 2014). In high-income
countries, cleft palates are usually repaired by 18 months of
age to reduce problems with speech (Chapman et al., 2008;
Deshpande et al., 2014). These children have a team of
plastic surgeons, otolaryngologists, audiologists, speech
pathologists, dentists, orthodontists, and nursing specialists
working together to ensure successful surgical, dental,
speech, and hearing outcomes (American Cleft Palate-
Craniofacial Association, 1993; Furr et al., 2011; Maine et
al., 2012).
Unfortunately, many children in low- and middle-
income countries (LMIC) have no access to surgeons who
can perform palate repairs (Bermudez, 2000; Mossey, 2004;
Maine et al., 2012). Consequently, these children depend on
humanitarian organizations to provide free reconstructive
surgery (Bermudez, 2000; ReSurge International, 2014).
Submitted January 2016; Revised April 2016, August 2016,
August 2016; Accepted September 2016.Month 2016
Address correspondence to: Kimberly M. Daniels, Department of
Plastic and Oral Surgery, 300 Longwood Avenue, Boston, MA 02115.
Email: Kimberly.Daniels@childrens.harvard.edu
DOI: 10.1597/16-007

0
0 Cleft Palate–Craniofacial Journal, Month 0000, Vol. 00 No. 00
Most organizations sponsor short-term trips intended to
provide patient care rather than research. Follow-up care is
provided on repeat trips or by local providers if the patients
return for follow-up. Patients often cannot return for
follow-up appointments due to poor infrastructure, limited
resources, and the distance they live from the surgical site
(Mossey, 2004; Maine et al., 2012; Jansen et al., 2014). This
makes collecting detailed, accurate patient records and
assessing their surgical outcomes difficult. As a result, few
studies have evaluated the outcomes of palate repairs in
LMICs following mission trips.
Palatal fistula is defined as an opening in the soft or hard
palate connecting the oral cavity to the nasal cavity. It is a
common complication following palatoplasty, with a
reported risk of 2% to 30% (Parwaz et al., 2009; Maine
et al., 2012; Deshpande et al., 2014; Schönmeyr et al., 2016).
Like a cleft palate, a palatal fistula also makes it difficult to
speak and eat properly (Deshpande et al., 2014). Previously
identified risk factors for fistula formation include older age
at the time of surgery, surgeon skill and experience, larger
cleft size, Veau classification, surgical technique used to
close the palate, tension along the suture line, and the
patient’s oral health (Emory et al., 1997; Turner et al., 1998;
Andersson et al., 2008; Bindingnavele et al., 2008;
Deshpande et al., 2014; Schönmeyr et al., 2016).
This study compares the probability of developing a
fistula after primary palatoplasty in three cohorts: (1) a
humanitarian plastic surgery organization, ReSurge Inter-
national (international), (2) West China Stomatology
Hospital (West China), and (3) the University of California
San Francisco Craniofacial Center (UCSF). Outcomes
were compared to determine if there are factors that
influence fistula formation in these populations.
METHODS
This was a retrospective study of three cohorts that
underwent primary palatoplasty between 2005 and 2009.
Patients with alveolar clefts were excluded from this study.
The international cohort consisted of 201 patients identified
through China’s Family Planning Committee and ReSurge
International’s records. A total of 116 patients from Akesu,
Qiqihaer, Ganzhou, and Chongqing were located by study
staff and asked to participate in this study between October
2009 and June 2010. Informed oral assent was obtained
from all patients over 6 years of age (as required by UCSF’s
Committee on Human Subjects Protection and the NIH).
Informed consent was also obtained from the parents or
guardians of all patients under 18 years in this cohort prior
to evaluation. Patients were excluded from analyses if
ReSurge did not perform their primary palate surgery, that
is, it was performed by other groups. Fifteen board-certified
or board eligible (five Webster Fellows) American and
Canadian surgeons performed the palatoplasties. One
fellow participated each year over the period 2005 to
2009. All fellows completed a residency in plastic surgery
and their participation in these procedures was during a
postdoctoral year. Postoperative evaluations were conduct-
ed between 1 and 5 years after palate surgery.
Each international patient answered a questionnaire and
underwent a brief physical examination. The questionnaire
assessed demographics, socioeconomic status, and medical
history. Examinations determined presence of a fistula,
quantified dental caries, and evaluated speech and hearing.
A medical student (E.Y.Y.) was trained over a 3-month
period by WYH and his team at UCSF to perform these
evaluations and acquire all photos of the palate accurately
and in the same way. She used a palatal mirror after it was
warmed (to avoid fogging) to capture a photo of each
child’s palate. She went to China for 1 year, accompanied
by a translator for each interview and evaluation. Palatal
fistulae were confirmed using the photos of the palate and
questions about symptoms caused by fistulae by a third-
party blinded plastic surgeon not involved in patients’
original surgical care.
For comparison, fistula risk and demographic informa-
tion were collected from the medical records of a second
cohort of 123 children who underwent palatoplasty
between 2006 and 2009 at UCSF and evaluated postoper-
atively annually with the first follow-up at 6 months. One
plastic surgeon (W.Y.H.) operated on all UCSF patients
and participated in their postoperative evaluations. He and
his team determined the presence or absence of a fistula for
this cohort. Our third cohort—a random sample from all
economic classes of 253 subjects operated on by four
surgeons between 2005 and 2009—was collected from the
West China Stomatology Hospital in the Sichuan Province
of China. Fistula risk and demographic information was
collected from the medical records of this group as well.
Similar to UCSF, in this hospital, presurgical orthopedics is
included in the treatment plan of the patient when needed.
Our colleague (S.B.) ensured that fistulae were assiduously
looked for at follow-up and all fistulae noted were reported
in patients’ medical records. The earliest follow-up visit was
at 1 month for one patient in this cohort. All other patients
had a follow-up visit at least 2 months after palatoplasty
with an average follow-up period of 15 months for this
cohort. These patients were not contacted because they
were evaluated by their surgeons. Informed consent was
waived by the UCSF committee on human protection
because only routinely collected data were reviewed at these
two institutions. The West China Stomatology Hospital’s
Human Subjects Protection Committee also approved of
this study.
This study was approved by the Human Subjects
Committees at UCSF. Analysis of the de-identified data
was approved by the Harvard University IRB. China does
not have an IRB process but China’s Family Planning
Committee, a government agency, agreed to participate
and our Human Subjects Committees decided this was
satisfactory.
 Daniels et al., PALATAL FISTULA RISK IN UNITED STATES AND CHINA 0

TABLE 1 Descriptive Statistics

Variable ReSurge International West China UCSF P-value*

N 116 253 123

Excluded 20 3 3
Fistula 35.4% 12.8% 2.5% ,0.001
Male 57 (58.7%) 134 (53.6%) 73 (60.8 %) 0.726
Age ,0.001
,2 years 15 (15.5%) 104 (41.6%) 109 (90.8%)
2–4 years 27 (27.8%) 69 (27.6%) 9 (7.5%)
4 years 55 (56.7%) 77 (30.8%) 2 (1.7%)
Veau class ,0.001
I 9 (9.3%) 23 (9.2%) 7 (5.8%)
II 11 (11.3%) 124 (49.6%) 41 (34.2%)
III 53 (54.6%) 79 (31.6%) 48 (40.0%)
IV 23 (23.7%) 24 (9.6%) 21 (17.5%)
* P-values are all from chi-square tests of association.
This data was collected from study participants in all three cohorts and compared here. There is a statistically significant association between the cohorts and fistula presentation, age
at palatoplasty, and Veau class.

Analysis cleft or fistula instead of a palatal cleft (2), or if surgeons

A power analysis was done to be sure this study was
adequately powered. Chi-square tests were used to
compare the Veau classification distributions and the
fistula proportions. Age distributions between the
international, Huaxi, and UCSF cohorts were compared
with a Kruskall-Wallace test. Where significant differ-
ences among the three cohorts were found, a Scheffe’s
test or a Bonferroni correction was used to determine
the cohorts that were significantly different from one
another.
Logistic regression was used to create a multivariable
predictive model to assess the effects of the surgical
cohort, gender, age at palate surgery, and Veau class on
the odds of developing a fistula. Within the internation-
al cohort, univariate logistic regression models were
used to estimate each variable’s effect on the odds of
developing a palatal fistula. A sensitivity analysis
predicted the number of patients who would have to
be fistula-free among the patients lost to follow-up to
change our results. Wald confidence intervals and P-
values are reported for each variable in the univariate
analysis and multivariable analysis. P-values less than
from groups other than ReSurge performed their primary
palatoplasty (15). Medical records were available for 120
(97.6%) UCSF patients. Of the 1612 patients who had their
palatoplasty at the West China Stomatology Hospital, 775
(48.1%) returned for a follow-up clinic visit and 253 were
randomly selected from this group. Complete data was
available for 250 subjects.
Demographics of the three cohorts are compared in
Table 1; descriptive statistics of the international popula-
tion are in Table 2. Of the international patients, 35.4% had
a fistula at the time of evaluation, 12.8% of West China
patients had a fistula, and only 2.5% of UCSF patients had
a fistula; all were significantly different from each other (P
,0.001). The median age of the international patients at the
time of palatoplasty was 4.5 years (IQR: 2.47 to 8.02 years).
In the West China cohort, it was 2.73 years (IQR: 1.61 to
4.73 years), and at UCSF it was 9.8 months (IQR: 9.5 to
11.3 months). The age distributions were all significantly
different (P ,0.001). Most children in the international and
TABLE 2 ReSurge Descriptive Statistics
Variable ReSurge International*

0.05 were considered statistically significant. Data were N 97

analyzed using SAS 9.3 (SAS Institute Inc, Cary, NC) Median travel time (mins) to physician 15 (IQR: 5, 30)

and STATA 12.0 software (StataCorp LP, College Dental caries

Proportion 84%
Station, TX).
Average number 5.79 6 4.6
Proportion who have seen dentist 9.3%
RESULTS Proportion living in rural farming region 87.6%
Received pre-surgical orthodontics 4 (4.1%)

A total of 1936 patients underwent primary palatoplasty
during the study period: 201 with ReSurge, 1612 at the
West China Stomatology Hospital, and 123 at UCSF. In
the international group, the follow-up rate was 57.7%.
There were 96 (82.7%) patients included in the analysis
from the international group. Patients in this group were
excluded due to missing data (3 patients), if they had a nasal
Problems gaining weight before surgery
1 71 (73.2%)
2 7 (7.22%)
3 8 (8.3%)
4 11 (11.3%)
Median income/month (Yuan) 2000 (IQR: 1000, 3000)
*thinsp;This data was collected from the ReSurge Cohort via interview and
evaluation. None of the ReSurge subjects received speech therapy or regular primary
care.
0 Cleft Palate–Craniofacial Journal, Month 0000, Vol. 00 No. 00

TABLE 3 Multivariable Analysis of Predictors of Presenting with a Fistula

Unadjusted Analysis Multivariate Analysis

Parameter Crude OR Confidence Interval* P-value* Adjusted OR Confidence Interval* P-value*

Surgical group
ReSurge versus UCSF 20.213 (5.959, 68.561) ,0.001 15.581 (5.055, 68.538) ,0.001
West China versus UCSF 5.575 (1.671, 18.594) 0.005 7.027 (2.321, 30.542) 0.002
Age at surgery (years) 1.071 (1.024, 1.122) 0.004 1.018 (0.963, 1.075) 0.534
Gender, M versus F 1.370 (0.805, 2.330) 0.246 1.042 (0.578, 1.877) 0.892
Veau, III, IV versus I, II 6.399 (3.182, 12.871) ,0.001 5.630 (2.677, 11.837) ,0.001
* Wald Confidence Intervals and P-values.

UCSF cohorts were Veau class III and IV (78.3% and
57.5%, respectively); 58.8% of the West China cohort was
Veau class I or II. The distribution of cleft types across the
Veau classification varied among the three groups (P
,0.001).
A multivariable model analyzed the effects of gender,
Veau class, age at primary palate surgery, and the cohort/
surgical group on the presence of a fistula (Table 3).
Although age at palatoplasty was associated with the odds
of having a fistula in the unadjusted analysis, it was not
significant after controlling for the surgical group, gender,
and Veau class. The strongest effects were seen from the
surgical group and the Veau class. The odds of a child in the
international cohort developing a fistula were 15.6 times the
odds a child from UCSF doing so in the multivariable
analysis. In the same model, the odds a child in the West
China cohort presenting with a fistula were 7.0 times the
odds of a child at UCSF.
Univariate models were analyzed within the international
data set to look for predictors of fistula development (Table
4). None of the variables were significantly associated with
fistula development.
DISCUSSION
This study found that patients operated on by the
international team had the highest odds of fistula
formation, followed by the children at the West China
Stomatology hospital. Those at UCSF had the lowest odds
of developing a fistula during the study period. The
outcomes of local surgeons (West China and UCSF
surgeons) were better than those of international teams
with respect to fistula formation. The surgical group and
Veau classes III and IV were associated with increased risk
of fistula formation in a multivariable model controlling for
the patient’s age at palatoplasty and gender. The incidence
of higher Veau classes in the international group may
account for some of the difference found. More detailed
analysis of the international cohort did not identify other
variables associated with fistula formation.

TABLE 4 Univariate Analysis of Each Variable’s Association With the Presence of a Fistula Within the ReSurge Cohort

Lower 95% Upper 95%

Variable Odds Ratio Confidence Bound Confidence Bound Wald P-value

Veau classification*
II versus I 4.571 0.409 51.138 0.217
III versus I 3.778 0.437 32.672 0.227
IV versus I 8.727 0.935 81.492 0.057
Gender, M versus F 1.894 0.774 4.635 0.162
Number of dental carries 1.002 0.910 1.104 0.962
Neighborhood type, urban city versus rural farming 0.325 0.067 1.579 0.164
Highest grade† 0.923 0.780 1.093 0.354
Monthly income,‡ 1 versus 0 0.449 0.116 1.735 0.245
Age at palate repair 1.005 0.894 1.131 0.928
Has seen a dentist,§ 1 versus 0 1.520 0.380 6.085 0.554
Travel time to a physician 1.006 0.989 1.023 0.517
Frequency of seeing a physician|| || 1.279 0.897 1.823 0.173
Oral health impression 0.898 0.610 1.323 0.587
Trouble gaining weight#
1 versus 0 2.556 0.528 12.360 0.243
2 versus 0 1.917 0.440 8.346 0.386
3 versus 0 0.426 0.085 2.130 0.299
All data were collected from interviews with the parent or primary caregiver that accompanied the child to the assessment.
* Veau class was also analyzed as a dichotomous variable (class III, IV versus I, II) and this resulted in an OR: 1.851 (0.608, 5.633) and a Wald P-value of 0.2782.
†
Highest grade represents the highest grade of education completed by either of the parents of the subject.
‡
Monthly income was dichotomized where 1 ¼ top 25% and 0 ¼ otherwise.
§
Represents whether or not the child has ever seen a dentist and coded as yes ¼ 1, no ¼ 0.
|| ||
Collected on a 5- point scale from 1 to 5 in order of increasing frequency.
Collected on a 1 to 5 scale in order of decreasing severity.
#
Collected on a 4-point scale in order of increasing difficulty gaining weight.
 Daniels et al., PALATAL FISTULA RISK IN UNITED STATES AND CHINA
The outcomes of palatoplasty procedures in LMIC have
been studied to a limited extent (Maine et al., 2012;
Abdurrazaq et al., 2013; Deshpande et al., 2014; Hardwicke
et al., 2014; Zhang et al., 2014; Fayyaz et al., 2015;
Schönmeyr et al, 2015; Schönmeyr et al., 2016). Our study,
the Maine et al. study, and a study from Lagos University
Teaching Hospital in Nigeria found higher rates of fistula
formation in LMIC compared to the United States (Maine
et al., 2012; Abdurrazaq et al., 2013). An Indian study of
1608 patients found a 13.6% fistula risk, similar to our
12.8% risk in the Huaxi cohort (Schönmeyr et al., 2016).
Nonetheless, recent literature review showed ‘‘[no] differ-
ence in incidence of fistula formation between populations
from Europe, the Americas, or Asia’’ (Hardwicke et al.,
2014). The only Chinese study we found reported a fistula
rate of 0.95% in their study population; however, their
study population was not defined and cannot be compared
to our West China data (Zhang et al., 2014). While the risk
of fistula after primary palatoplasty varies for many reasons
between 0.95% and 54% in the literature, comparable
palatoplasty outcomes have been reported between LMIC
and high-income countries (Maine et al., 2012; Hardwicke
et al., 2014; Zhang et al., 2014).
Less is known about palatoplasty outcomes from
surgical missions. Maine et al. evaluated Ecuadorian
patients who underwent cleft palate repair for the presence
of a fistula and found that the risk of palatal fistula was the
same (56%) whether the surgeon was visiting or Ecuador-
ian (Maine et al., 2012). Both groups used the same mission
model used by many groups. For nonmission patients
(private patients) of the Ecuadorian surgeons, fistula
formation was 6% (Maine et al., 2012). We also found
similar results from our Chinese cohorts. A humanitarian
organization in Pakistan found that fistula risk decreased
with increased surgeon experience (Fayyaz et al., 2015).
They report a fistula rate of 50% when resident surgeons
performed the surgery, which was not the case in our
international group (Fayyaz et al., 2015). When performed
by experienced surgeons, their fistula risk was 9.6%, similar
to that found in Finland (Lithovius et al., 2014; Fayyaz et
al., 2015). ReSurge International vets all surgeons and
allows only board-certified plastic surgeons to operate on
missions (Maine et al., 2012). All surgeons who apply to
work under the auspices of ReSurge International provide
recommendation letters from other plastic surgeons
confirming the surgeon’s level of experience with cleft
repairs (Maine et al., 2012). Although Webster Fellows
participated in ReSurge procedures, they were always
supervised by the board-certified, vetted plastic surgeons
who were responsible for the patients in our study. None of
these Fellows operated outside of the scope of practice
standard to the United States.
While the vetting of these surgeons suggests that they
have the skill and experience to repair cleft palates in their
own practice with good results, it does not guarantee
experience repairing cleft palates in a low-resource setting in
0
older patients who may have wider, more technically
difficult palates to repair (Rossell-Perry et al., 2015;
Schönmeyr et al., 2016). Peruvian surgeons using the same
mission model had a 25% risk of fistula on their missions
but only a 3.82% risk in their private patients (Rossell-
Perry et al., 2015). We did not have specific information on
the surgeons performing the palatoplasties in the interna-
tional or West China cohorts and so cannot draw any
conclusions about the impact of the surgeon or their
experience and skill level on the risk of their patients
presenting with a fistula. More research is required to
determine why patients in the mission model have more
fistulae than patients who receive their surgery in other
ways. Understanding this difference will allow patients
treated on mission trips to have improved outcomes and
reduced incidence of fistula formation. Having such data
would allow all humanitarian organizations to learn from
the successes and failures of other hospitals and for
nongovernmental organizations (NGO) to improve their
quality of care.
The multivariable model included the surgical group that
performed the palatoplasty, gender, the age at primary
palate surgery, and Veau class as independent variables.
Fistula presentation was the dependent variable. Crude
estimates of all of these variables, except gender, were
significantly associated with the presentation of a fistula;
however, when combined into a multivariable model, only
the surgical group and Veau class were associated with
having a fistula. This finding is supported by other studies
showing that increased Veau class contributes to fistula risk
(Maine et al., 2012; Hardwicke et al., 2014; Lithovius et al.,
2014; Ahmed et al., 2015; Schönmeyr et al., 2016).
We could not identify why some children in the
international cohort had fistulas and others did not. This
may be due to a lack of variability in socioeconomic status,
dental health, medical care, and types of food consumed
(Table 2). The majority of patients lived in rural
communities where they grew most of their own food,
had access to electricity and water, had poor oral health,
and saw a physician only when they were sick.
Limitations
The follow-up rate for the international group was
57.7%, which is high for mission surgery but was
achieved only by sending someone (E.Y.Y.) to China
for a year to collect the data. Though similar to other
global health studies in LMIC, this follow-up rate still
brings up concerns related to selection bias due to loss
to follow-up (Maine et al., 2012). The West China
cohort is limited in its generalizability because the
cohort of patients we randomly sampled represented
48.1% of the West China population who returned for a
follow-up visit and, therefore, our results are represen-
tative only of patients who returned for follow-up. Both
these cohorts dealt with a lack of follow-up. The
0 Cleft Palate–Craniofacial Journal, Month 0000, Vol. 00 No. 00
difficulties of treating patients, conducting research, and
following up with patients in low-resource settings have
been well-documented in the literature (Maine et al.,
2012; Jansen et al., 2014; Hackenberg et al., 2015;
Schönmeyr et al., 2016). It is possible that patients in
either of these cohorts without problems and no fistula
did not return for follow-up or that patients with
significant problems did not do so because travel was
expensive, far away, or they felt nothing more could be
done. Some may not even know how to recognize a
surgical complication. Therefore, the bias could be
positive or negative. Our sensitivity analysis showed
that in the unlikely event that none of the international
patients who were lost to follow-up had a fistula, they
would still have a significantly higher fistula rate
(16.9%) than UCSF. Furthermore, if fewer than 4.2%
of the 525 West China patients not included in our
random sample had a fistula, then our results would be
reversed and their outcomes would not be statistically
different from UCSF.
Due to the retrospective nature of this study, data
were not available on postoperative care, surgical
technique, tension on suture lines at the time of
operation, the number of upper respiratory infections,
antibiotic use, the surgeon’s skill and experience, or the
patient’s nutritional status at time of surgery. Nor were
data available on oral hygiene and caries in patients in
either the Huaxi or UCSF cohorts. Differences in these
variables may contribute to the fistula risk (Andersson
et al., 2008; Furr et al., 2011; Lithovius et al., 2014;
Ahmed et al., 2015; Fayyaz et al., 2015). Operating on
patients before their teeth erupt, as well as better dental
care (as was done at UCSF) may explain some of the
difference in the incidence of fistula formation among
international, Huaxi, and UCSF patients. The use of
antibiotics before and after surgery and the develop-
ment of postoperative fistula formation is a fertile field
for study. Finally, the generalizability of these cohorts is
limited because the West China Stomatology Hospital
and the UCSF Craniofacial Center are large centers for
palatoplasty surgery with patients traveling great
distances to seek care. This makes it more difficult to
define their patient population.
CONCLUSIONS
This study found significant differences among the
proportion of fistula in the UCSF, Huaxi, and international
cohorts with and without controlling for gender, Veau
class, and the age at primary palate repair. The fistula rates
were 35.4%, 12.8%, and 2.5% for the international, Huaxi,
and UCSF cohorts, respectively. We do not believe these
differences are due to surgical skill and experience, as the
surgeons are well-vetted by ReSurge. Prospective, detailed
follow-up is necessary to determine the cause for these
differences among these populations and determine why
fistula rates vary widely among humanitarian groups,
countries, and local environments. The success of a surgical
mission must be determined by both patient outcomes and
the number of surgeries performed. It is essential that
outcomes achieved on missions are objectively evaluated
and the principles of constant improvement be implement-
ed. Constant evaluation and revision of strategic concepts,
programs, and policies are critical to the success of
humanitarian organizations in their efforts to improve
global care and standards. It is only through this critical
evaluation that innovative ways to improve the care
provided will continue to be developed.
REFERENCES
Abdurrazaq TO, Micheal AO, Lanre AW, Olugbenga OM, Akin LL.
Surgical outcome and complications following cleft lip and palate
repair in a teaching hospital in Nigeria. Afr J Paediatr Surg.
2013;10:345–357.
Ahmed MK, Maganzini AL, Marantz PR, Rousso JJ. Risk of
persistent palatal fistula in patients with cleft palate. JAMA Facial
Plast Surg. 2015;17:126–130.
American Cleft Palate-Craniofacial Association. Parameters for
evaluation and treatment of patients with cleft lip/palate or other
craniofacial anomalies. Cleft Palate Craniofac J. 1993;S30(suppl 1).
Available at http://www.acpa-cpf.org/uploads/site/Parameters_
Rev_2009.pdf. Accessed October 10, 2015.
Andersson EM, Sandvik L, Semb G, Abyholm F. Palatal fistulas after
primary repair of clefts of the secondary palate. Scand J Plast
Reconstr Surg Hand Surg. 2008;42:296–299.
Bermudez LE. Operation Smile: plastic surgery with few resources.
Lancet. 2000;356(suppl):s45.
Bindingnavele VK, Bresnick SD, Urata MM, Huang G, Leland HA,
Wong D, Hammoudih J, Reinisch J. Superior results using the
islandized hemipalatal flap in palatoplasty: experience with 500
cases. Plast Reconstr Surg. 2008;122:232–239.
Centers for Disease Control and Population. Facts about cleft lip and
palate. Available at http://www.cdc.gov/ncbddd/birthdefects/
cleftlip.html. Accessed September 20, 2015.
Chapman KL, Hardin-Jones MA, Goldstein JA, Halter KA, Havlik
RJ, Schulte J. Timing of palatal surgery and speech outcome. Cleft
Palate Craniofac J. 2008;45:297–308.
Daniels KM, Yu EY, Maine RG, Corlew DS, Bing S, Hoffman WY,
Gregory GA. Palatal fistula risk after primary palatoplasty: a
retrospective comparison of a humanitarian organization and
tertiary hospitals. The Lancet. 2015;385:S37.
Derijcke A, Eerens A, Carels C. The incidence of oral clefts: a review.
Br J Oral Maxillofac Surg. 1996;34:488–494.
Deshpande GS, Campbell A, Jagtap R, Restrepo C, Dobie H, Keenan
HT, Sarma H. Early complications after cleft palate repair: a
multivariate statistical analysis of 709 patients. J Craniofac Surg.
2014;25:1614–1618.
Emory RE, Clay RP, Bite U, Jackson IT. Fistula formation and repair
after palatal closure: an institutional perspective. Plast Reconstr
Surg. 1997;99:1535–1538.
Fayyaz GQ, Gill NA, Ishaq I, Ganatra MA, Mahmood F, Kashif M,
Alam I, Chen PK, Laub DR. A model humanitarian cleft mission:
312 cleft surgeries in 7 days. Plast Reconstr Surg Glob Open.
2015;3:e313.
Furr MC, Larkin E, Blakeley R, Albert TW, Tsugawa L, Weber SM.
Extending multidisciplinary management of cleft palate to the
developing world. J Oral Maxillofac Surg. 2011;69:237–241.
Hackenberg B, Ramos MS, Campbell A, Resch S, Finlayson SR,
Sarma H, Howaldt HP, Caterson EJ. Measuring and comparing the
 Daniels et al., PALATAL FISTULA RISK IN UNITED STATES AND CHINA
cost-effectiveness of surgical care delivery in low-resource settings:
cleft lip and palate as a model. J Craniofac Surg. 2015;26:1121–
1125.
Hardwicke JT, Landini G, Richard BM. Fistula incidence after
primary cleft palate repair: a systematic review of the literature.
Plast Reconstr Surg. 2014;134:618e–627e.
Jansen LA, Carillo L, Wendby L, Dobie H, Das J, Restrepo C,
Campbell A. Improving patient follow-up in developing regions. J
Craniofac Surg. 2014;25:1640–1644.
Lithovius RH, Ylikontiola LP, Sándor GK. Incidence of palatal fistula
formation after primary palatoplasty in northern Finland. Oral
Surg Oral Med Oral Pathol Oral Radiol. 2014;118:632–636.
Maine RG, Hoffman WY, Palacios-Martinez JH, Corlew DS,
Gregory GA. Comparison of fistula rates after palatoplasty for
international and local surgeons on surgical missions in Ecuador
with rates at a craniofacial center in the United States. Plast
Reconstr Surg. 2012;129:319e–326e.
Mossey P. Report of a WHOM on International Collaborative Research
on Craniofacial Anomalies. Geneva: WHO; 2004.
Parwaz MA, Sharma RK, Parashar A, Nanda V, Biswas G, Makkar
S. Width of cleft palate and postoperative palatal fistula – do they
correlate? J Plast Reconstr Aesthet Surg. 2009;62:1559–1563.
0
ReSurge International. ReSurge International. Available at: http://
www.resurge.org/home/home.cfm. Accessed March 5, 2014.
Rossell-Perry P, Segura E, Salas-Bustinza L, Cotrina-Rabanal O.
Comparison of two models of surgical care for patients with cleft lip
and palate in resource-challenged settings. World J Surg.
2015;39:47–53.
Schönmeyr B, Wendby L, Campbell A. Early surgical complications
after primary cleft lip repair: a report of 3108 consecutive cases.
Cleft Palate Craniofac J. 2015;52:706–710.
Schönmeyr, B., Wendby, L., Campbell, A. Surgical complications in
1408 primary cleft palate repairs operated at a single center in
Guwahati, Assam, India. Cleft Palate Craniofac J. 2016;53:278–
282.
Thanawirattananit P, Prathanee B, Thanaviratananich S. Audiological
status in patients with cleft lip and palate at Srinagarind Hospital. J
Med Assoc Thai. 2012;95:S93–S99.
Turner C, Zagirova AF, Frolova LE, Courts FJ, Williams WN. Oral
health status of Russian children with unilateral cleft lip and palate.
Cleft Palate Craniofac J. 1998;35:489–494.
Zhang Z, Fang S, Zhang Q, Chen L, Liu Y, Li K, Zhao Y. Analysis of
complications in primary cleft lips and palates surgery. J Craniofac
Surg. 2014;25:968–971.