BackgroundHumanitarian surgical organisations provide cleft palate repair for patients without access to surgical care. Despite decades of experience, very little research has assessed the outcomes of these trips. This study investigates the fistula rate in patients from two cohorts in rural China and one in the USA. MethodsThis retrospective study compared the odds of fistula presentation among three cohorts whose palates were repaired between April, 2005, and November, 2009. The primary cohort included 97 Chinese patients operated on in China by surgeons from ReSurge International. A second Chinese cohort of 250 patients was operated on at Huaxi University Hospital by Chinese surgeons. The third cohort of 120 patients from the University of California San Francisco (UCSF) was included for comparison over the same time period; data was taken from medical records. Age, fistula presentation, and Veau Class were compared between the three cohorts with χ2 tests. Logistic regression was used to analyse predictors of fistula presentation among the three cohorts. This study received institutional review board approval from the UCSF, the Harvard School of Public Health, and physicians at Huaxi University Hospital, and written consent was obtained from study participants in China. FindingsThe fistula risk was 35·4% in ReSurge patients, 12·8% for patients at Huaxi University Hospital, and 2·5% for patients at UCSF (p<0·001). At the time of surgery 15·5% of the ReSurge patients were younger than 2 years old, whereas 90·8% of the UCSF children and 41·6% of the Huaxi children were (p<0·001). In the ReSurge cohort, 20·6% of patients had a Veau class of I or II, wheras 40·8% and 58·9% of UCSF and Huaxi patients, respectively, were in class I or II (p<0·001). Age and Veau Class were associated with fistula formation in a univariate analysis. (Veau Class III or IV vs I or II, odds ratio [OR] 6·399 [95% CI 3·182–12·871]; age, OR 1·071 [95% CI 1·024–1·122]). A multivariate model controlling for the surgical group, age at palatoplasty, and sex showed an association between Veau Class and the odds of fistula presentation (Class III or IV vs I or II, OR 5·630 [95% CI 2·677–11·837). In this model, UCSF patients and Huaxi patients had 0·064 and 0·451 times the odds of developing a fistula, respectively, compared with ReSurge patients (p<0·001 both). InterpretationChinese children undergoing palatoplasty on surgical missions have higher post-operative odds of palatal fistula than do children treated by local physicians. Children in low-resource settings have higher complication rates than do children in high-resource settings. Older age at palatoplasty and a Veau class III and IV are associated with post-palatoplasty fistula. Furthermore demographic, socioeconomic, and cultural differences could play a part in palatoplasty fistula outcomes between these three populations. More research is needed to determine the effects of post-operative care, the skill of the providers, and the technique used in the surgery that play a role on fistula outcomes after primary palatoplasty, particularly in low-resource environments. FundingNone.