Understanding Early Fontan Failure in Vietnam: Incidence and Clinical Predictors.

BackgroundWe aimed to determine the incidence of early Fontan failure (EFF) in a contemporary series of patients from Vietnam and identify risk factors for EFF.MethodsA total of 145 consecutive patients underwent the Fontan procedure at E hospital in Vietnam from August 2012 through December 2019. We analyzed predictive factors for EFF based on clinical evaluation, literature review, and our institutional database. The primary outcome assessed was EFF.ResultsThe incidence of EFF was 9.66%. In a univariate analysis of preoperative data, the anatomic diagnosis of unbalanced atrioventricular (AV) septal defect, situs inversus, AV valve regurgitation, large aortopulmonary circulation on Doppler echocardiogram, elevated pulmonary artery pressure (PAP), and elevated pulmonary vascular resistance (PVR) were significantly associated with EFF. Four intraoperative risk factors influencing EFF included pulmonary artery reconstruction, AV valve repair, bleeding, and elevated PAP. Postoperative edema was also significantly associated with EFF. A total of 22 potential independent variables were put into a model with multivariate logistic regression analysis. A final reduced model following utilizing a stepwise backward selection strategy indicated preoperative elevated PAP (odds ratio [OR]: 1.84, 95% CI: 1.12-3.00,  = .016), AV valve repair at Fontan (OR: 65.85, 95%CI: 1.95-2228.14,  = .020), and postoperative elevated PAP (OR: 1.66, 95%CI: 1.19-2.33,  = .004) were independent predictors for EFF.ConclusionEarly Fontan failure was relatively high in our case series and was associated with a significant mortality rate. Patients with single ventricle physiology having preoperative elevated PAP, intraoperative AV valve repair, and postoperative elevated PAP were identified as independent risk factors to predict EFF. Particularly, in the context of a developing country with limited health care resources, a comprehensive management strategy that targets the variable predictors for EFF should be incorporated with previously known risk factors to avoid EFF.