The Fontan operation has low mortality, but is associated with significant postoperative morbidity, including prolonged chest tube output (PCTO), which is associated with prolonged hospital length of stay (PLOS). We sought to identify variables present early in the clinical course that could predict patients at high risk for PCTO and PLOS. Retrospective data were collected on 84 Fontan (extracardiac conduit) operations from 1/2008 to 12/2013 at a single institution. PCTO was defined as ≥8 days (>75th percentile); PLOS was defined as ≥12 days postoperatively (>75th percentile). Multivariate regression was used to determine covariates associated with PCTO and PLOS. Median age was 3.5 years (IQR 3–5); weight was 14.5 kg (IQR 13–17). There was no mortality. LOS was 9 days (IQR 3–11), and duration of chest tube drainage 6 days (IQR 5–8) at 15 ml/kg/day (IQR 9–20). In univariate analysis, only systemic right ventricle, 24-h 5 % albumin administration, 24-h fluid balance, and 12-h inotrope score were associated with PCTO. In multivariate analysis, only 5 % albumin administration in first 24 h (p < 0.001) and PCTO were independently associated with PLOS. ROC curve analysis showed patients receiving >25 ml/kg of 5 % albumin in first 24-h predicted PLOS (94 % specificity, 93 % sensitivity, AUC = 0.95, p < 0.001). Increased colloid in the first 24-h post-CPB strongly predicts PCTO and PLOS after Fontan operation, potentially providing an early identification of a cohort with unfavorable Fontan physiology. A better understanding of the role of colloid resuscitation after Fontan is necessary, and efforts to reduce perioperative colloid administration could decrease hospital morbidity