BACKGROUND Children with complex congenital heart disease entail risk when undergoing noncardiac operations and other procedures requiring general anesthesia. To address concerns regarding intraoperative instability, need for postoperative mechanical ventilation, and postoperative hospital length of stay (LOS), we present our 5-year experience with 71 patients with complex congenital heart disease who underwent 252 surgical procedures. METHODS We reviewed the records of all patients from July 2006 to January 2011 who underwent a cardiac procedure with a Risk Adjustment for Congenital Heart Surgery-1 score of 6, and included all who underwent noncardiac procedures during this interval. Perioperative data were gathered to identify patients at risk for induction and maintenance instability, need for postoperative mechanical ventilation, and postoperative hospital LOS. Univariate predictors of these outcome variables were evaluated and entered into stepwise regression algorithms to determine independent variables. RESULTS We identified 252 procedures that were performed on 71 patients during the study interval. These procedures were performed under 173 general anesthesias. Using each general anesthesia as a case, induction instability was independently associated with stage of palliation before cavopulmonary shunt, case complexity, and preoperative use of angiotensin-converting enzyme inhibitor in a multivariate logistic regression. Maintenance instability was independently associated with case complexity and preoperative use of digoxin and inotropes. Among the 145 cases where the patient was not intubated before the procedure, postoperative need for mechanical ventilation was associated only with preoperative hospital LOS exceeding 14 days. Finally, the resulting linear regression model showed postoperative hospital LOS was independently associated with preoperative hospital LOS exceeding 14 days, presence of moderate ventricular dysfunction, preoperative use of an inotrope, and negatively associated with use of digoxin. CONCLUSIONS Within this population, we have identified independent risk factors for specific clinical outcomes. Patients before stage II palliation, undergoing more invasive procedures, and receiving inotropes, angiotensin-converting enzyme inhibitors, or digoxin appear to be at risk for intraoperative hemodynamic instability. Patients with preoperative hospital LOS exceeding 14 days appear to be at greater risk for requiring postoperative mechanical ventilation. Patients with preoperative LOS exceeding 14 days, with ventricular dysfunction, receiving inotropes, and not receiving digoxin appear to be at risk for protracted hospitalization. Application of these results should assist clinicians in assessing perioperative risk.