Real-time complication monitoring in pediatric cardiac surgery

Background: As overall mortality rates have fallen in pediatric cardiac surgical procedures, complication monitoring is becoming an increasingly important metric of patient outcome. Currently there is no standardized method available to monitor severity-adjusted complications in congenital cardiac surgical procedures. Methods: Complications associated with pediatric cardiac surgical procedures were prospectively collected from consecutive cases in a single pediatric cardiac surgical unit from October 1, 2009 to September 31, 2011. Complications were accounted for by frequency and severity and then stratified by surgical complexity, using the Risk Adjustment for Congenital Heart Surgery (RACHS) method, giving an average morbidity burden per RACHS category. "Expected" morbidity burden for each RACHS category was derived from year 1 (2009-2010) data. Observed minus expected (O:E) plots were then generated for the entire series of complications from year 2 (2010-2011) data. Separate O:E plots were also created for 5 complication classes and monitored for increases. Results: There were 181 index surgical procedures performed in 178 patients. Two hundred and seventeen complications occurred in 80 procedures. The frequency and severity of complications increased with surgical complexity. The overall O:E plot was flagged twice for unanticipated increases in severity-adjusted complications. When the class-specific O:E plots were monitored for increases, the overall flags were found to originate from increased rates of infections and cardiac/operative complications. Conclusions: The O:E plot provides a simple and effective system to monitor complication rates over time based on severity-adjusted complication data. Grouping complications into classes allows us to identify specific subsets of complications that can be focused on to improve patient outcomes.