Predictors of Prolonged Hospital Length of Stay Following Stage II Palliation of Hypoplastic Left Heart Syndrome (and Variants): Analysis of the National Pediatric Cardiology Quality Improvement Collaborative (NPC-QIC) Database

The objective of this study is to identify predictors of prolonged hospital length of stay (LOS) for single ventricle patients following stage 2 palliation (S2P), excluding patients who underwent a hybrid procedure. We explore the impact of demographic features, stage 1 palliation (S1P), interstage I (IS1) management, S2P, and post-surgical care on hospital LOS following S2P. We conducted a retrospective analysis of the National Pediatric Cardiology Quality Improvement Collaborative (NPC-QIC) database. The NPC-QIC database is an established registry of patients with hypoplastic left heart syndrome (HLHS) and its variants. It contains detailed information regarding the demographic features, S1P, IS1, S2P, and interstage 2 (IS2) management of children with HLHS and related single ventricle cardiac malformations. Between 2008 and 2012, there were 477 participants with recorded LOS data in the NPC-QIC registry. Excluding the 29 patients who underwent hybrid procedure, there were 448 participants who underwent a Norwood (or Norwood-variant procedure) as S1P. In order to be included in the NPC-QIC database, participants were discharged to home following S1P and prior to S2P. We found that postoperative LOS among the 448 S2P procedure recipients is most strongly influenced by the need for reoperation following S2P, the need for an additional cardiac catheterization procedure following S2P, the use of non-oral methods of nutrition (e.g., nasogastric tube, total parental nutrition, gastrostomy tube), and the development of postoperative complications. Factors such as age at the time of S2P, the presence of a major non-cardiac anomaly, site participant volume, IS1 course, the type and number of vasoactive agents used following S2P, and the need for more than 1 intensive care unit (ICU) hospitalization (following discharge to the ward but prior to discharge to home) were significant predictors by univariate analysis but not by multivariate analysis. We excluded participants undergoing the hybrid procedure as S1P from this analysis given that the S2P following the initial hybrid is typically a more complicated procedure. Hospital LOS following S2P among children undergoing the Norwood or Norwood-variant procedure as S1P is most strongly influenced by events following S2P and not demographic or S1P factors. Factors most predictive of prolonged LOS include the need for reoperation, the need for an additional cardiac catheterization procedure following S2P, the need for non-oral methods of nutrition, and the development of postoperative complications.