A risk assessment scoring system predicts survival following the Norwood procedure

Abstract

No one set of characteristics has been consistently predictive of perioperative mortality and morbidity associated with the Norwood procedure. The purpose of the current study is to further validate a scoring system shown to be predictive of mortality following the Norwood procedure. We performed a retrospective review of all infants with the diagnosis of hypoplastic left heart syndrome (HLHS) who underwent the Norwood procedure at St. Louis Children's Hospital from July 1, 1994, to December 31, 2002. A weighted score for each of six factors comprised the scoring system. The factors included ventricular function, tricuspid regurgitation, ascending aortic diameter, atrial septal defect blood flow characteristics, blood type, and age. A score of > or = 7 points indicated lower reconstructive mortality risk, and a total score of < 7 points indicated a higher mortality risk. A total of 57 patients were analyzed. Twenty-five infants (44%) had a low risk score. These infants had a significantly greater survival at 48 hours compared to infants with a score of < 7 (92 vs 75%, p < 0.05). Infants with a high risk score had a significantly greater relative risk of mortality at 48 hours [OR = 2.04; confidence interval (CI) 1.04-4.00; p = 0.036]. The area under the receiver operating characteristic (ROC) curve is 0.8534 (95% CI, 0.78-0.922). This suggests that the scoring system has a very good degree of discriminatory power in selecting children who did not survive. Based on the results of the ROC, a cutoff score of >7 gives the best sensitivity and specificity for survival. When applied retrospectively, the survival outcomes predicted by our scoring system significantly correlated with actual outcomes. This supports the conclusion that a specific population of HLHS patients may have a higher mortality risk independent of surgical technique and postoperative care based on factors that can be assessed preoperatively.