The German Registry of Acute Aortic Dissection Type A score for 30-day mortality prediction in Type A Acute Aortic Dissection surgery: a systematic review and meta-analysis

OBJECTIVES: The German Registry of Acute Aortic Dissection Type A (GERAADA) score is a risk score for predicting 30-day mortality after an operation for type A acute aortic dissection (TAAAD). This meta-analysis sought to assess the performance of the GERAADA model and compare it to the performance of the European System for Cardiac Operative Risk Evaluation II (EuroSCORE II). METHODS: A systematic search of 3 online databases was conducted to identify studies that externally validated the GERAADA score. A random-effect meta-analysis was conducted, pooling area under the curve (AUC) data, operative mortality observed/expected (O/E) ratios and observed-expected (O-E) differences-of the GERAADA model in all studies and of the EuroSCORE II when available. RESULTS: Eleven studies were selected, including a total of 10 360 patients. The observed in-hospital mortality rate was 12.2%. Pooled expected mortality rates estimated by the GERAADA score and the EuroSCORE II were 18.4% and 5.8%, respectively. The pooled analyses for the GERAADA scores showed moderate discrimination [AUC 0.70, 95% confidence interval (CI) 0.66-0.73] and good calibration [observed-expected (O-E) differences -12.3, 95% CI -27.1 to 2.58; O/E ratio 0.81, 95% CI 0.57-1.05]. Results from 5 studies (2133 patients) investigating both scores simultaneously revealed similar AUC results (P = 0.50), significantly lower O-E differences (P = 0.03) and a trend towards O/E ratios closer to 1 (P = 0.08) with the GERAADA score compared to the EuroSCORE II. CONCLUSIONS: The GERAADA score seemed to offer a better calibration for predicting 30-day postoperative death following TAAAD operations, even though further studies are needed to confirm these findings. The moderate discriminatory capacity of both scores highlights the challenges of predicting outcomes in complex cardiovascular conditions like TAAAD.