The Barts Surgical Infection Risk tool (B-SIR): external validation and comparison with existing tools to predict surgical site infection after cardiac surgery

Abstract

ObjectiveWe previously developed and internally validated the Barts Surgical Infection Risk (B-SIR). We sought to explore the external validity of the B-SIR tool and compare with the Australian Clinical Risk Index (ACRI) and Brompton and Harefield Infection Score (BHIS). Study design and settingThis multicentre retrospective analysis of prospectively collected local data included adult (≥18years) patients undergoing cardiac surgery between January 2018 and December 2019. Pre-pandemic data was used as a reflection of standard practice. Area under the curve (AUC) was used to validate and compare the predictive power of the scores and calibration was assessed using Hosmer-Lemeshow test and calibration plots. ResultsFrom three centres, 6,022 patients were included in the complete case analysis. The mean age was 66 years, 75% were men and 3.19% developed SSI. The B-SIR has an AUC of 0.686 (95% CI: 0.649 to 0.723) similar to the developmental study (AUC=0.682; 95% CI: 0.652 to 0.713). This is significantly higher than BHIS AUC=0.610 (95% CI: 0.045 to 0.109; p<0.001) and ACRI AUC=0.614 (95% CI: 0.041 to 0.103; p<0.001). After re-calibration using a correction factor, the B-SIR model gave accurate risk predictions (Hosmer-Lemeshow test p=0.423). Multiple imputation result (AUC=0.676; 95%CI: 0.639 to 0.712) is similar to development data and is higher than ACRI and BHIS. ConclusionExternal B-SIR validation indicates B-SIR predicts SSI after cardiac surgery better than ACRI and BHIS risk tools. This suggests B-SIR could be useful to use routinely in practice.