Abstract Background Infective endocarditis (IE) remains a heterogeneous condition with high rates of adverse outcomes, and surgery is indicated in approximately 50% of IE. Risk scores play important roles in the decision-making for cardiac surgery, but few logistic scores have been developed and calibration evaluated for IE surgery, with variable performance. We aimed to develop and validate a novel risk score to predict operative mortality for IE surgery. Methods Consecutive patients in the prospective IE surgery registry at our clinic during 2010-2022 were studied (age 55.6±16.1 years, female 639 (30.0%), divided into the derivation (2010-2018) and validation (2019-2022) cohorts. Multivariable logistic regression was performed to develop a novel logistic risk score, and discriminative and calibration performances assessed for both cohorts. Results Operative mortality occurred in 5.1% (72/1401) and 4.1% (30/730) of the derivation and validation cohorts respectively. Novel risk score based on results of the multivariable logistic regression analysis for operative mortality are shown in the Figure, with key parameters of older age, critical pre-operative state, invasive IE, higher number of valves requiring surgery, history of diabetes, lower albumin, lower renal function and lower left ventricular ejection fraction. In the derivation and validation cohorts respectively, the novel score’s areas under curve (95% confidence intervals) for operative mortality were 0.80 (0.75-0.85) and 0.78 (0.69-0.87) respectively, and observed/expected ratios to assess calibration were 1.01 and 0.92 respectively, for the derivation and validation cohorts. Conclusion We developed a novel operative mortality logistic risk score for IE surgery, with moderate-to-high discrimination and good calibration. Further studies are necessary to externally validate and compare our score with other prior scores, and determine how our score can guide IE management to improve clinical outcomes.
Wang et al. (Sat,) studied this question.