Potential of the Danish Comorbidity Index for Acute Myocardial Infarction to discriminate mortality in patients with infective endocarditis.

Introduction: Current risk score models for predicting mortality in infective endocarditis (IE) include data often unavailable in registries, limiting their use for confounding adjustment in population-based research.

Methods: This study assessed the Danish Comorbidity Index for Acute Myocardial Infarction (DANCAMI) for its ability to predict 30-day, 1-year, and 5-year mortality in IE patients, compared to the Charlson Comorbidity Index (CCI) and Elixhauser Comorbidity Index (ECI). The study included all adult Danish patients with first-time IE from 1995 to 2021. The area under the receiver operating characteristic curve (AUC) was estimated using logistic regression to measure discriminatory performance for all-cause and cardiovascular mortality at the specified time intervals. A baseline model included age and sex, while extended models incorporated continuous comorbidity scores.

Results: We identified 8966 patients with IE. Mortality rates were 12% at 30 days, 26% at 1 year, and 36% at 5 years. For all-cause mortality, AUCs for the baseline versus DANCAMI models were 0.64 vs. 0.69 at 30 days, 0.66 vs. 0.73 at 1 year, and 0.72 vs. 0.79 at 5 years. For cardiovascular mortality, AUCs for baseline versus DANCAMI models were 0.67 vs. 0.69 at 30 days, 0.67 vs. 0.69 at 1 year, and 0.70 vs. 0.71 at 5 years. CCI and ECI demonstrated comparable AUCs to the DANCAMI model.

Conclusion: DANCAMI improved discrimination of short- and long-term mortality in IE patients and may be used for confounder adjustment similarly to CCI and ECI.