Comparison of infective endocarditis in patients with and without malignancy: A propensity score matched analysis.

Abstract

e24030 Background: Cancer patients are at a higher risk of infective endocarditis (IE) due to immunosuppressed state, indwelling catheters and recurrent infections. Due to the clinical scenario being relatively uncommon, there is lack of real-world data on prevalence, trends and outcomes of IE in cancer patients. Methods: We extracted data from the National Inpatient Sample (NIS) Database from 2016 to 2020. We included all admissions ≥18 years old with a principal diagnosis of IE with and without a cancer diagnosis using International Classification of Diseases Tenth Revision (ICD-10) codes. Propensity score matching was performed to balance hospital characteristics, patient demographics and comorbidities between the two groups. Univariate logistic regression was performed in the balanced cohorts. The primary outcome was inpatient mortality. Analysis was performed using STATA-MP, version 14.2. Results: A total of 70,225 admissions for IE were identified between 2016-2020, of which 2,294 (3.27%) had a history or active cancer. The annual number of admissions for IE did not show a significant change (p trend = 0.56); however, the proportion of IE patients with cancer doubled from 2.33% to 4.61% (p trend < 0.01) over the study period. After propensity score matching, there was no significant difference in in-hospital mortality (odds ratio [OR] 1.30, 95% CI 0.79-2.16, p = 0.31) between IE patients with and without cancer. Cohort with cancer had lower odds of experiencing respiratory failure (OR 0.52, 95% CI 0.33-0.83, p < 0.01) and stroke (OR 0.35, 95% CI 0.25-0.49, p < 0.01), although venous thromboembolism was more common in cohort with cancer (OR 3.03, 95% CI 1.76-5.21, p < 0.01). There was no difference in acute kidney injury or shock between the two groups. Cohort with cancer was less likely to undergo valve replacement surgery (OR 0.42, 95% CI 0.31-1.24, p < 0.01). The median length of stay (11 days vs 14 days, p < 0.01) and hospital charges ($70,698 vs $86,763, p < 0.01) were higher in the cohort without cancer. Conclusions: Underlying cancer was seen in growing proportion of admissions for IE over the study period. The pattern of complications was different between patients with and without cancer, which underscores the importance of cancer history in management of IE. Although there was no difference in mortality between patients with and without cancer, patients with cancer were less likely to undergo valve replacement surgery during the admission. Further research is needed to explore the specific factors influencing treatment decisions and long-term outcomes in cancer patients with IE.