Effect of thrombocytopenia in patients with infective endocarditis: an insight from the National Inpatient Sample database

Abstract

Abstract Introduction Infective endocarditis (IE) is one of the feared diseases in septic patients, and incidences are rising due to the intravenous drug abuse epidemic. Sepsis causes an escalation of the platelet destructions leading to thrombocytopenia (1). Few independent hospital-based studies have proposed increase mortality with thrombocytopenia in patients with IE (2–5). We aim to evaluate the significance of thrombocytopenia in IE subjects from the national inpatient sample (NIS) database. Method We analyzed the NIS database from Jan-2016 to Dec-2018 using Stata 16.0. NIS is the largest publicly available all-payer inpatient care database in the United States, containing data on more than seven million hospital stays per year. We identified patients with IE with or without thrombocytopenia using ICD-10 codes. The primary outcome of interest was in-hospital mortality comparison. We adjusted potential confounders (age, sex, diabetes, hypertension, etc.) with multivariate logistic regression analysis. Further analysis was done after balancing the population co-morbidity using a Greedy propensity match for accuracy. Results A total of 174,495 subjects were included in this study with a diagnosis of IE. Among these individuals, 33,285 patients had a concurrent diagnosis of thrombocytopenia. The mean ages were 53±19.5 years for the thrombocytopenia group and 55±19.8 years for others. Females were equally represented in both cohorts. There were 4,945 (14.86%) vs 2,835 (2.01%) mortalities reported in with and without thrombocytopenia group respectively. After propensity matching, there was a pronounced increase in mortality [Odds ratio (OR): 1.93 (1.72 – 2.15), p-value: <0.001] in the group with thrombocytopenia comparing to others. Complications such as Major bleeding requiring blood transfusion [OR: 1.45 (1.35–1.57)], acute myocardial infarction [OR: 1.56 (1.35–1.70)], complete heart block [OR: 1.44 (1.16–1.53)], cardiac arrest [OR: 1.44 (1.25–1.72)], acute respiratory failure [OR: 1.51 (1.39–1.73)] and pressor support requirement [(OR: 1.73 (1.57–2.01)] were notably higher in the cohort of thrombocytopenia with statistically significant p-value (<0.001). The difference in length of stay between both cohorts after propensity match wasn't statistically significant. Conclusion In conclusion, IE patients with thrombocytopenia have higher incidences of in-patient mortality and poor outcomes than cohort without thrombocytopenia. Some of the adverse consequences could be temporally explained by complications related to underlying thrombocytopenia. Further investigations are needed to delineate the outcome in this group of subjects. Funding Acknowledgement Type of funding sources: None.