Contemporary Trends in Native Valve Infective Endocarditis in United States (from the National Inpatient Sample Database)

Abstract

Infective Endocarditis (IE) is associated with high mortality and morbidity. The data on contemporary trends and health care utilization remain scarce for IE. Consequently, we used the National Inpatient Sample database from 2002 to 2016 to study burden of IE. Risk-adjusted rates were calculated using an Analysis of Covariance with the Generalized Linear Model. Trends were assessed with linear regression and Pearson's Chi-square modeling, where appropriate. Binomial logistic regression was used for computing predictors of in-hospital mortality. We identified 523,432 hospitalizations for native valve IE. Risk-adjusted mortality decreased from 16.7% in 2002 to 9.7% in 2016 (p <0.01). The risk-adjusted length of stay decreased from 17.4 days in 2002 to 13.4 days in 2016 (p <0.01). Mean cost of stay adjusted for risk factors and inflation increased from 112,702$ in 2002 to 164,767$ in 2016 (p <0.01). Valve replacement increased from 10.2% in 2002 in to 13.4% in 2016, (p <0.01). Independent predictors of mortality included age (OR, 1.02 [1.02 to 1.020], p <0.01), female gender (OR, 1.07 [1.05 to 1.09], p <0.01), Blacks (OR, 1.28 [1.24 to 1.31], p <0.01), Hispanics (OR, 1.15 [1.11 to 1.19], p <0.01) and patients with co-morbid conditions like congestive heart failure (OR, 1.78 [1.74 to 1.82], p <0.01), renal failure (OR, [1.69 [1.65 to 1.73], p <0.01) and weight loss (OR, 1.40 [1.36 to 1.43], p <0.01). In summary, in-hospital mortality from native valve IE has been decreasing but total hospitalization and average cost of stay has increased.