Impact of Opioid Epidemic on Infective Endocarditis Outcomes in the United States: From the National Readmission Database

Infective endocarditis (IE) is associated with marked morbidity and mortality in the United States and parallels the opioid pandemic. Few studies explore this interaction and its effect on clinical outcomes. We analyzed contemporary patients admitted with IE to determine predictors of readmission in the United States. The 2017 National Readmission Database was used to identify index admissions in adults with the diagnosis of IE, based on the International Classification of Disease, 10th Revision codes. The primary outcome of interest was 30-day readmission. Secondary outcomes were mortality, hospital charges, and predictors of hospitalization readmission. Of 40,413 index admissions for IE, 5,558 patients (13.8%) were readmitted within 30 days. Patients who were readmitted were younger (55 ± 20 vs 61 ± 19 years, p0.001) and more likely to have end-stage renal disease (12.2% vs 10.5%, p0.001), hepatitis C virus (19.4% vs 12.6%, p0.001), HIV (1.8% vs 1.2%, p = 0.001), opioid abuse (23.9% vs 15%, p0.001), cocaine use (7.3% vs 4.4%, p0.001), and other substance abuse (8.5 vs 5.6, p0.001). Patients readmitted were less likely to have diabetes mellitus (27.8% vs 29.4%, p = 0.01), hypertension (56.9% vs 64%, p0.001), heart failure (37.7% vs 40%, p0.001), chronic kidney disease (31.2% vs 32%, p0.001), and peripheral vascular disease (3.6% vs 4.6%, p = 0.001). The median cost of index admission for the total cohort was $84,325 (39,922 to 190,492). After adjusting for age, diabetes mellitus, heart failure, hypertension, and end-stage renal disease, opioid abuse (odds ratio [OR] 1.34; 95% confidence interval [CI] 1.23 to 1.46; p0.001), cocaine use (OR 1.32; 95% CI 1.17 to 1.48; p0.001), other substance abuse (OR 1.16; 95% CI 1.04 to 1.30; p = 0.008), and hepatitis C virus (OR 1.32; 95% CI 1.21 to 1.43; p0.001) correlated with higher odds of 30-day readmission. These factors may present targets for future intervention.