Infective endocarditis in the U.S., 1998-2009: a nationwide study

PLoS One. 2013;8(3):e60033. doi: 10.1371/journal.pone.0060033. Epub 2013 Mar 20.

Abstract

Background: Previous studies based on local case series estimated the annual incidence of endocarditis in the U.S. at about 4 per 100,000 population. Small-scale studies elsewhere have reported similar incidence rates. However, no nationally-representative population-based studies have verified these estimates.

Methods and findings: Using the 1998-2009 Nationwide Inpatient Sample, which provides diagnoses from about 8 million U.S. hospitalizations annually, we examined endocarditis hospitalizations, bacteriology, co-morbidities, outcomes and costs. Hospital admissions for endocarditis rose from 25,511 in 1998 to 38, 976 in 2009 (12.7 per 100,000 population in 2009). The age-adjusted endocarditis admission rate increased 2.4% annually. The proportion of patients with intra-cardiac devices rose from 13.3% to 18.9%, while the share with drug use and/or HIV fell. Mortality remained stable at about 14.5%, as did cardiac valve replacement (9.6%). Other serious complications increased; 13.3% of patients in 2009 suffered a stroke or CNS infection, and 5.5% suffered myocardial infarction. Amongst cases with identified pathogens, Staphylococcus aureus was the most common, increasing from 37.6% in 1998 to 49.3% in 2009, 53.3% of which were MRSA. Streptococci were mentioned in 24.7% of cases, gram-negatives in 5.6% and Candida species in 1.0%. We detected no inflection in hospitalization rates after changes in prophylaxis recommendations in 2007. Mean age rose from 58.6 to 60.8 years; elderly patients suffered higher rates of myocardial infarction and death, but slightly lower rates of Staphylococcus aureus infections and neurologic complications. Our study relied on clinically diagnosed cases of endocarditis that may not meet strict criteria. Moreover, since some patients are discharged and readmitted during a single episode of endocarditis, our hospitalization figures probably slightly overstate the true incidence of this illness.

Conclusions: Endocarditis is more common in the U.S. than previously believed, and is steadily increasing. Preventive efforts should focus on device-associated and health-care-associated infections.

MeSH terms

  • Age Factors
  • Endocarditis / epidemiology*
  • Endocarditis / microbiology*
  • Hospitalization / statistics & numerical data
  • Humans
  • Incidence
  • Linear Models
  • Staphylococcus aureus*
  • United States / epidemiology

Grant support

The authors have no support or funding to report.