Incidence, Microbiology, and Outcomes in Patients Hospitalized With Infective Endocarditis

Abstract

Background: Despite improvements in management, infective endocarditis remains associated with high mortality and morbidity. We describe temporal changes in the incidence, microbiology, and outcomes of infective endocarditis and the effect of changes in national antibiotic prophylaxis guidelines on incident infective endocarditis.

Methods: Using a Scotland-wide, individual-level linkage approach, all patients hospitalized with infective endocarditis from 1990 to 2014 were identified and linked to national microbiology, prescribing, and morbidity and mortality datasets. Linked data were used to evaluate trends in the crude and age- and sex-adjusted incidence and outcomes of infective endocarditis hospitalizations. From 2008, microbiology data and associated outcomes adjusted for patient demographics and comorbidity were also analyzed. An interrupted time series analysis was performed to evaluate incidence before and after changes to national antibiotic prophylaxis guidelines.

Results: There were 7638 hospitalizations (65±17 years, 51% females) with infective endocarditis. The estimated crude hospitalization rate increased from 5.3/100 000 (95% CI, 4.8-5.9) to 8.6/100 000 (95% CI, 8.1-9.1) between 1990 and 1995 but remained stable thereafter. There was no change in crude incidence following the 2008 change in antibiotic prophylaxis guidelines (relative risk of change 1.06 [95% CI, 0.94-1.20]). The incidence rate in patients >80 years of age doubled from 1990 to 2014 (17.7/100 000 [95% CI, 13.4-23.3] to 37.9/100 000 [95% CI, 31.5-45.5]). The predicted 1-year age- and comorbidity-adjusted case fatality rate for a 65-year-old patient decreased in women (27.3% [95% CI, 24.6-30.2] to 23.7% [95% CI, 21.1-26.6]) and men (30.7% [95% CI, 27.7-33.8] to 26.8% [95% CI, 24.0-29.7]) from 1990 to 2014. Blood culture data were available from 2008 (n=2267/7638, 30%), with positive blood cultures recorded in 42% (950/2267). Staphylococcus (403/950, 42.4%) and streptococcus (337/950, 35.5%) species were most common. Staphylococcus aureus and enterococcus had the highest 1-year mortality (adjusted odds ratio 4.34 [95% CI, 3.12-6.05] and 3.41 [95% CI, 2.04-5.70], respectively).

Conclusions: Despite changes in antibiotic prophylaxis guidelines, the crude incidence of infective endocarditis has remained stable. However, the incidence rate has doubled in the elderly. Positive blood cultures were observed in less than half of patients, with Staphylococcus aureus and enterococcus bacteremia associated with worse outcomes.

Keywords: antibiotic; antibiotic prophylaxis; incidence; infective endocarditis; mortality; outcomes.

Figure 1.

Estimated incidence rate per 100 000 in the population (A) and stratified by age groups (B). Blue circles in A represent the absolute crude rates with the size of the circles proportional to the absolute count. The solid blue line represents the estimated incident rate from generalized additive modeling using the Poisson distribution. The dashed blue lines represent the corresponding upper and lower 95% CI bounds.

Figure 2.

Plot showing the observed incident rates per 100 000 by year of hospitalization (black dots), in relation to the introduction of national antibiotic prophylaxis guidelines. The shaded gray box indicates introduction of National Institute of Health and Care Excellence antibiotic prophylaxis guidelines. The black line shows the predicted incident rate using the model described in Text VI in the Data Supplement, incorporating the change in guidelines from 2008 onward. The overlying red line shows the predicted incident rate assuming the counterfactual of no change in antibiotic prophylaxis guidelines in 2008.

Figure 3.

Forest plot showing odds ratio from logistic regression evaluating the association between all-cause mortality at 1 year and patient demographics, comorbidity, and microbiology.