By Michael H. Crawford, MD

Professor of Medicine, Chief of Clinical Cardiology, University of California, San Francisco

Dr. Crawford reports no financial relationships relevant to this field of study.

SYNOPSIS: The authors of a population-wide study of hospitalizations and deaths from infective endocarditis (IE) in England confirmed the high risk of IE in certain cardiac conditions, but showed that other conditions thought to be low risk also are at higher risk and found new higher-risk categories not previously identified.

SOURCES: Thornhill MH, Jones S, Prendergast B, et al. Quantifying infective endocarditis risk in patients with predisposing cardiac conditions. Eur Heart J 2018;39:586-595.

Sun YP, O’Gara PT. Cardiovascular conditions predisposing to infective endocarditis: Time to reconsider the current risk classification system? Eur Heart J 2018;39:596-598.

Data on the relative risks of developing infective endocarditis (IE) or dying from IE with different predisposing cardiac conditions in a large population cohort are lacking. Thus, investigators from England surveyed all patients admitted to English hospitals between 2000 and 2013 with a condition associated with an increased risk of IE and followed them for five years to assess subsequent admissions for IE. This information was compared to a reference group of the entire population of England (> 51 million).

The incidence of IE in the whole English population was 36.2 cases/million/year with an admission-related mortality of 6.3 cases/million/year. The incidence was highest in those with a previous history of IE and prosthetic valves or repaired valves (14,359, 4,637, and 4,710 cases/million/year, respectively). IE admission deaths also were highest in these groups (2,940, 1,092, and 907 cases/million/year, respectively). Admissions for IE and subsequent deaths were high in patients with congenital heart conditions (CHC) with a shunt or conduit, but lower in those with unrepaired cyanotic CHC and considerably lower in those post repair using prosthetic material. In fact, the incidence of IE or death on admission for IE was higher in those with rheumatic fever or nonrheumatic valve disease compared to those with cyanotic CHC repaired with prosthetic material.

Among conditions in which the risk of IE has not been characterized, hypertrophic cardiomyopathy had a significantly higher risk of IE than the controls (odds ratio [OR], 33; 95% confidence interval [CI], 23-45; P < 0.0001) but an insignificant risk of IE death (OR, 4; 95% CI, 0.2-18; P = 0.17). However, implanted electrophysiology (EP) devices had a relatively high incidence of IE (OR, 10; 95% CI, 9-11; P < 0.0001) and a higher risk of IE death (OR, 10; 95% CI, 9-12; P < 0.001) compared to controls.

The authors concluded that some conditions considered at moderate risk of IE and not candidates for antibiotic prophylaxis should be re-evaluated.


After consideration of the lack of quality data and with pressure from the dental industry, the guidelines were changed in 2007 to state that the use of prophylactic antibiotics for traumatic procedures in areas of the body not amenable to sterilization should be restricted to a few very high-risk conditions. Subsequent studies on the effect of these changes on the incidence of IE have shown conflicting results. This study was designed to use the National Health Service data from England to establish the incidence of IE and hospital death from IE in the English population from 2000 until 2013 with a minimum five-year follow-up and to relate these data to the patients’ cardiac conditions that put them at risk for IE.

The results confirm the high-risk status of prior IE, valve surgery (replacement, repair), or CHC with shunts or conduits, but not unrepaired cyanotic CHC and repaired CHC with prosthetic material. There were too few heart transplant patients to accurately assess IE incidence (considered high risk if they have valve regurgitation). Unrepaired cyanotic CHC had a risk of IE similar to rheumatic fever, non-rheumatic valve disease, and congenital valve abnormalities (e.g., bicuspid valve). Repaired CHC with prosthetic material had a lower risk than these valve diseases. The surprise was the moderate risk of IE in hypertrophic cardiomyopathy (HCM). However, many patients with HCM have mitral valve regurgitation. Not surprising was the moderate risk of IE with implanted EP devices.

There were limitations to this study. The authors used ICD-10 codes to categorize the patients rather than the actual medical records. There are no data on comorbidities, the infecting organisms, therapy, or surgery. Also, nonrheumatic valve disease is a broad category that could have subgroups with different results. Finally, the authors only examined hospital admissions and deaths, so the incidence estimates probably are underestimated.

There always has been a proviso that the antibiotic prophylaxis guidelines were just that, and clinical judgment regarding individual patients should be employed. These data from England will help inform which patient considerations outside the restrictive guidelines would be appropriate.

Clearly, high-risk patients (OR, > 76) are those with prior IE, valve replacement/repair, and CHC with shunts/conduits. Intermediate-risk patients (OR, 10-66) would include (in decreasing order by OR): congenital valve abnormalities, unrepaired cyanotic CHC, rheumatic fever, nonrheumatic valve disease, HCM, CHC repaired with prosthetic material, and EP devices.

Whether the next edition of the antibiotic prophylaxis guidelines will alter the recommendations based on these data is unknown, but clinicians should be aware that many believe we have been too restrictive and the patient categories requiring prophylaxis should be revised.