By Michael H. Crawford, MD, Editor

SYNOPSIS: A large, prospective observational study of echocardiography in all patients with Enterococcus faecalis bacteremia showed a high incidence of infective endocarditis.

SOURCES: Dahl A, Iversen K, Tonder N, et al. Prevalence of infective endocarditis in Enterococcus faecalis bacteremia. J Am Coll Cardiol 2019;74:193-201.

Fernández-Hidalgo N, Escolà-Vergé L. Enterococcus faecalis bacteremia: Consider an echocardiography, but consult an infectious diseases specialist. J Am Coll Cardiol 2019;74:202-204.

The authors of previous studies have found that Enterococcus faecalis (Ef) is the third most common organism causing infective endocarditis (IE) and the most frequent cause of IE after transcatheter aortic valve replacement (TAVR). However, little information exists concerning the prevalence of IE with Ef bacteremia. Investigators from Denmark performed echocardiograms on all patients with Ef bacteremia in 10 community hospitals in Denmark serving about 1.5 million people. Only patients who died before blood culture results were available or were terminally ill were excluded. A multidisciplinary team considering all clinical data, including PET (when indicated), made the diagnosis of IE. The portal of entry of Ef was determined using international standards. For this study, patients were entered for three years (2014-2016). Follow-up was 100%, and all patients were followed for a minimum of one year. Ef bacteremia was diagnosed in 421 patients; after exclusions, 344 patients underwent echocardiography, including transesophageal echo (TEE) in 74% of patients.

Definite IE was diagnosed in 90 patients, with a mean age of 74 years. The Charlson Comorbidity Index (CCI) was similar in those with and without IE. Community-acquired infection was more common in those with IE (44%), whereas hospital-acquired infection was more common in those with bacteremia alone (63%). Patients with IE had monomicrobial Ef bacteremia more often (90%) and ≥ 3 positive blood culture bottles (84%). Echocardiography showed vegetations in 83% of those diagnosed with IE, and TEE was superior for detecting vegetations (positive when negative by transthoracic echo in 47%).

In addition to community acquisition, monomicrobial Ef, and ≥ 3 positive blood cultures, IE was also more common with unknown origin of infection, prosthetic heart valves, and immunosuppression. Cardiac surgery was performed in 20% of IE cases, and 10% had devices removed. The authors concluded that the high prevalence of IE in Ef bacteremia patients suggests that in all such patients, echocardiography should be strongly considered.


The main finding of this study was that the prevalence of IE in Ef bacteremia is higher (26%) than that reported in prior retrospective and case-controlled studies (5-13%). A major difference between this new study and older ones was the deployment of echocardiography in all Ef bacteremia patients. The authors of prior studies used echocardiography at much lower rates, especially TEE (12%) compared to 74% in this new study. In some ways, the universal use of echocardiography would be expected to increase the frequency of IE diagnoses. One of the clinical considerations used in this study was the modified Duke criteria. The Duke criteria were first proposed in 1994 and were modified in 2000. Subsequent studies have shown that the criteria are heavily dependent on echocardiography findings, such that echocardiography has become the cornerstone of IE diagnosis. Thus, studies not relying on echocardiography to diagnose IE would be expected to have lower prevalence rates.

The key questions raised by this study: Is echocardiography required in all patients with Ef bacteremia? If so, should it be TEE? Studies of the more common Staphylococcus bacteremia, which has similar rates of IE (about 25%), have shown the power of echocardiography for diagnosis, but most clinicians do not recommend echocardiography for all cases of hospital-acquired Staphylococcus sepsis. If they do recommend echocardiography, not all patients undergo TEE. As the authors of an accompanying editorial noted, considerable clinical judgment is required in patients with bacteremia. First, echocardiography should be conducted only if diagnosing IE would alter management. Some of these hospitalized patients already are taking antibiotics and are so sick that further interventions are out of the question.

The results of the Dahl et al study suggest an approach based on the six risk factors for IE they identified. Low-risk patients with none of these factors had a 3% risk of IE. In those with one or two risk factors, IE risk was 14%. The risk was 56% in those with three to six risk factors. One could hold off on echocardiography in the low-risk group, start with transthoracic echocardiography in the moderate-risk group, and go right to TEE in the high-risk group. Of course, such a schema would have to be tested in prospective randomized trials to be highly confident of its utility, but the concept is attractive.

Two major strengths of this study were the authors’ use of an expert multidisciplinary team to diagnose IE and the fact that possible IE was not counted. A potential weakness of this study was that it included a Scandinavian population. It is unknown if the results of this study would be applicable to other populations. Also, this study was conducted in community hospitals. The results could be different in tertiary referral hospitals.

Finally, five of the risk factors Dahl and colleagues identified were well known already (prosthetic heart valve, community-acquired infection, unknown origin, ≥ 3 positive blood cultures, and monomicrobial Ef). However, the authors added one more useful risk factor: immunocompromised patients.