Clinical Characteristics and Outcomes of Patients With Cutibacterium acnes Endocarditis

Key Points Question What are the clinical characteristics and outcomes of patients with Cutibacterium acnes endocarditis? Findings In this case series including 105 patients, C acnes endocarditis was seen predominantly among male patients with prosthetic heart valves and was characterized by absence of fever, close to normal inflammatory markers, and a prolonged time to positive blood culture results. Surgical treatment was required in a high proportion of patients. Meaning This study suggests that C acnes endocarditis is difficult to diagnose because typical endocarditis features are often absent; it is recommended to extend blood culturing for male patients with new-onset prosthetic valve dysfunction, peripheral embolization, or signs of heart failure.


Introduction
Cutibacterium acnes is a gram-positive bacterium, most known for its role in acne vulgaris. C acnes is generally considered nonpathogenic, and positive blood culture results are often regarded as contaminated. 1 In recent years, more attention has been paid to C acnes as a cause of infective endocarditis (IE). Experienced clinicians hypothesize that patients with C acnes endocarditis are difficult to diagnose because typical IE features, such as fever and high levels of inflammatory markers, are often absent. However, these statements are supported only by case reports or small case series. 2,3 Therefore, the aim of this study is to examine the clinical characteristics and outcomes of patients with C acnes IE in a large cohort of more than 100 patients.

Methods
We performed a retrospective case series in 7 hospitals in the Netherlands and France (4 university hospitals and 3 teaching hospitals). Patients with a diagnosis of definite IE according to the modified Duke criteria 4 between January 1, 2010, and December 31, 2020, were included. Cases with an isolated infected pacemaker or internal cardioverter defibrillator lead cases were excluded because we were interested only in valve endocarditis. Cases were identified by positive blood or valve and prosthesis culture results, retrieved from the medical microbiology databases. The Medical Ethics Review Committee Leiden The Hague Delft waived the need for consent because of the retrospective nature of this study in which participants are not subjected to any form of action. In addition, data were deidentified. Recurrence of IE was defined as IE relapse and/or a new episode of C acnes IE after initial treatment. Data are presented following the reporting guideline for case series. 5  Hemoglobin level, median (IQR), g/dL 7.6 (6.  These patients had left-sided PVE with small vegetations. Mortality rates as well as C acnes recurrence rates are presented in Table 5.

Statistical Analysis
All but 1 patient received empirical and, when susceptibility results were available, targeted antibiotic therapy after diagnosis as advised by the European Society of Cardiology guideline. 7 The latter patient had a severe stroke at presentation with an unfavorable prognosis.   antibiotic therapy because of patient preference and a chronically infected ascending aorta prosthesis. Clindamycin and amoxicillin, respectively, were given.

Discussion
To our knowledge, this study is the world's largest case series of C acnes IE. It confirms the hypothesis that the clinical presentation of patients with C acnes IE differs significantly from that of other bacterial causative agents. In our study, C acnes IE was characterized by absence of fever (66.7%) and close to normal CRP levels. In contrast, in the EURO-ENDO (European Endocarditis) registry, 77% of patients with IE had fever, and the median CRP level was 6.5 mg/dL. 8 Aside from the diagnostic challenges inherent to PVE, 9 the diagnostic process in C acnes IE is further impaired by the prolonged time to positive blood culture results. So, for male patients with prosthetic heart valves presenting with signs of heart failure or peripheral embolization, there should be a low threshold for performing blood cultures (>5 days' incubation period). Echocardiography remains the criterion standard in detecting IE, but additional cardiac CT or 18 FDG PET-CT is useful in suspected PVE with periannular  We hypothesize that C acnes IE occurs mostly among male patients because the sebaceous glands, where C acnes predominates, 11 proliferate under stimulation of androgens. 12 Because the male chest is rich with sebaceous glands, we suggest the surgical wound is contaminated with C acnes during index (valve) surgery. Subsequently, we assume the low virulence and slow growth of C acnes 13 results in patients presenting several years after index surgery. We hypothesize that additional preoperative measures, such as topical agents, can be effective in C acnes IE prevention for patients with acne on the thorax. However, further research is needed on this topic.
Although levels of inflammatory markers are generally low at presentation, the high share of aortic root abscesses emphasizes the indolent but not innocent nature of C acnes IE. If surgery is indicated, 7 it should be offered to patients because mortality is high when it is indicated but not performed. 14

Limitations
Our study has several limitations. First, the retrospective design of our study creates the risk of information bias. Although we comprehensively assessed all patients' records to and from referring hospitals, we cannot rule out possible inaccuracies in the provided data. Second, because we opted not to perform statistical analysis because of the small sample size, we cannot provide the best antibiotic regimen (with or without adding rifampicin) to treat C acnes IE and to minimize the risk of IE recurrence. Also, although we observed a trend of higher C acnes IE recurrence rates after conservative treatment, we must look at this finding in the light of a lack of underlying statistical evidence. Third, C acnes is a known skin contaminant, so blood or tissue culture results could have been false positive. Nevertheless, all included cases had multiple positive blood or tissue culture results, combined with positive echocardiography findings and/or surgical evidence, which means that all cases were definite IE by C acnes.

Conclusions
In this case series, we found that C acnes endocarditis was seen predominantly among male patients with prosthetic heart valves. Diagnosing C acnes endocarditis is difficult due to its atypical presentation, with frequent absence of fever and inflammatory markers. The prolonged time to positivity of blood culture results further delays the diagnostic process. Not performing surgery when indicated seems to be associated with higher mortality rates. However, each patient's operability must always be taken into account. For patients with PVE with small vegetations, there should be a low threshold for surgery because these patients seem to be prone to short-term endocarditis relapse.