Background/Purpose: Anti-neutrophil cytoplasmic antibodies (ANCAs), particularly those directed against proteinase 3 (anti-PR3) or myeloperoxydase (anti-MPO), are considered a highly specific hallmark of ANCA-associated vasculitis. Thus, false-positive ANCA tests have been reported in the context of drug exposure or infection, and several small case studies suggested that ANCAs can be detected in infective endocarditis (IE). This situation, combined with the multisystem protean presentation of IE, may lead to inappropriate diagnosis and therapy. Because the frequency of ANCAs in IE is unknown, we assessed the prevalence of ANCAs in a relatively large number of cases with IE.

Methods: The study was conducted in the framework of a prospective cohort study of consecutive cases with IE launched in 2005 in a single university hospital. Data on the principal demographic, clinical, laboratory (including serum creatinin and hematuria) and microbiological features were collected by a standardized questionnaire. Sera were stored for all patients who gave informed consent for blood sampling. Among the patients for whom sera were stored, we selected those with blood sampled not later than 30 days after the initiation of antibiotic therapy. All selected sera were tested for ANCAs in a central laboratory using indirect immunofluorescence (IIF) assays, and positive results were confirmed by ELISA. IIF involved ethanol-fixed neutrophils and categorized positive tests for C-ANCA (cytoplasmic pattern) or P-ANCA (perinuclear pattern). ELISA for anti-PR3 and anti-MPO specificities involved use of commercially available kits. In addition, we tested all sera for antinuclear antibodies by use of a commercially available indirect immunofluorescence test.

Results: Sera from 109 patients (81 [74%] men, mean age: 57.4 yrs [SD: 15.3]) were tested. All patients fulfilled Duke’s criteria for definite or probable endocarditis, and 33 (30%) had prosthetic valves. The major causative pathogens were Staphylococcus aureus (n = 30), Streptococci (n = 23), Streptococcus bovis (n = 10) and Enterococci (n = 7). C-ANCAs were found in 13 patients (12%), P-ANCAs in 11 (10%) and 1 case (1%) showed both patterns. ELISA revealed anti-PR3 in 4 cases (3%) and anti-MPO in 4 (3%), some with very high titers. The 8 anti-PR3/anti-MPO–positive IE cases involved various pathogens and both native and prosthetic valves. Testing for antinuclear antibodies was positive (titer > 1:160) in 17 (16%) cases.

Conclusion: This study is the first to assess the prevalence of ANCAs in IE and suggests that ANCAs, including those with anti-PR3 or anti-MPO specificities, occur in a significant subset of cases. These observations considerably substantiate the consideration of IE as a potential cause of false ANCA positivity.

ACR Meeting Abstracts - https://acrabstracts.org/abstract/prevalence-of-anti-neutrophil-cytoplasmic-antibodies-in-infective-endocarditis-an-analysis-of-109-cases/