Respective effects of early cerebral and abdominal magnetic resonance imaging on clinical decisions in infective endocarditis

Abstract

Aims: Extracardiac complications of endocarditis influence diagnosis, therapeutic plans, and prognosis. The aim of this study was to assess how early combined cerebral and abdominal magnetic resonance imaging (MRI) affects the diagnosis and management of adults with endocarditis.

Methods and results: In a single-centre prospective study, 58 patients with endocarditis underwent systematic cerebral and abdominal MRI within 7 days following admission. Diagnostic classification (Duke's modified criteria) and therapeutic plans were established by two experts just before and after MRI and then compared. Endocarditis was initially classified as definite in 29 patients, possible in 27, and excluded in 2. MRI detected cerebral lesions in 47 patients (81%) (ischaemic lesions in 25, microbleeds in 32, and silent aneurysms in 6), and abdominal lesions in 20 patients (34%). Based solely on MRI results without taking microbleeds into account, experts upgraded endocarditis diagnostic classification in 8 out of 29 (28%) non-definite endocarditis cases either to definite in 6 or to possible in 2. This upgrading was exclusively due to cerebral MRI in four patients and to cerebral and/or abdominal MRI in four patients. Experts modified endocarditis therapeutic plans in 11 (19%) out of the 58 patients, based solely on cerebral MRI, including modification of surgical plans in six (10%) patients. Overall, early MRI led experts to modify classification and/or therapeutic plans in 16 (28%) patients.

Conclusion: MRI identified cerebral and/or abdominal asymptomatic lesions in many patients with endocarditis, but more frequently cerebral. Both cerebral and abdominal MRI findings affected diagnosis, but only cerebral MRI affected clinical management plans. CLINICALTRIALS.GOV IDENTIFICATION NUMBER: NCT 00144885. IRB AUTHORIZATION NUMBER: 0511114 (Comité Protection des Personnes Paris-Ile de France 1).

Trial registration: ClinicalTrials.gov NCT00144885.