Cerebral imaging in infectious endocarditis: A clinical study

Abstract

Background: Because neurological failure is the most frequent extra-cardiac complication in Infectious Endocarditis (IE), a brain computerised tomography (CT) scan is usually performed. The benefits of magnetic resonance imaging (MRI) have not been clearly established. This study aims to clarify the prevalence and type of cerebral lesions in IE detected using MRI and to compare them with those detected using CT scans. Methods: In the Grenoble University Hospital, patients diagnosed with definite or possible endocarditis according to Duke’s criteria were screened from 2010–2012. Brain CT and MRI were performed as soon as possible after diagnosis. Results: Of the 62 patients with IE who underwent at least one cerebral imaging within 3 weeks of diagnosis, Streptococcus (29) and Staphylococcus (14) were the main micro-organisms present. Twenty-eight (45%) patients underwent cardiac surgery. Eight (13%) died before discharge. Twenty (32%) had neurological symptoms. A brain CT-scan was performed on 53 (85%) patients and a MRI was performed on 43 (69%) patients. CT was pathological in 26 (49%) patients, whereas 32 (74%) MRI demonstrated abnormalities. The MRI lesions were classified as follows: ischaemia (48%), microbleeds (34%), haemorrhages (16%), abscesses (9%) and microbial aneurysms (4%). Of the 37 patients who underwent both MRI and CT examinations, ischaemia (48% vs 35%) and microbleeds (34%) demonstrated the difference between the two imaging methods. Conclusion: Through the early diagnosis of cerebral damage, even in asymptomatic cases, MRI may have a role in the IE management, influence any surgical decision and assist in prognosis assessment.

Keywords:

• Infectious endocarditis
• magnetic resonance imaging
• brain ischaemia
• cardiac surgical procedures
• cerebral microbleeds

Acknowledgements

We would like to thank Miss Mylène Maillet for her guidance and advice and the University Hospital of Grenoble for grants.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.