The Clinical Significance of Cerebral Microbleeds in Infective Endocarditis Patients

Abstract

We assessed the clinical features of cerebral microbleeds (CMBs) and their association with clinical outcomes in active infective endocarditis patients. From January 2009 to June 2015, 132 active IE patients diagnosed per the modified Duke's criteria were retrospectively reviewed. Brain magnetic resonance imaging was performed in 102 patients, and 74 patients whose image data were available to assess CMBs were enrolled. CMBs were defined as hypointense lesion <10 mm in diameter, seen on T2* or susceptibility-weighted imaging. Forty patients had CMB and 34 did not. Patients with CMB were older, and the proportion of prior antiplatelet therapy, staphylococcal infection, and prosthetic valve endocarditis were higher than in patients without CMB. Surgery was performed in 25 (63%) patients with CMB and 24 (71%) patients without CMB. There was no significant difference in the de novo stroke incidence postoperatively (16% vs 17%, P = 0.95). Although all-cause mortality rate tended to be higher in patients with CMB, there were no significant differences in the in-hospital mortality rate and estimated 1-year major adverse event rate between the 2 groups (13% vs 12%, P = 0.92; 20% vs 19%, P = 0.35). Cox regression analysis adjusting age and operative risk did not show that CMB was a significant risk factor for all-cause death and major adverse event. Patients with CMB were older than those without, and microbleeds were associated with antiplatelet therapy, staphylococcal infection, and prosthetic valve endocarditis. However, the mid-term clinical outcomes of patients with CMB and those without were comparable.