Silent brain infarcts in 755 consecutive patients with a first-ever supratentorial ischemic stroke. Relationship with index-stroke subtype, vascular risk factors, and mortality

Abstract

Background and purpose: We wanted to establish independent associations of various clinical variables, computed tomographic (CT) scan features, presenting stroke subtypes, and outcome with the presence of silent infarcts on CT.

Methods: We studied 755 consecutive patients in a prospective registration of patients with first-ever supratentorial atherothrombotic, cardioembolic, or lacunar stroke or stroke of undetermined cause by multiple logistic regression analysis.

Results: Two hundred six patients (27%) with a first symptomatic territorial or small deep ischemic stroke had one or more silent infarcts on CT. Of all silent lesions, 169 (82%) were small and deep. Silent infarcts were significantly more strongly associated with a lacunar than atherothrombotic (odds ratio [OR], 1.59; 95% confidence interval [CI], 1.02 to 2.47; P = .039) or cardioembolic (OR, 1.89; 95% CI, 1.2 to 2.99; P = .005) index stroke. Silent territorial lesions were more strongly associated with cardioembolic than with lacunar stroke but not with atherothrombotic stroke. In this respect, no differences were found between the atherothrombotic and undetermined-cause group. Advanced age and hypertension were the only risk factors that were significantly associated with silent infarcts (OR, 1.76; 95% CI, 1.14 to 2.71; P = .011; and OR, 1.58; 95% CI, 1.13 to 2.21; P = .007; respectively), mainly because of a strong independent association of these risk factors with silent small deep infarcts (OR, 1.75; 95% CI, 1.10 to 2.79; P = .018; and OR, 1.57; 95% CI, 1.09 to 2.24; P = .014; respectively). A cardioembolic source or atrial fibrillation in specific was not independently associated with any type or number of silent infarcts. Significant carotid stenosis (diameter reduction > 50%) was not significantly associated with any type of silent lesion. Initial severe handicap (Rankin Scale score > 3), 30-day case fatality rate, and 1-year mortality were not affected by the presence of silent infarcts.

Conclusions: The strong association of silent small deep lesions with first symptomatic small deep infarcts suggests a common underlying mechanism (presumably small-vessel vasculopathy), whereas cardiogenic embolism and large-vessel thromboembolism are the most likely causes in both silent and first symptomatic territorial infarcts. Single or multiple silent infarcts do not predict a cardioembolic stroke mechanism in first symptomatic supratentorial brain infarcts. As silent infarcts do not predict the cause of carotid embolic stroke in first symptomatic brain infarcts, their presence should not influence the decision on carotid surgery. Silent infarcts do not affect the degree of initial handicap, 30-day case fatality, or 1-year mortality. The significance of silent infarcts for predicting possible future cognitive decline and risk of recurrent stroke deserves further study.