What causes false clinical prediction of small deep infarcts?

Abstract

Our goal was to identify factors that play a role in false clinical diagnosis of small deep infarcts. In 350 prospectively registered patients with a first supratentorial ischemic stroke, we clinically differentiated between lacunar and nonlacunar syndromes. Using computed tomography (CT), we distinguished small deep and territorial infarcts and also recorded leukoaraiosis and asymptomatic infarcts. Degree of initial handicap, potential source of cardioembolic stroke, and hypertension were also noted. One hundred forty-seven patients had a lacunar and 203 a nonlacunar syndrome. Forty-two (12%) had a lesion visualized by CT that was compatible with a recent infarct but was considered inappropriate for the clinical syndrome: nineteen had a nonlacunar syndrome but a small deep infarct, and 23 had a lacunar syndrome but a territorial infarct. Patients with a nonlacunar syndrome but a small deep infarct were more severely disabled (a modified Rankin scale rating of 5) (odds ratio [OR], 4.31; 95% confidence interval [CI], 1.25 to 14.88) and had a cardioembolic source (OR, 4.07; 95% CI, 1.04 to 15.95), leukoaraiosis (OR, 3.79; 95% CI, 1.32 to 10.05), or asymptomatic infarcts visualized by CT (OR, 4.13; 95% CI, 1.45 to 11.71) compared with 124 patients with a correctly diagnosed small deep infarct. Twelve of 19 patients with a nonlacunar syndrome but a small deep infarct had a lesion in the left hemisphere, and 9 of these 12 had "aphasia." Patients with a lacunar syndrome but a territorial infarct more often had a cardioembolic source (OR, 4.02; 95% CI, 1.15 to 14.03) and a pure motor syndrome (OR, 4.52; 95% CI, 1.55 to 13.18) than those with lacunar syndrome but a small deep infarct, although 21 (91%) were in the right hemisphere. Of the first 103 patients with lacunar stroke diagnosed by two of the study neurologists, 5 had an inappropriate lesion compared with 14 of the later 40 diagnosed by colleagues without a specific interest in cerebrovascular diseases (OR, 0.09; 95% CI, 0.03 to 0.26). (1) Diagnosis of lacunar syndromes should not be influenced by deficit severity or the presence of a potential cardiac source of embolism. (2) Speech disorders should carefully be classified. (3) Routine tests of nondominant higher functions may be inadequate. (4) Doctors interested in cerebrovascular neurology have a lower failure rate in differentiating small deep infarcts from territorial infarcts than those less well-trained or interested in neurology. (5) Among the lacunar syndromes, pure motor syndrome may be the least specific predictor of a small deep infarct.