Premium
Clinical‐Sonographic Index (CSI): A Novel Transcranial Doppler Diagnostic Model for Middle Cerebral Artery Stenosis
Author(s) -
Jung KeunHwa,
Lee YongSeok
Publication year - 2008
Publication title -
journal of neuroimaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.822
H-Index - 64
eISSN - 1552-6569
pISSN - 1051-2284
DOI - 10.1111/j.1552-6569.2007.00181.x
Subject(s) - medicine , transcranial doppler , stenosis , odds ratio , confidence interval , middle cerebral artery , logistic regression , magnetic resonance imaging , radiology , magnetic resonance angiography , cardiology , nuclear medicine , ischemia
BACKGROUND Transcranial Doppler sonography is useful for the diagnosis of middle cerebral artery (MCA) stenosis. Although the previous studies have focused on the elevated mean flow velocity (MFV) or asymmetry of MFV, the lack of clinical correlation might limit diagnostic accuracy. We try to develop and validate a new diagnostic model including more comprehensive clinical and sonographic parameters.METHODS Consecutive patients with magnetic resonance angiography (MRA)‐verified MCA stenosis were included, and compared with control subjects with normal MCA. The age, sex, corresponding symptoms (CS) to sonographic side, diabetes mellitus (DM), and hypertension were included for analysis. As sonographic parameters, MFV (cm/sec), asymmetry index (AI,%), and difference of pulsatility index (ΔPI) were analyzed. Clinical‐sonographic index (CSI) model was built with significant parameters by multivariate logistic regression analysis.RESULTS One hundred and seven patients (M:F = 53:54, age: 61.6 ± 11.6 years), and 100 control subjects ( M:F = 49:51 , age: 54.9 ± 14.5 years) were included. In logistic regression, MFV (odds ratio [OR], 1.057; 95% confidence interval [95% CI], 1.030–1.084), AI (OR, 1.067; 95% CI, 1.031–1.104), ΔPI (OR, 41.754; 95% CI, 2.771–626.999), CS (OR, 15.904; 95% CI, 5.055–50.042), and DM (OR, 3.949; 95% CI, 1.132–13.783) were independent predictors of MCA stenosis. CSI was simplified for clinical use , CSI = MFV (cm/sec) + 3 * AI (%) + 180 *ΔPI + 90 * CS (presence = 1, absence = 0) + 30 * DM (presence = 1, absence = 0) . The area under the receiver operator characteristic (ROC) curve of MCA stenosis versus MFV, ΔPI, AI, and CSI was .641, .668, .865 and .953. According to ROC curve, cut‐off point for MCA stenosis was suggested as CSI > 180 (sensitivity: 87%, specificity: 92%).CONCLUSION CSI might be useful to enhance diagnostic accuracy.