The accuracy of transcranial Doppler in the diagnosis of middle cerebral artery stenosis

Abstract

Background and purpose: It was the aim of this study to systematically review available literature on the accuracy of transcranial Doppler (TCD) compared with angiography for the diagnosis of > or =50% middle cerebral artery stenosis in patients with transient ischemic attack or ischemic stroke.

Methods: We performed a systematic review that included original articles published on TCD accuracy from 1982 until the end of December 2005 using angiography as the gold standard. The following measures of diagnostic accuracy were obtained from each primary study: sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Weighted mean averages were then calculated from individual results for different velocity cutoffs.

Results: Six papers met our selection criteria. Using laboratory-specific variable mean flow velocity cutoffs, self-reported best accuracy results yield a mean weighted average sensitivity of 92%, specificity of 92%, PPV of 88% and NPV of 98% for 80 cm/s cutoff. For 100 cm/s cutoff, the sensitivities were 100%, specificity 97%, PPV 88% and NPV 100%.

Conclusions: Although limited to few reports, this analysis demonstrates fair TCD performance against angiography. Since increasing velocity cutoffs do not yield decreasing sensitivity and increasing specificity, further studies are required to determine optimal velocity values and possibly other criteria such as velocity ratios to develop a screening test with balanced performance parameters.