The Evaluation Of Flow Velocity In The Unilateral Middle Cerebral Artery Stenosis By Transcranial Doppler

Objective: To determine the optimal velocity values in diagnosing the degree ofunilateral middle cerebral artery (MCA) stenosis by Transcranial Doppler (TCD), and findother useful parameters to improve diagnostic accuracy using magnetic resonanceangiography (MRA) as the reference standard.Methods: Nonconsecutive patients with unilateral MCA stenosis referred for TCD gaveinformed consent to also undergo MRA of the intracranial arteries. The degree of middlecerebral artery stenosis was categorized into four groups: normal (normal caliber and signal),mild(<50%), moderate (30~69%), severe (70~99%, or no flow detected), using MRA forconfirmation. The peak systolic velocity (PSV), end-diastolic velocity (EDV), mean flowvelaocity (MFV), and the TCD spectrum of each MCA were recorded. The receiveroperating characteristic curve (ROC curve) was used to determine the optimal velocity fornormal, mild, moderate and severe groups. The coincidence rate of unilateral MCA stenosisdiagnoses by TCD and MRA was calculated with Kappa number.Results:302patients (70.9%males and29.1%females; mean age:56.6±12.5yrs) atthe hospital from January2010to October2011were included in this study. Among604arteries, there were319(52.8%)normal，94(15.6%)mild,66(10.9%)moderate and125(20.7%)severe. The velocity among these four groups were significant (P<0.001,respectively). The optimal velocity values for normal and stenosis were: PSV=160cm/s (areaunder the curve0.923, sensitivity91.6%, specificity93.1%, positive predictive value92.2%,negative predictive value92.5%), MFV=100cm/s (0.911，88.4%,93.7%,92.6%,90.1%); formild and moderate were: PSV=200cm/s(0.756,78.8%,72.3%,66.7%,82.9%),MFV=120cm/s(0.714,75.8%,67.0%,61.7%,79.7%); for moderate and severe were:PSV=280cm/s(0.793,63.2%,95.5%,96.3%,57.8%), MFV=180cm/s(0.787,66.4%,90.9%,93.3%,58.8%). Using PSV, MFV as the diagnostic criteria respectively, the Kappanumber=0.668,0.641. The optimal PSVaffected-healthvalue for mild and moderate was70cm/s(0.772,78.8%,75.5%,69.3%,83.5%), for moderate and severe was120cm/s(0.760,80.8%,71.2%,84.2%,66.2%). Optimal combined criteria for moderate stenosis werePSV>200cm/s and PSVaffected-health>70cm/s (specificity87.2%), for severe stenosis werePSV>280cm/s and PSVaffected-health>120cm/s (specificity95.5%). The spectral waveform abnormalities were common in moderate and severe stenosis arteries.Conclusions: TCD can distinguish normal and stenosis MCA, and different degrees ofmiddle cerebral artery stenosis. Using PSV or MFV as the diagnostic criteria, TCD has highcoincidence rate with MRA in diagnosis of MCA stenosis, PSV criteria had more accuracy.Combined PSVaffected-healthcriteria and the abnormal spectrum may improve the specificity ofTCD in the diagnosis of moderate or severe stenosis.