Objectives Todynamicallyobservethechangesofhemodynamicparametersinpatients with severe stenosis of unilateral middle cerebral artery (MCA)by transcranial Doppler ultrasound (TCD) andtoevaluateandanalyzetherelatedfactorsforinfluencingthestenoticprocess.Methods Atotalof 113 consecutive patients with severe stenosis of unilateral MCA screened by TCD and confirmed by computed tomography angiography (CTA)and digital subtraction angiography (DSA)were enrolled retrospectively. They were divided into either a progressive group (n =43 )or a non-progressive group (n=90)according to the variation of MCA hemodynamic parameters. The effects of age,sex,major risk factors for cerebrovascular disease,clinical symptoms,clinical medication,and drug compliance on the stenotic process were documented and analyzed. Results (1)The comparison of detection rate of the risk factors for cerebrovascular disease:The patients with a history of smoking (72. 1%[n=31])in the progressive group was significantly higher than that (51. 1%[n=46])in the non-progressive group (P=0.022). The period of smoking of the patients in the progressive group were longer than that in the non-progressive group (28 ± 12 years vs. 21 ± 10 years,P=0. 011). (2)Comparison of MCA hemodynamic parameters:The distal pulsatility indexes of MCA stenosis at the first diagnosis in the progressive group were all lower than those in the non-progressive group (0. 66 ± 0. 10 vs. 0. 70 ± 0. 13;t= -2. 096,P=0. 038),and the distal pulsatility indexes of MCA stenosis at the end point in the patients of the progressive group were lower than those in the non-progressive group (0. 61 ± 0. 15 vs. 0. 74 ± 0. 15). There were significant differences (t=-2. 718,P= 0. 008). The peak systolic velocity (PSV)of the progressive MCA stenotic segments at the end point in 10 patients of the progressive group was higher than that in the non-progressive group (299 ± 23 cm/s vs. 244 ± 50 cm/s,t=3. 437;P=0. 001),while PSV of MCA in 33 patients with occlusion in the progressive group were significantly lower than those in the non-progressive group (56 ± 18 cm/s vs. 244 ± 50 cm/s,t= -20. 905;P=0. 000). (3)The regular medication:The patients using statins (atorvastatin calcium)were significantly lower than those of the non-progressive group (2. 3%[n=1] vs. 54. 4%[n=49],χ2 =33. 690;P<0. 01). (4)During the follow up period,the recurrence rates of transient ischemic attack and stroke of the progressive group were significantly higher than those of the non-progressive group (27. 9%[n=12]vs. 6. 7%[n=6],32. 6%[n=14]vs. 2. 2%[n=2];all P<0.01). (5)Multivariate Logistic regression analysis showed that smokers (OR,4. 403,95%CI 1. 094-14.017),cerebrovascular event recurrence (OR,10. 648,95%CI 2. 530 -41. 261),and irregularly taking statins (OR,5. 675,95%CI 1. 631-152. 740)were all closely associated with the progress of severeMCAstenosis.Conclusion EvaluationofthehemodynamicchangesofsevereMCAstenosiswith TCD follow up study can be used as an important basis for clinical assessment of the outcomes. Stop smoking and regularly taking statins may help to delay the progress of MCA stenosis.