Improved Cerebral Vasomotor Reactivity After Exercise Training in Hemiparetic Stroke Survivors

Background and Purpose-Animal studies provide strong evidence that aerobic exercise training positively influences cerebral blood flow, but no human studies support the use of exercise for improving cerebral hemodynamics. This randomized study in stroke survivors assessed the effects of treadmill aerobic exercise training (TM) on cerebral blood flow parameters compared to a control intervention of nonaerobic stretching. Methods-Thirty-eight participants (19 in TM group and 19 in control group) with remote stroke (>6 months) and mild to moderate gait deficits completed middle cerebral artery blood flow velocity measurements by transcranial Doppler ultrasonography before and after a 6-month intervention period. Middle cerebral artery blood flow velocity was assessed bilaterally during normocapnia and hypercapnia (6% CO2). Cerebral vasomotor reactivity (cVMR) was calculated as percent change in middle cerebral artery blood flow velocity from normocapnia to hypercapnia (cVMR percent) and as an index correcting percent change for absolute increase in end tidal CO2 (cVMR index). Results-The TM group had significantly larger improvements than did controls for both ipsilesional and contralesional cVMR index (P <= 0.05) and contralesional cVMR percent (P <= 0.01). Statin users in the TM group (n=10) had higher baseline cVMR and lower training-induced cVMR change, indicating that cVMR change among those not using statins (n=9) primarily accounted for the between-group effects. There was a 19% increase in VO2 peak for the TM group compared to a 4% decrease in the control group (P<0.01), and peak fitness change correlated with cVMR change (r=0.55; P<0.05). Conclusions-Our data provide the first evidence to our knowledge of exercise-induced cVMR improvements in stroke survivors, implying a protective mechanism against recurrent stroke and other brain-related disorders. Statin use appears to regulate cVMR and the cVMR training response. (Stroke. 2011; 42: 1994-2000.)