The major purpose of this study was to simultaneously evaluate dCA before and shortly after cerebral vasodilatation evoked by infusion of acetazolamide (ACZ). It was questioned if and to what degree dCA was changed after ACZ infusion. Using 15 mg kg(-1) ACZ infusion cerebrovascular reactivity (CVR) was assessed in 29 first ever lacunar stroke patients (19 M/10 F). During the CVR-test, the electrocardiogram, non-invasive finger arterial blood pressure (ABP) and middle cerebral artery blood flow velocity (CBFV) were recorded. DCA based on spontaneous blood pressure variations was evaluated in 24 subjects by linear transfer function analysis. Squared coherence, gain and phase angle in the frequency range of autoregulation (0.04-0.16 Hz) were compared before and after ACZ infusion. After ACZ infusion, median phase angle decreased significantly (p < 0.005 Wilcoxon) to 0.77 rad compared to a pre-test baseline value of 1.05 rad, indicating less efficient dCA due to ACZ. However, post-test phase values are still mostly within the normal range. Poor and statistically non-significant correlations were found between CVR and absolute dCA phase angle. It can be concluded that CVR testing with body weight adjusted infusion of ACZ lowers dCA performance but by no means exhausts dCA, suggesting that in this way maximal CVR is not determined. Characterizing dCA based on transfer function analysis of ABP to CBFV needs no provocation and adverse patient effects are minimal. The poor correlation between CVR and dCA phase angle supports an interpretation that CVR and dCA study different mechanisms of cerebrovascular control.