Effects of vasodilating medications on cerebral haemodynamics in health and disease: systematic review and meta-analysis

Abstract

Objectives: Vasodilating antihypertensives prevent stroke and potentially cerebral small vessel disease but their effects on cerebrovascular haemodynamics beyond blood pressure lowering are unclear.

Methods: We searched PubMed, Medline, Embase, Cinahl, Psychinfo, Health Business Elite and Health Management Information Consortium for randomized studies of vasodilating medications, compared to no treatment or nonvasodilators, that reported effects on cerebral blood flow (CBF), mean blood flow velocity (MFV) or cerebrovascular reactivity. Absolute and standardized mean differences (SMD) were combined by inverse-variance weighted fixed or random-effects meta-analysis stratified by study design, population characteristics and vasodilator class.

Results: In 35 studies reporting 57 comparisons, there was a reduction in SBP (-4.13 mmHg, -7.55 to -0.71, P = 0.018) but no change in MFV (ΔMFV 1.11, confidence interval -0.93 to 3.14, P = 0.29, 23 comparisons). MFV increased in patients with underlying conditions (3.41, 0.24 to 6.57, P = 0.04) but not in healthy study participants (-1.27, -5.18 to 2.64, P = 0.68), with no differences by vasodilating drug class. Cerebral pulsatility index was reduced across all studies (Δ pulsatility index -0.04, -0.07 to -0.02, P = 0.001; Δ pulsatility index -SMD -0.32, -0.47 to -0.16, P < 0.001), except in studies reporting responses to single drug doses (Δ pulsatility index 0.00, -0.09 to -0.08, P = 0.93). Despite evidence of reporting and publication bias, there was an apparent consistent reduction in CBF with vasodilators (CBF-SMD -0.24, -0.46 to -0.02, P = 0.03) with a significant increase in cerebrovascular reactivity-SMD (0.48, 0.13-0.83, P = 0.007).

Conclusions: Despite reducing SBP, vasodilators did not significantly impair absolute CBF but improved cerebrovascular pulsatility and reactivity, suggesting therapeutic potential in preventing stroke and cerebral small vessel disease.

FIGURE 1

Forest plot of difference in mean flow velocity between groups randomized to a vasodilating medication vs. a non-vasodilating control group. Control groups included patients treated with either placebo, a non-vasodilating medication or no change in treatment, with effect sizes combined by both fixed effects and random effects meta-analysis weighted by the inverse variance. ΔMFV, change in mean flow velocity; Cilow, confidence interval lower limit; Cihigh, confidence interval upper limit; GTN, glyceryl trinitrate; HCTZ, hydrochlorothiazide; N, number of study participants; P-het, P value for heterogeneity.

FIGURE 2

Forest plot of difference in cerebral arterial pulsatility index between groups randomized to a vasodilating medication vs. a non-vasodilating control group. Control groups included patients treated with either placebo, a non-vasodilating medication or no change in treatment, with effect sizes combined by both fixed effects and random effects meta-analysis weighted by the inverse variance. All studies report the MCA pulsatility index except for Zhang et al.[19] who reported basilar artery pulsatility index. ΔPI, change in mean flow velocity; Cilow, confidence interval lower limit; Cihigh, confidence interval upper limit; GTN, glyceryl trinitrate; N, number of study participants; P-het, P value for heterogeneity.

FIGURE 3

Forest plot of difference in cerebral blood flow between groups randomized to a vasodilating medication vs. a non-vasodilating control group. Control groups included patients treated with either placebo, a non-vasodilating medication or no change in treatment, with effect sizes combined by both fixed effects and random effects meta-analysis weighted by the inverse variance. Differences between groups are expressed as the SMD. ΔCBF, change in cerebral blood flow velocity; Cilow, confidence interval lower limit; Cihigh, confidence interval upper limit; GTN, glyceryl trinitrate; HCTZ, hydrochlorothiazide; N, number of study participants; P-het, P value for heterogeneity; SMD, standardized mean difference.

FIGURE 4

Forest plot of difference in reactivity of cerebral blood flow or blood flow velocity to a CO₂ challenge between groups randomized to a vasodilating medication vs. a non-vasodilating control group. Control groups included patients treated with either placebo, a non-vasodilating medication or no change in treatment (A) or either placebo or no change in treatment (B), with effect sizes combined by both fixed effects and random effects meta-analysis weighted by the inverse variance. Differences between groups are expressed as the SMD. ΔCVR, change in cerebral vascular reactivity; Cilow, confidence interval lower limit; Cihigh, confidence interval upper limit; HCTZ, hydrochlorothiazide; N, number of study participants; P-het, P value for heterogeneity; SMD, standardized mean difference.

FIGURE 5

Forest plot of difference in SBP and DBP between groups randomized to a vasodilating medication vs. a non-vasodilating control group. Control groups included patients treated with either placebo, a non-vasodilating medication or no change in treatment, with effect sizes combined by both fixed effects and random effects meta-analysis weighted by the inverse variance. Differences between groups are expressed as the SMD. Cilow, confidence interval lower limit; Cihigh, confidence interval upper limit; GTN, glyceryl trinitrate; HCTZ, hydrochlorothiazide; N, number of study participants; P-het, P value for heterogeneity; SMD, standardized mean difference.