Indomethacin markedly blunts cerebral perfusion and reactivity, with little cognitive consequence in healthy young and older adults

Abstract

Key points Cognitive function depends on adequate cerebrovascular perfusion and control. However, it is unknown whether acutely‐reduced cerebral blood flow (CBF) impairs cognition in healthy adults. In the present study, we used a placebo‐controlled, single‐blinded, randomized cross‐over design to test the hypothesis that acutely‐reduced CBF (using a pharmacological aid; indomethacin) would impair cognition in young and older healthy adults. At baseline, older adults had lower cognitive performance and CBF, but similar cerebrovascular reactivity to CO2 and dynamic cerebral autoregulation compared to young adults. In both young and older adults, cognitive performance on a mental switching task was slightly (7%) reduced after indomethacin, but not significantly associated with reductions in CBF (∼31%). These results indicate that cognitive performance is broadly resilient against a ∼31% reduction in CBF per se in healthy young and older adults. Abstract Cognitive function depends on adequate cerebrovascular perfusion and control. However, it is unknown whether acutely‐reduced cerebral blood flow (CBF) impairs cognition in healthy adults. Using a placebo‐controlled, single‐blinded, randomized cross‐over design, we tested the hypothesis that acutely‐reduced CBF (using indomethacin [1.2 mg kg–1 oral dose]) would impair cognition in young (n = 13; 25 ± 4 years) and older (n = 12; 58 ± 6 years) healthy adults. CBF and cerebrovascular control were measured using middle cerebral artery blood velocity (MCAvmean) and its reactivity to hypercapnia (CVRHYPER) and hypocapnia (CVRHYPO), respectively. Cognitive function was assessed using a computerized battery including response time tasks. Baseline comparisons revealed that older adults had 14% lower MCAvmean and 15% lower cognitive performance (all P ≤ 0.048), but not lower CVRHYPER/HYPO (P ≥ 0.26). Linear and rank‐based mixed models revealed that indomethacin decreased MCAvmean by 31% (95% confidence interval = –35 to –26), CVRHYPER by 68% [interquartile range (IQR) = –94 to –44] and CVRHYPO by 50% (IQR = –83 to –33) (treatment‐effect; all P < 0.01), regardless of age. Baseline CVRHYPER/HYPO values were strongly associated with their indomethacin‐induced reductions (r = 0.70 to 0.89, P < 0.01). Mental switching performance was impaired 7% (IQR = 0–19) after indomethacin (P = 0.04), but not significantly associated with reductions in MCAvmean (Young: rho = –0.31, P = 0.30; Older: rho = 0.06, P = 0.86). In conclusion, indomethacin reduced MCAvmean and impaired cognition slightly; however, no clear association was evident in younger or older adults. Older adults had poorer cognition and lower MCAvmean, but similar CVRHYPER/HYPO.

Contrasting Measures of Cerebrovascular Reactivity Between MRI and Doppler: A Cross-Sectional Study of Younger and Older Healthy Individuals

“…However, the correlation between TCD CVR measures (average r = 0.43) taken was still much greater than the relationship of the TCD CVR measures to the CVR measures derived from PCA MRI (Supplementary Figure 5A, r = −0.19), which would have been expected to provide similar CVR measures. To fully be able to interpret these results it would be necessary to also assess test re-test reliability in the MRI CVR measures to try and disentangle if it is participant physiology (Shoemaker et al, 2020) or measurement technique that is driving this variability and is contributing to the lack of correlation between modalities. Nevertheless, the variability that we observed with the TCD-CVR is consistent with other studies that have reported CoV for this measure (Shoemaker et al, 2020) and/or examined its variability (Burley et al, 2020).…”

“…To fully be able to interpret these results it would be necessary to also assess test re-test reliability in the MRI CVR measures to try and disentangle if it is participant physiology (Shoemaker et al, 2020) or measurement technique that is driving this variability and is contributing to the lack of correlation between modalities. Nevertheless, the variability that we observed with the TCD-CVR is consistent with other studies that have reported CoV for this measure (Shoemaker et al, 2020) and/or examined its variability (Burley et al, 2020). Finally, the PCA MRI-derived flow measure of CVR in the MCA, which is not affected by the operator to the same extent as TCD acquisition, also did not correlate with the BOLD CVR metrics (Supplementary Figure 5B).…”

The enigma of cerebral blood flow and cognition

“…The novel findings reported by Shoemaker et al . (2020), demonstrate the profound resiliency of cognitive performance regardless of acute reductions in cerebral perfusion, importantly, in both young and older adults (Shoemaker et al . 2020).…”

“…The results reported by Shoemaker et al . (2020) may be extended in several directions. First, older adults use neural compensation mechanisms by recruiting a set of brain areas to cope with the increasing attention load required to perform complex tasks (Ansado et al .…”

“…For example, in syncope prone older adults, McHenry et al (1961) demonstrated that cerebral oxygen extraction increased from ∼43% at rest to 77% before presyncope, with a reduction in CBF by ∼50%. Given the recent findings reported by Shoemaker et al (2020) where a 30% reduction in CBF did not elicit noticeable changes in cognition (at least related to CBF), it could be suggested that the healthy brain is able to compensate for acute and transient decreases in blood flow beyond a 30% reduction without any observable functional deficits in brain/cognitive function. It is unreasonable to suggest that oxygen can be extracted up to the critical value of ∼70% (as observed by McHenry et al 1961) without cognitive deficit because neural function is probably affected prior to reaching maximal extraction.…”