Impact of bariatric surgery on cerebral vascular reactivity and cognitive function: a non-randomized pilot study

Background Bariatric surgery is an effective long-term weight loss strategy yielding improvements in neurocognitive function; however, the mechanism(s) responsible for these improvements remains unclear. Here, we assessed the feasibility of using magnetic resonance imaging (MRI) to evaluate whether cerebral vascular reactivity (CVR) is impaired in severely obese bariatric surgery candidates compared with normal weight healthy controls and whether CVR improves following bariatric surgery. We also investigated whether changes in CVR were associated with changes in cognitive function. Methods Bariatric surgery candidates ( n = 6) were compared with normal weight healthy controls of a similar age ( n = 10) at baseline, and then reassessed 2 weeks and 14 weeks following sleeve gastrectomy bariatric surgery. Young reference controls ( n = 7) were also studied at baseline to establish the range of normal for each outcome measure. Microvascular and macrovascular CVR to hypercapnia (5% CO 2 ) were assessed using blood-oxygen-level-dependent (BOLD) MRI, and changes in the middle cerebral artery (MCA) cross-sectional area, respectively. Cognitive function was assessed using a validated neurocognitive software. Results Compliance with the CVR protocol was high. Both macro- and micro-cerebrovascular function were highest in the young reference controls. Cognitive function was lower in obese bariatric surgery candidates compared with normal weight controls, and improved by 17% at 2 weeks and 21% by 14 weeks following bariatric surgery. To our surprise, whole-brain CVR BOLD did not differ between obese bariatric surgery candidates and normal weight controls of similar age (0.184 ± 0.101 vs. 0.192 ± 0.034 %BOLD/mmHgCO 2 ), and did not change after bariatric surgery. In contrast, we observed vasoconstriction of the MCA during hypercapnia in 60% of the obese patients prior to surgery, which appeared to be abolished following bariatric surgery. Improvements in cognitive function were not associated with improvements in either CVR BOLD or MCA vasodilation after bariatric surgery. Conclusions Assessing CVR responses to a hypercapnic challenge with MRI was feasible in severely obese bariatric patients. However, no changes in whole-brain BOLD CVR were observed following bariatric surgery despite improvements in cognitive function. We recommend that future large trials assess CVR responses to cognitive tasks (rather than hypercapnia) to better define the mechanisms responsible for cognitive function improvements following bariatric surgery.