Standing middle cerebral artery velocity predict cognitive performance on working memory tasks in adults with varying levels of cognitive abilities

Abstract

Introduction: The pathophysiology of dementia is multifactorial and reduced cerebral blood flow (CBF) is a predictor of cognitive impairment. The majority of research on CBF and dementia has been completed in the supine position, however during upright-posture the vascular system must counter gravitational forces and the redistribution of blood volume to the working muscles to maintain CBF. Aim: Determine if diastolic middle cerebral artery velocity (MCAdv) during standing is associated with lower performance on a working memory task (WMT, 1-back or 2-back) in older adults. Methods: Fifty-three participants (73±6yrs, 17-females: 14 cognitively intact, 30 mild cognitive impairment, 9 dementia) were analyzed. MCAdv profiles the lowest level of flow within a cardiac cycle. Beat-by-beat MCAdv (transcranial Doppler ultrasound) and mean arterial pressure (MAP, plethysmography) were averaged for 30-sec during supine-rest and standing (figure 1). K-means algorithms identified three homogeneous groups (increased-MCAdv, maintained-MCAdv, and reduced-MCAdv) from standing MCAdv (figure 2). Participants completed a 90-second WMT and a performance-based signal detection variable was calculated [sensitivity-A’=0.5+((hit rate−false alarm rate) × (1+hit rate−false alarm rate))/(4 × hit rate × (1−false alarm rate))], and a speed versus accuracy exchange was calculated [sensitivity-A’ score=100xsensitivity-A’/target reaction time(ms)x100]. ANCOVA (covariate: age), and post-hoc analyses were completed on standing MCAdv and WMT. Multiple linear regressions were completed for WMT with age and either MAP or MCAdv (supine and standing), and simple linear regressions were also completed. Significance was set to p≤0.05. Results: The reduced-MCAdv group had lower WMT scores (p<0.02) compared to the increased- or maintained-MCAdv groups (figure 2). There was a linear correlation between standing-MCAdv and WMT which was not observed while supine (figure 3). Similarly, regression modelling predicted WMT [F(3,49)=2.533, p=0.068, R2=0.134] with standing-MCAdv, supine-MCAdv, and age, and only standing-MCAdv was a significant predictor variable (p=0.013). This relationship was unique to MCAdv and was not observed with MAP. Discussion: Decreased MCAdv when standing, demonstrates a linear relationship with reduced working memory in older adults. This was not observed in the supine position and was not present with measures of peripheral arterial flow (MAP), indicating a unique relationship between dynamic measures of MCAdv and memory tasks, which was not observed at supine rest or in MAP.