Different Predicting Values Of VO2max Metrics On Cognitive Performances In Older Adults With Cardiovascular Risk Factors

The use of better normalized cardiorespiratory fitness (VO2max) metrics (predicted value (VO2/pred) or lean mass (VO2/LM)) could improve our understanding of the relationship between VO2max and cognition in older adults with cardiovascular risk factors (CVRFs). PURPOSE: Determine the associations between two normalized VO2max metrics (VO2/pred and VO2/LM) and cognitive functions in older individuals with CVRFs. We hypothesized that the VO2/pred will correlate better with cognitive functions than VO2/LM. METHODS: 141 participants (70 ± 6 years) with CVRFs underwent a cardiopulmonary exercise test. Measured VO2max were expressed in percentage of predicted values (VO2/pred) and normalized by lean mass (VO2/LM) measured by bioimpedance. Participants completed a neuropsychological test battery assessing the MoCA, working memory (WM), processing speed (PS), executive function (EF), and verbal memory (VM) for which composite z-score were computed. Participants were classified in terciles for both metrics (VO2/pred : low fit (Lp), medium fit (Mp), high fit (Hp); VO2/LM: low fit (Lm), medium fit (Mm), high fit (Hm)). RESULTS: There was no age difference among the VO2/pred groups, but the Hp group had a higher proportion of women (p < 0.001) and more years of education (p = 0.032). Significant differences between VO2/pred groups were found in the MoCA (p = 0.004), WM (p = 0.039) and EF (p = 0.008) after adjusting for sex and education. VO2/pred was correlated with the MoCA (r = 0.348, p = 0.017), WM (r = 0.349, p = 0.007), PS (r = -0.303, p = 0.026) EF (r = -0.534, p < 0.001) and VM (r = 0.551, p = 0.001) after adjusting for education, sex, and age. For VO2/LM, there was no sex or education difference, but the Hm group was younger (p < 0.001). Significant differences between VO2/LM groups were observed in the MoCA (p = 0.002), WM (p = 0.032) and EF (p < 0.001) after adjusting for age. VO2/LM was correlated with the MoCA (r = 0.346, p = 0.014), WM (r = 0.349, p = 0.008), PS (r = -0.344, p = 0.004) and EF (r = -0.545, p < 0.001) when adjusting for education, sex, and age. CONCLUSIONS: Both higher aerobic fitness metrics are linked to higher MoCA scores and better cognitive functions. However, predicted value normalization appears to be a better predictor of VM while lean mass normalization seems to better predict EF in older adults with CVRFs.