Adiposity, BMI, and Cognition in Elderly Adults
Adiposity, BMI, and Cognition in Elderly Adults
To the best of the knowledge of the authors, this is the first study to investigate the relationship between cognitive performance and BMI, WC, DEXA body fat, and DEXA abdominal fat in a large community-dwelling elderly cohort without dementia. No linear relationships were found between any of the estimates of adiposity and cognition. Contrary to what was hypothesized, BMI and DEXA body fat had a statistically significant nonlinear relationship with executive function, showing a relationship with a downward curvature. When categorized using BMI, results showed that overweight individuals had significantly higher levels of global cognition and executive function than normal-weight individuals. Results for BMI showed a consistent pattern, with best cognitive performance by overweight participants, followed by obese individuals, and worst performance by those of normal weight. When DEXA body fat was categorized according to sex, results showed that women in the middle and highest tertiles had significantly higher levels of executive function than those in the lowest tertile, suggestive of an association between greater adiposity and better cognition. Executive function and memory were significantly better for men in the middle tertile of DEXA body fat than those in the lowest tertile, suggesting cognitive benefits associated with body fat, but only up to a certain point. DEXA abdominal fat and WC were not associated with cognitive performance.
Contrary to what was hypothesized, DEXA body fat showed no additional benefit over BMI when investigating adiposity and cognitive function during aging, although BMI added to the predictive power of the model using DEXA, specifically when examining executive function. These results indicate that BMI is a better measure for predicting cognitive performance in elderly adults. These results are surprising because the DEXA body fat measure was expected to be the best measure of body fat and thus, based on the hypothesis, to be able to predict cognition more strongly. A previous study suggested a low correlation between cognitive performance and body fat, which may indicate a confounding effect of sarcopenia, frailty, lean mass, illness, or other conditions having greater influence on cognition than body fat, and perhaps BMI takes these confounding effects into account better than DEXA body fat. The relationship between DEXA and BMI alone in elderly adults needs to be explored further because the correlation was only 0.57, although the direction of association with cognition was similar between the two measures.
Although research involving children, adolescents, and adults has repeatedly found adiposity to be associated with poorer cognitive performance, the current findings, together with two other studies involving older-old adults, suggest that this association disappears in those aged 70 and older. Instead, being of normal weight or having a low percentage of body fat could impair cognitive performance. There are compelling potential reasons for this dramatic shift in the association between obesity and cognition throughout the lifespan. For instance, it has been proposed that weight loss in elderly adults contributes to poor cognitive performance and may be a sign of dementia. However, the current study controlled for weight change that might have occurred in the past 4 years, which could be indicative of an underlying medical condition. Some cross-sectional studies have reported that age-related sarcopenia from inactivity or from medical conditions (cachexia) is associated with cognitive deficits. A study found that poor cognitive function assessed using the Short Portable Mental Status Questionnaire was associated with low fat-free mass in a large cross-sectional study of community-dwelling women aged 75 and older (N = 3,027). It could therefore be that older adults with higher BMI and higher body fat composition do not suffer as greatly from sarcopenia and its effects on cognition.
Previous studies have found that the relationship between WC and cognition disappears after the age of 70, because this relationship was attenuated with age. It seems that visceral fat, which is relevant in younger samples because of the risk of metabolic syndrome and cognitive impairment, does not affect the brain in elderly adults, and further research is needed to understand the mechanisms.
The current findings appear paradoxical because adiposity induces low-grade elevation in inflammatory markers, which have repeatedly been shown to be negatively associated with cognitive performance and predict cognitive decline in elderly adults. It would be expected that obese individuals would have higher levels of inflammatory markers and consequently lower cognitive performance, but it was found that overweight individuals in the current sample had lower inflammatory markers than those who were normal weight and obese at baseline (unpublished data). This in turn suggests that, although the current findings involving obese individuals and better cognition are contrary to predictions based on low-grade inflammation, less inflammation may explain the association between overweight and better cognition in this sample. Because inflammatory markers were not measured during the same wave as the DEXA measures, this cannot be confirmed in the current sample.
There are other possible biological mechanisms that may explain better cognition in overweight and obese individuals. Overweight individuals retain testosterone or estrogen in the body fat, which in turn could help prevent cognitive impairment. Higher testosterone or estrogen has been linked to higher cognitive function, giving overweight and obese people a cognitive advantage. It is also possible that the findings reflect a "survivor" effect in the sample, which means that overweight and obese individuals are more likely to die at younger ages and that this elderly sample would then include the "survivors," who are healthier. A study that pooled data from 19 studies and included 1.46 million white men and women aged 19–84 found obese individuals to be 2.5 times as likely to die young. Together with this "survivor" effect, it is possible that "healthy" obesity, which has been suggested to involve up to 30% of obese people, explains some of the positive associations.
This study has major strengths, including sample size, use of a comprehensive battery of neuropsychological tests, and incorporation of multiple covariates, including APOE, symptoms of depression, and weight change. The study is limited by its cross-sectional design and a lack of data pertaining to the duration of obesity. The data indicate that the average body weight of the sample increased by 1.4 kg over the 4 years of the study, but body weight and adiposity over the lifespan may influence cognition. Nevertheless, future longitudinal observations in this well-defined population can help delineate whether these relationships are maintained.
In conclusion, this study supports the growing evidence that being overweight, or having extra body fat, in older-old adults (≥70) is associated with better cognitive function, and the simpler BMI demonstrated this association better than the more-elaborate DEXA measure. Longitudinal studies are needed to examine the long-term effects of adiposity and changes in adiposity on cognitive performance and change in cognition in individuals aged 70 and older.
Discussion
To the best of the knowledge of the authors, this is the first study to investigate the relationship between cognitive performance and BMI, WC, DEXA body fat, and DEXA abdominal fat in a large community-dwelling elderly cohort without dementia. No linear relationships were found between any of the estimates of adiposity and cognition. Contrary to what was hypothesized, BMI and DEXA body fat had a statistically significant nonlinear relationship with executive function, showing a relationship with a downward curvature. When categorized using BMI, results showed that overweight individuals had significantly higher levels of global cognition and executive function than normal-weight individuals. Results for BMI showed a consistent pattern, with best cognitive performance by overweight participants, followed by obese individuals, and worst performance by those of normal weight. When DEXA body fat was categorized according to sex, results showed that women in the middle and highest tertiles had significantly higher levels of executive function than those in the lowest tertile, suggestive of an association between greater adiposity and better cognition. Executive function and memory were significantly better for men in the middle tertile of DEXA body fat than those in the lowest tertile, suggesting cognitive benefits associated with body fat, but only up to a certain point. DEXA abdominal fat and WC were not associated with cognitive performance.
Contrary to what was hypothesized, DEXA body fat showed no additional benefit over BMI when investigating adiposity and cognitive function during aging, although BMI added to the predictive power of the model using DEXA, specifically when examining executive function. These results indicate that BMI is a better measure for predicting cognitive performance in elderly adults. These results are surprising because the DEXA body fat measure was expected to be the best measure of body fat and thus, based on the hypothesis, to be able to predict cognition more strongly. A previous study suggested a low correlation between cognitive performance and body fat, which may indicate a confounding effect of sarcopenia, frailty, lean mass, illness, or other conditions having greater influence on cognition than body fat, and perhaps BMI takes these confounding effects into account better than DEXA body fat. The relationship between DEXA and BMI alone in elderly adults needs to be explored further because the correlation was only 0.57, although the direction of association with cognition was similar between the two measures.
Although research involving children, adolescents, and adults has repeatedly found adiposity to be associated with poorer cognitive performance, the current findings, together with two other studies involving older-old adults, suggest that this association disappears in those aged 70 and older. Instead, being of normal weight or having a low percentage of body fat could impair cognitive performance. There are compelling potential reasons for this dramatic shift in the association between obesity and cognition throughout the lifespan. For instance, it has been proposed that weight loss in elderly adults contributes to poor cognitive performance and may be a sign of dementia. However, the current study controlled for weight change that might have occurred in the past 4 years, which could be indicative of an underlying medical condition. Some cross-sectional studies have reported that age-related sarcopenia from inactivity or from medical conditions (cachexia) is associated with cognitive deficits. A study found that poor cognitive function assessed using the Short Portable Mental Status Questionnaire was associated with low fat-free mass in a large cross-sectional study of community-dwelling women aged 75 and older (N = 3,027). It could therefore be that older adults with higher BMI and higher body fat composition do not suffer as greatly from sarcopenia and its effects on cognition.
Previous studies have found that the relationship between WC and cognition disappears after the age of 70, because this relationship was attenuated with age. It seems that visceral fat, which is relevant in younger samples because of the risk of metabolic syndrome and cognitive impairment, does not affect the brain in elderly adults, and further research is needed to understand the mechanisms.
The current findings appear paradoxical because adiposity induces low-grade elevation in inflammatory markers, which have repeatedly been shown to be negatively associated with cognitive performance and predict cognitive decline in elderly adults. It would be expected that obese individuals would have higher levels of inflammatory markers and consequently lower cognitive performance, but it was found that overweight individuals in the current sample had lower inflammatory markers than those who were normal weight and obese at baseline (unpublished data). This in turn suggests that, although the current findings involving obese individuals and better cognition are contrary to predictions based on low-grade inflammation, less inflammation may explain the association between overweight and better cognition in this sample. Because inflammatory markers were not measured during the same wave as the DEXA measures, this cannot be confirmed in the current sample.
There are other possible biological mechanisms that may explain better cognition in overweight and obese individuals. Overweight individuals retain testosterone or estrogen in the body fat, which in turn could help prevent cognitive impairment. Higher testosterone or estrogen has been linked to higher cognitive function, giving overweight and obese people a cognitive advantage. It is also possible that the findings reflect a "survivor" effect in the sample, which means that overweight and obese individuals are more likely to die at younger ages and that this elderly sample would then include the "survivors," who are healthier. A study that pooled data from 19 studies and included 1.46 million white men and women aged 19–84 found obese individuals to be 2.5 times as likely to die young. Together with this "survivor" effect, it is possible that "healthy" obesity, which has been suggested to involve up to 30% of obese people, explains some of the positive associations.
This study has major strengths, including sample size, use of a comprehensive battery of neuropsychological tests, and incorporation of multiple covariates, including APOE, symptoms of depression, and weight change. The study is limited by its cross-sectional design and a lack of data pertaining to the duration of obesity. The data indicate that the average body weight of the sample increased by 1.4 kg over the 4 years of the study, but body weight and adiposity over the lifespan may influence cognition. Nevertheless, future longitudinal observations in this well-defined population can help delineate whether these relationships are maintained.
In conclusion, this study supports the growing evidence that being overweight, or having extra body fat, in older-old adults (≥70) is associated with better cognitive function, and the simpler BMI demonstrated this association better than the more-elaborate DEXA measure. Longitudinal studies are needed to examine the long-term effects of adiposity and changes in adiposity on cognitive performance and change in cognition in individuals aged 70 and older.
Source...