Abstract
BACKGROUND: Obesity is a risk factor for dementia within the old population however not within the middle-aged population, that is referred to the "obesity paradox". This study explored the association of body mass index (BMI) and waist circumference (WC) with post-stroke cognitive impairment (PSCI) in middle-aged (40-65 years) versus old population (≥ 65 years). METHODS: The current study enrolled 1735 individuals over the age of 40 who had their first ischemic stroke from the Impairment of Cognition and Sleep (ICONS) subgroup of the China National Stroke Registry-3 (CNSR-3). BMI and WC were used for the diagnosis of obesity and central obesity, respectively. PSCI was diagnosed according to the Montreal Cognitive Assessment (MoCA). The main clinical outcome was the incidence of PSCI assessed at three months after stroke. Multivariable regression analysis was performed to evaluate the association between obesity and three-month PSCI. Stratified analysis was also performed to explore the effect of age on the relationship between obesity and PSCI. RESULTS: In the general population, multivariable logistic regression found that the adjusted odds ratio (OR) with 95% confidence interval (CI) of general obesity was 1.45 (1.06-1.98) and that of central obesity was 1.54 (1.24-1.91) for the three-month incidence of PSCI. Stratified analysis by age showed that the adjusted OR with a 95% CI of general obesity was 1.84 (1.24-2.72) in middle-aged patients and 0.89 (0.52-1.54) in elderly patients (p-value for interaction = 0.05). Central obesity was associated with PSCI in all age groups: 1.57 (1.18-2.09) in middle-aged patients and 1.52 (1.08-2.15) in elderly patients (p-value for interaction= 0.93). CONCLUSION: General obesity was related to an increased risk of PSCI in middle-aged but not elderly patients, whereas central obesity was associated with an increased risk of PSCI in all age groups, suggesting that the obesity paradox arises only obesity is outlined by BMI.