To investigate whether C-reactive protein (CRP) can modulate the relationship between obesity phenotypes and stroke risk in middle-aged and elderly individuals through a cohort study.

Utilizing data from the China Health and Retirement Longitudinal Study (CHARLS), participants from the baseline survey in 2011 and follow-ups in 2013, 2015, and 2018 were selected. Based on the participants’ weight status (overweight/obese) and metabolic abnormalities, they were categorized into four obesity phenotypes: Metabolically Healthy Normal Weight (MHNO), Metabolically Healthy Overweight/Obese (MHO), Metabolically Unhealthy Normal Weight (MANO), and Metabolically Unhealthy Overweight/Obese(MAO). Additionally, participants were grouped into three categories based on CRP quartiles. The Cox proportional hazards regression model was employed to analyze the relationship between obesity phenotypes, CRP, and stroke, incorporating interaction terms between obesity phenotypes and CRP in the regression model. Finally, stratified analyses were conducted to explore whether CRP modulates the relationship between obesity phenotypes and stroke.

Totally 6 868 participants were included in this study, with 421 (6.13%) experiencing a stroke during the 7-year follow-up period. The Cox proportional hazards regression model indicated that, compared to the MHNO group, the risk of stroke increased by 66%, 85%, and 151% for the MHO, MANO, and MAO groups, respectively, with hazard ratios (HR) and 95% confidence intervals (CI) of 1.66 (1.15-2.38), 1.85 (1.39-2.46), and 2.51 (1.92-3.29), respectively. Furthermore, stratified analysis revealed that in the CRP T1 group, the MHO and MANO groups did not show an increased risk of stroke compared to the MHNO group (P > 0.05). In the CRP T2 group, the risk of stroke for the MHO and MANO groups increased by 111% and 127%, with HR and 95%CI of 2.11 (1.11-4.02) and 2.27 (1.29-3.98), respectively. Similarly, in the CRP T3 group, the risks for the MHO and MANO groups increased by 93% and 89%, with HR and 95%CI of 1.93 (1.06-3.50) and 1.89 (1.17-3.05), respectively.

CRP can modulate the relationship between obesity phenotypes and stroke in middle-aged and elderly individuals.