Predictive Value of Traditional and Novel Obesity Indices for Stroke and Its Subtypes Across Sexes and Glucose Status: Toward Precision Prevention Strategies.

Novel obesity indices outperformed traditional metrics (BMI, waist circumference) for predicting total and ischemic stroke.

• Eight novel indices were compared: weight-adjusted waist index, a body shape index, body roundness index, visceral adiposity index, conicity index, lipid accumulation product, waist-to-hip ratio, and waist-to-height ratio.
• Traditional measures compared were body mass index and waist circumference.
• Cox proportional hazards and Fine-Gray competing risk models were used to generate hazard ratios with 95% CIs.
• Predictive discrimination was assessed via Harrell's C-statistic.
• The study analyzed 398,270 stroke-free UK Biobank participants in a retrospective cohort design.

Weight-adjusted waist index showed the strongest associations with ischemic stroke risk among all obesity indices examined.

• Men in the highest quartile of weight-adjusted waist index had a 25% increased ischemic stroke risk (HR, 1.25 [95% CI, 1.11–1.36]).
• Women in the highest quartile had a 36% increased ischemic stroke risk (HR, 1.36 [95% CI, 1.20–1.57]).
• The association was stronger in women than in men.
• Analyses were stratified by sex and glucose metabolic status.

Weight-adjusted waist index significantly surpassed BMI in discriminating ischemic stroke risk in both men and women.

• In men, C-statistic for weight-adjusted waist index was 0.712 (95% CI, 0.705–0.720) versus BMI 0.688 (95% CI, 0.680–0.696), P<0.0001.
• In women, C-statistic for weight-adjusted waist index was 0.748 (95% CI, 0.739–0.757) versus BMI 0.726 (95% CI, 0.717–0.735), P=0.0007.
• Both differences in C-statistics were statistically significant.
• Women showed higher absolute C-statistic values than men for both weight-adjusted waist index and BMI.

No significant associations were observed between obesity indices and hemorrhagic stroke.

• This null finding applied across the obesity indices examined.
• The contrast with ischemic stroke findings suggests subtype-specific relationships between adiposity and stroke.
• Both Cox proportional hazards and Fine-Gray competing risk models were used in the analysis.

The predictive performance of weight-adjusted waist index was robust across sexes and glucose-regulated populations, with particular relevance for normoglycemic individuals.

• Analyses were stratified by glucose metabolic status in addition to sex.
• Findings specifically support utility in normoglycemic individuals for ischemic stroke prediction.
• The study addressed limitations of current anthropometric tools by comparing multiple indices across these subgroups.
• Results support sex-specific adiposity assessment in stroke prevention strategies.

The study cohort consisted of 398,270 stroke-free UK Biobank participants analyzed retrospectively.

• The design was a retrospective cohort study.
• Participants were free of stroke at baseline.
• Analyses were stratified by both sex and glucose metabolic status.
• The large sample size enabled subgroup analyses across multiple obesity indices and stroke subtypes.