**Background:** Stroke is the second leading cause of death worldwide, affecting approximately 13.7 million people annually. High-density lipoprotein cholesterol (HDL-C) is known to protect against cardiovascular disease through reverse cholesterol transport, anti-inflammatory, antioxidant, and antithrombotic mechanisms. However, the relationship between HDL-C and stroke remains controversial, with some studies showing inverse associations, others showing no association, and some meta-analyses suggesting high HDL-C may increase stroke risk. Few studies have examined the dose-response relationship across different sex and racial groups.
**Methods:** This cross-sectional study analyzed data from 8,643 participants (4,222 men, 4,421 women) aged ≥40 years from the National Health and Nutrition Examination Survey (NHANES) 2007–2016 cycles. Participants with missing stroke data (n=27,078), missing HDL-C data (n=29,718), cancer (n=1,304), or age <40 (n=4,971) were excluded. HDL-C was measured by direct immunoassay (Roche Modular P analyzer 2007–2012; Roche Cobas 6000 and Modular P 2013–2016). Stroke was defined by self-reported physician diagnosis. Covariates included age, sex, race/ethnicity, LDL-C, total cholesterol, triglycerides, HbA1c, income-to-poverty ratio, smoking, education, BMI, alcohol consumption, diabetes, and hypertension. Multivariate logistic regression models, smooth curve fitting, generalized additive models, and two-piecewise linear regression were used. Three models were constructed: Model 1 (unadjusted), Model 2 (adjusted for age, sex, race), and Model 3 (fully adjusted for all covariates).
**Key Results:** The study included 8,643 participants (51.15% women; mean age 58.35±11.68 years for women, 58.48±12.04 for men). Baseline characteristics differed significantly across HDL-C quintiles (p<.05). Higher HDL-C groups had higher total cholesterol, more women, more White and Black participants, higher education, more alcohol consumption, higher income, and lower BMI, triglycerides, and HbA1c. In the fully adjusted model, HDL-C as a continuous variable was inversely associated with stroke (OR 0.69, 95% CI: 0.49–0.96, p<.05). Compared to Q1 (<1.04 mmol/L), stroke risk was reduced by 33% in Q2 (OR 0.67, 95% CI: 0.5094–0.8860), 39% in Q3 (OR 0.61, 95% CI: 0.4211–0.8962), and 61% in Q4 (OR 0.39, 95% CI: 0.1906–0.8141). Q5 (≥2.32 mmol/L) showed a non-significant 32% reduction (OR 0.68, 95% CI: 0.3388–1.3771). A nonlinear relationship was identified with an inflection point at 1.55 mmol/L. Below this threshold, each unit increase in HDL-C reduced stroke odds by 64% (OR 0.36, 95% CI: 0.21–0.62, p=.0002). Above 1.55 mmol/L, the association was not significant (OR 1.29, 95% CI: 0.79–2.09, p=.3057). Subgroup analyses showed significant inverse associations in men (OR 0.58, 95% CI: 0.34–0.98, p=.0424) and Whites (OR 0.49, 95% CI: 0.29–0.82, p=.0070), but not in women (OR 0.77, 95% CI: 0.50–1.18, p=.2316), Mexican Americans (OR 0.64, 95% CI: 0.21–2.00, p=.4436), Blacks (OR 1.03, 95% CI: 0.61–1.75, p=.9035), or other races (OR 0.43, 95% CI: 0.14–1.35, p=.1463).
**Clinical Implications:** This study demonstrates a nonlinear, threshold-dependent relationship between HDL-C and stroke risk in adults over 40. The protective effect is most pronounced at HDL-C levels below 1.55 mmol/L, with no significant benefit or harm at higher levels. The findings suggest that maintaining HDL-C within an appropriate range may reduce stroke incidence, particularly in men and White populations. However, the cross-sectional design precludes causal inference, and the study is limited by self-reported stroke, inability to distinguish ischemic from hemorrhagic stroke, small sample size at extreme HDL-C levels, and geographic/ethnic restrictions to the U.S. population. Future prospective studies are needed to confirm these findings and explore race-specific differences.