Dietary omega-6 fatty acid subtypes and circadian syndrome in US adults: a NHANES study with exploratory Mendelian randomization analyses

Abstract

Circadian syndrome (CircS) combines metabolic abnormalities with short sleep and depressive symptoms. Whether dietary omega-6 polyunsaturated fatty acid subtypes have distinct associations with CircS remains unclear. We analyzed 11,459 adults from the National Health and Nutrition Examination Survey (NHANES) 2005–2018 using survey-weighted logistic regression to examine dietary total omega-6 polyunsaturated fatty acids, linoleic acid (LA), and arachidonic acid (AA) in relation to CircS. Sensitivity analyses used energy-density exposures and additional adjustment for body mass index, red and processed meat intake, and diagnosis history. Complementary two-sample Mendelian randomization (MR) evaluated genetically predicted circulating omega-6-related traits against selected cardiometabolic and behavioral phenotypes; it did not directly test dietary AA or CircS. Total omega-6 polyunsaturated fatty acid and LA intake were not associated with CircS after full adjustment. Participants in the highest AA quartile had higher odds of CircS than those in the lowest quartile (odds ratio [OR] 1.41, 95% confidence interval [CI] 1.15–1.73; P for trend = 0.003). The association was also observed per 1-standard-deviation increment (OR 1.14, 95% CI 1.06–1.22) and in the energy-density analysis (OR 1.23, 95% CI 1.03–1.48). Additional adjustment for red and processed meat intake modestly attenuated the estimate (OR 1.35, 95% CI 1.09–1.67). Multiple imputation yielded a similar AA estimate (OR 1.43, 95% CI 1.19–1.72). Component-level associations for AA were concentrated in metabolic features, including central obesity, elevated fasting glucose, and reduced HDL-C, whereas short sleep and depressive symptoms showed no FDR-significant trends. MR findings showed the most consistent associations for triglyceride and glycemic traits; however, substantial heterogeneity and sensitivity to outlier correction indicated that several lipid-related associations should be interpreted as exploratory rather than causal evidence. Higher dietary AA intake was positively associated with CircS in this cross-sectional analysis. Component-level findings were concentrated in metabolic features rather than short sleep or depressive symptoms, suggesting that the observed association may primarily reflect metabolic clustering within CircS. The association remained present after excluding the central-obesity component, although residual confounding cannot be excluded. The exploratory MR analyses used different exposures and outcomes and do not validate the dietary AA–CircS association.

The authors' abstract, as published at the source. Lipids in Health and Disease, 2026 · DOI ↗