BackgroundEmerging evidence suggests an association between remnant cholesterol (RC) and vascular damage and hypertension. However, this association has not been explored in a large-scale population in China, and a temporal relationship between RC and hypertension also needs to be investigated.MethodsWe conducted a retrospective cross-sectional study in 2,199,366 individuals and a longitudinal study in 24,252 individuals with repeated measurements of lipid profile and blood pressure in at least a 3-year follow-up. The logistic model was used to explore the association between lipid components and hypertension in the cross-sectional analysis. The Cox model was used to analyze the association between high RC (HRC) at baseline and the subsequent incidence of hypertension or the association between hypertension at baseline and incidence of HRC. The cross-lagged panel model was applied to analyze the temporal relationship between RC and hypertension.ResultsRC level as a continuous variable had the highest correlation with hypertension among lipid profiles, including RC, low-density lipoprotein cholesterol, total cholesterol, non-high-density lipoprotein cholesterol, and triglycerides, with an odds ratio of 1.59 (95% confidence interval: 1.58–1.59). In the longitudinal cohort, HRC at baseline was associated with incident hypertension. We further explored the temporal relationship between RC and hypertension using the cross-lagged analysis, and the results showed that RC increase preceded the development of hypertension, rather than vice versa.ConclusionsRC had an unexpected high correlation with the prevalence and incidence of hypertension. Moreover, RC increase might precede the development of hypertension, suggesting the potential role of RC in the development of hypertension.