Abstract P3076: Incident hypertension by biomarker patterns: REasons for Geographic and Racial Differences in Stroke (REGARDS) stud

Background: We recently empirically identified 3 novel patterns of cardiovascular risk factor biomarkers with factor analysis: Renal-inflammatory, thrombo-inflammatory, and dyslipidemic-inflammatory. Hypertension is a major cardiovascular disease risk factor. Whether these patterns are associated with incident hypertension are unknown. Objectives: To determine the association between novel risk patterns and incident hypertension. Methods: REGARDS recruited 30,239 Black and White adults from the 48 contiguous US states in 2003-2007 with a second visit in 2013-2016. Baseline fasting lipids, C-reactive protein (CRP), cystatin C, urine albumin/creatinine ratio, blood urea nitrogen (BUN), complete blood count, and serum albumin were measured in all or most participants. A robust set of other baseline biomarkers was measured in a sex-race stratified sample of 4,400 participants who attended both visits. Factor scores for each pattern were developed and applied to each of the 2,723 participants with all available baseline biomarkers. Hypertension was defined as baseline blood pressure (BP) ≥140/90 mm Hg or self-reported use of antihypertensive medications. We excluded participants with prevalent hypertension. Tertiles of each factor score were computed in the analytical population. Modified Poisson regression estimated the adjusted relative risk (RR) of incident hypertension by factor pattern tertile, testing for pairwise interactions between each factor. Because of a significant thrombo-inflammatory*dyslipidemic-inflammatory interaction on hypertension (P=0.06; interaction significance threshold <0.10), these groups were stratified across each other’s tertiles. Results: Among the 1,223 included (mean [SD] age 61 [8] years, 25% Black race, 64% women), 34% developed incident hypertension. There was no difference in age, sex, race, or baseline systolic BP-adjusted RR of hypertension by tertile of the renal-inflammatory pattern ( Figure ). Relative to the lowest tertile combination, there was a higher RR of incident hypertension across higher dyslipidemic- and thrombo-inflammatory pattern tertiles (3 rd tertile of each: RR 1.67; 1.10 to 2.52). Discussion: Among Black and White adults, thrombo-inflammatory and dyslipidemic-inflammatory biomarker patterns were strongly associated with incident hypertension. Behavioral or medical therapies targeting specific biological pathways represented by these patterns may provide targets to prevent hypertension and its sequelae.