Peak exercise SBP and future risk of cardiovascular disease and mortality

Objectives: This study aimed to evaluate the risk of all-cause mortality and incident cardiovascular disease associated with peak systolic blood pressure (PeakSBP) at clinical exercise testing. Methods: Data from 10 096 clinical exercise tests (54% men, age 18-85 years) was cross-linked with outcome data from national registries. PeakSBP was compared with recently published reference percentiles as well as expressed as percentage predicted PeakSBP using reference equations.Natural cubic spline modelling and Cox regression were used to analyse data stratified by sex and baseline cardiovascular risk profile. Results: Median [IQR] follow-up times were 7.9 [5.7] years (all-cause mortality) and 5.6 [5.9] years (incident cardiovascular disease), respectively. The adjusted risk of all-cause mortality [hazard ratio, 95% confidence interval (95% CI)] for individuals with PeakSBP below the 10th percentile was 2.00 (1.59-2.52) in men and 2.60 (1.97-3.44) in women, compared with individuals within the 10th--90th percentile. The corresponding risk for incident cardiovascular disease was 1.55 (1.28-1.89, men) and 1.34 (1.05-1.71, women). For males in the upper 90th percentile, compared with individuals within the 10th--90th percentile, the adjusted risks of all-cause death and incident cardiovascular disease were 0.35 (0.22-0.54) and 0.72 (0.57-0.92), respectively, while not statistically significant in women. Spline modelling revealed a continuous increase in risk with PeakSBP values less than 100% of predicted in both sexes, with no increase in risk more than 100% of predicted. Conclusion: Low, but not high, PeakSBP was associated with an increased risk of mortality and future cardiovascular disease. Using reference standards for PeakSBP could facilitate clinical risk stratification across patients of varying sex, age and exercise capacity.Funding: County Council of Ostergotland [LIO-822461]; Swedish Heart and Lung FoundationSwedish Heart-Lung Foundation [20190064]; Research Foundation Flanders (FWO)FWO [1225021N]; Swedish Research CouncilSwedish Research CouncilEuropean Commission [201902081]