The shape of the systolic blood pressure response during graded exercise: methodology, correlates and predictive value

During graded exercise, systolic blood pressure (SBP) is expected to increase linearly, but other responses are observed. To date, a framework for algorithmic assessment of the shape of the SBP response is lacking, making its physiological and clinical relevance poorly understood. We aimed to algorithmically identify distinct SBP response shapes and analyze their association with clinical factors and all-cause mortality. We retrospectively analyzed SBP recordings from a cohort of 5633 patients (mean age, 55.1 years; 43% female) undergoing maximal cycle ergometry, who met strict quality criteria, including ≥30 mmHg increase in SBP during exercise. Per patient, test duration and SBP values were rescaled (0-100%) to retrieve the SBP response shape. Group-based trajectory modelling (GBTM) was used to classify SBP shapes by sex. Associations with clinical factors and all-cause mortality were evaluated using multinomial logistic regression and Cox survival analysis. In both sexes, GBTM identified three SBP response shapes: early, linear and late rise in SBP. Late rise was associated with higher resting SBP, lower peak SBP and smaller increases in SBP during exercise (P < 0.05). A late SBP rise related independently to older age, higher body mass index, beta blocker use and lower exercise capacity. A late SBP rise predicted increased all-cause mortality in men (HR_adjusted versus early SBP rise: 1.66, 95% CI: 1.10-2.50; P = 0.015) but not in women (1.06, 0.60-1.90; P = 0.84). In conclusion, a late SBP response was linked to a worse risk profile and independently predicted all-cause mortality in men, suggesting clinical relevance for SBP shape assessment during exercise.