Exaggerated exercise blood pressure response during treadmill testing as a predictor of future hypertension in men: A longitudinal study

BACKGROUND The purpose of this study was to evaluate receiver operating characteristic curves to identify optimal cutoff values of exercise systolic blood pressure (SBP) using both peak SBP and relative SBP (peak SBP minus resting SBP) as predictors of future hypertension (HTN). METHODS Participants were 3,742 healthy normotensive men who underwent symptom-limited treadmill testing at baseline. Incident HTN was defined as SBP/diastolic blood pressure greater than 140/90mm Hg and/or diagnosed HTN by a physician. RESULTS During an average 5-year follow-up, 364 (9.7%) new cases of HTN were observed. The most discriminatory cutoff values for peak SBP and relative SBP for predicting incident HTN were 181mm Hg (areas under the curve (AUC) = 0.644, sensitivity = 54%, and specificity = 69%) and 52mm Hg (AUC = 0.549, sensitivity = 64.3%, and specificity = 44.6%), respectively. Participants with peak SBP greater than 181mm Hg and relative SBP greater than 52mm Hg had 1.54-fold (95% CI: 1.23-1.93) and 1.44-fold (95% CI: 1.16-1.80) risks of developing HTN after adjusting for potential confounding variables. When these 2 variables were entered simultaneously into the Cox proportional hazards regression model with adjustment for potential confounding variables, only peak SBP (relative risk: 1.39, 95% CI: 1.02-1.89) was a predictor of the development of HTN. CONCLUSIONS The most accurate discriminators for peak and relative SBP during treadmill exercise testing to predict incident HTN were greater than 181 and 52mm Hg, respectively, in normotensive men. A peak SBP greater than 181mm Hg during treadmill exercise testing may provide a useful predictor for the development of HTN in clinical practice.