Low but not high exercise systolic blood pressure is associated with long-term all-cause mortality

Abstract

Objectives The risks associated with achieving a high peak systolic blood pressure (SBP) during clinical exercise testing remain controversial, although this issue has not been evaluated in relation to predicted SBP standards. This cohort study aimed to evaluate the long-term risk of all-cause mortality in males in relation to reference values of peak SBP and the increase in SBP during exercise from the Fitness Registry and the Importance of Exercise: A National Database (FRIEND). Methods We followed 7164 males (mean age: 58.2±10.6 years) over 95 998 person-years of follow-up (mean 13.4±5.4 years), who performed a maximal treadmill exercise test at baseline. SBP was measured at rest and at peak exercise. Risk of all-cause mortality over 20 years (Cox regression) was determined in relation to reference percentiles of peak SBP and increase in SBP with exercise: <10th (low), 10th-90th, >90th (high) percentiles. Results A high peak or a large increase in SBP with exercise was not associated with all-cause mortality. Subjects with a low peak SBP had a 20% higher unadjusted risk for all-cause death compared with those with a normal value (1.20 (1.11-1.31)), and a statistically non-significant 7% higher risk after adjustment for all baseline risk factors (1.07 (0.97-1.18)). The corresponding unadjusted and adjusted risks associated with a low increase in SBP were 1.24 (1.15-1.35) and 1.11 (1.02-1.21), respectively. Conclusions A low - but not high - peak SBP is associated with increased unadjusted risk of all-cause mortality. The FRIEND percentiles of exercise SBP can aid clinicians in individualising risk assessment.