A Hypertensive Response To Exercise Relative To Fitness In Type 2 Diabetes Is Associated With Increased Risk Of Cardiovascular Events And Mortality: Results From The Exercise Stress Test Collaboration (EXERTION)

PURPOSE: A hypertensive response to exercise (HRE) recorded during clinical exercise testing is independently associated with cardiovascular disease (CVD), but its interpretation may be clouded by functional capacity (fitness). Type-2 diabetes (T2DM) is associated with increased prevalence of HRE, but whether this confers additional CVD risk in T2DM when HRE is considered relative to fitness has never been determined and was the aim of this study. METHODS: Clinical exercise test records were analysed from 14,449 people (aged 52.4 ± 13.5 years, 58.7% male) who completed the Bruce treadmill protocol (stages 1-4, and peak) at 6 Australian hospitals. Records (including BP) were linked to administrative health datasets (hospital and emergency admissions, death register) to define clinical characteristics, classify T2DM (n = 1,321) vs. no-T2DM (n = 13,128) and determine the primary outcome of CVD events and death (mean follow-up, 56 ± 33 months, n = 1452 events). To consider exercise systolic BP relative to fitness, systolic BP was divided by peak METs and defined as HRE if values were ≥ 90th percentile for each exercise test stage. Survival analysis using cox-proportional hazards was undertaken to compare outcome rates across strata of those with (+) and without (-) HRE and T2DM. RESULTS: Compared to HRE- and T2DM- (reference), there was a stepwise increase in age and sex adjusted CVD event and death rate across the strata (HRE+ and T2DM-, HRE- and T2DM+, HRE+ and T2DM+), with those HRE+ and T2DM+ showing the highest risk at exercise test stage 1 (HR 1.95, 95%CI: 1.61-2.35), stage 2 (HR 2.05, 95%CI: 1.71-2.47), stage 3 (HR 2.52, 95%CI: 2.12-3.00), stage 4 (HR 3.04, 95%CI: 2.88-3.72) and at peak exercise (HR 1.71, 95%CI: 1.39-2.10). These associations were stronger than if HRE was not considered relative to fitness and persisted when the sample was restricted to those without a history of CVD at baseline (P < 0.01 at stages 1-4 and peak exercise). CONCLUSIONS: When the systolic BP response to exercise testing is considered relative to fitness, those with T2DM and HRE have increased risk of CVD events and death compared to those without. This suggests that clinicians supervising exercise testing should be alert to HRE and impaired functional capacity in people with T2DM and optimise therapy to address heightened CVD risk.