Objectives Peak exercise systolic blood pressure (SBP) is associated with future cardiovascular disease (CVD) and mortality. We aimed to evaluate the predictive value of different SBP patterns at the end of exercise with these outcomes.Methods We studied 6329 adults (45% women) referred for exercise testing, with test duration of 6-14 min, maximal effort and valid SBP measurements at the end of exercise. The two last SBPs were indexed to work rate (mmHg/Watt), defining responses as: drop (negative change), plateau (no change), slow (lower tertile of increase), intermediate (middle tertile) and steep (upper tertile). Data were cross-linked with nationwide disease and mortality registries. Associations with all-cause mortality and incident CVD were analysed using Cox proportional hazards regression (hazard ratio (HR), 95% confidence interval), using slow SBP increase as reference, adjusted for sex, age, body mass index, baseline CVD (mortality analysis only), beta-blockers and exercise capacity (peak Watt).Results The prevalence of SBP responses at the end of exercise were drop (1.1%), plateau (15.0%), slow (30.4%), intermediate (25.2%) and steep increase (28.3%). Follow-up was 8.8 +/- 3.4 years. Compared with a slow increase, the adjusted all-cause mortality risks were not statistically different for a drop (HR 1.16 (0.50-2.65)), plateau (HR 1.19 (0.85-1.66)), intermediate (HR 1.24 (0.93-1.66)) or steep SBP increase (HR 1.16 (0.89-1.52)). CVD risk was increased in those with a SBP drop (HR 3.10 (1.85-5.19), but not significantly for plateau (HR 1.17 (0.92-1.48)), intermediate or steep SBP increases (HRs 0.99-1.00).Conclusion Subjects with a slow SBP increase at the end of exercise tended to have the lowest mortality risk, although no SBP response pattern predicted all-cause mortality independently. CVD risk was strongly increased in patients with a drop in SBP and tended to be increased (non-significantly) also in patients with a plateau in SBP at the end of exercise, in comparison with increasing SBP.