This retrospective further analysis of the ACTION database evaluated the relationships between baseline blood pressure (BP), on-treatment BP (after 6 weeks) and subsequent cardiovascular outcomes. Analyses were performed using multivariate Cox proportional hazard models. Statistically significant (p < 0.001) and consistent patterns were noted between the risk of major cardiovascular endpoints and both baseline SBP and on-treatment SBP. The lowest risk of debilitating stroke was apparent in those patients with baseline SBP < 120mmHg, with a hazard ratio in this lowest BP group of 0.45 (95% confidence interval 0.28, 0.72), compared to the referent highest BP group (SBP < 150mmHg). Adjusting the model for treatment (nifedipine or placebo) did not modify the conclusions in any statistical or clinically meaningful way. Corresponding and similar results were obtained for pulse pressure but diastolic blood pressure (DBP) was not a consistently useful predictor of outcome. These data confirm the predictive importance of on-treatment SBP (but not DBP) and contribute to the debate about treatment-related BP targets. In this analysis, treatment with nifedipine gastrointestinal therapeutic system in high-risk patients with coronary artery disease was not associated with any increase in cardiovascular risk, even with baseline SBP5120mmHg.
Keywords: Angina; J-curve; blood pressure control; cardiovascular outcomes; nifedipine GITS.