Publication Date



Background: Observational data have described the association of blood pressure (BP) with mortality as ‘J‐shaped’, meaning that mortality rates increase below a certain BP threshold. We aimed to analyse the associations between BP and prognosis in a population of acute myocardial infarction (MI) patients with heart failure (HF) and/or systolic dysfunction. Methods and results: The datasets included in this pooling initiative are derived from four trials: CAPRICORN, EPHESUS, OPTIMAAL, and VALIANT. A total of 28 771 patients were included in this analysis. Arithmetic means of all office BP values measured throughout follow‐up were used. The primary outcome was cardiovascular death. The mean age was 65 ± 11.5 years and 30% were female. Patients in the lower systolic BP (SBP) quintiles had higher rates of cardiovascular death (reference: SBP 121–128 mmHg) [adjusted hazard ratio (HR) 2.49, 95% confidence interval (CI) 2.26–2.74 for SBP ≤112 mmHg, and HR 1.29, 95% CI 1.16–1.43 for SBP 113–120 mmHg]. The findings for HF hospitalization and MI were similar. However, stroke rates were higher in patients within the highest SBP quintile (reference: SBP 121–128 mmHg) (HR 1.38, 95% CI 1.11–1.72). Patients who died had a much shorter follow‐up (0.7 vs. 2.1 years), less BP measurements (4.6 vs. 9.8) and lower mean BP (–8 mmHg in the last SBP measurement compared with patients who remained alive during the follow‐up), suggesting that the associations of low BP and increased cardiovascular death represent a reverse causality phenomenon. Conclusion: Systolic BP values < 125 mmHg were associated with increased cardiovascular death, but these findings likely represent a reverse causality phenomenon.


Mary MacKillop Institute for Health Research

Document Type

Journal Article

Access Rights

ERA Access

Access may be restricted.