Time in blood pressure range and cardiovascular outcomes in HFmrEF/HFpEF: a pooled participant-level analysis of four large-scale trials

Background Hypertension is prevalent in patients with heart failure with preserved ejection fraction (HFpEF) and blood pressure (BP) control is one of the only Class I recommendations for the management of HFpEF in the latest AHA/ACC/HFSA guidelines; however, evidence supporting specific systolic BP (SBP) targets in patients with HFpEF remains limited. Aim We investigated the association between BP control and subsequent cardiovascular (CV) outcomes in contemporary trials including patients with HF with mildly reduced EF (HFmrEF)/HFpEF. Methods TOPCAT-Americas, PARAGON-HF, DELIVER, and FINEARTS-HF were randomized clinical trials testing spironolactone, sacubitril/valsartan, dapagliflozin, and finerenone, respectively, versus placebo or an active control (valsartan in PARAGON-HF), in patients with HF and a LVEF >40% (DELIVER), ≥40% (FINEARTS-HF), or ≥45% (TOPCAT-Americas, PARAGON-HF). For each participant, daily BPs were estimated by interpolation from standardized office measurements obtained at prespecified study visits: randomization and 1, 2, 4, 8, and 12 months (TOPCAT-Americas); randomization and 1, 4, 8, and 12 months (PARAGON-HF); randomization and 1 and 12 months (DELIVER); and randomization and 1, 3, 6, 9, and 12 months (FINEARTS-HF). The percentage of time spent within the target SBP range (110–<130mmHg) during the first year after randomization was then estimated. The continuous association between individual-level time-in-target range (TIR) during the first year post-randomization and subsequent risk of the primary outcome of first HF hospitalization (HFH) or CV death, its individual components, and all-cause death was assessed using landmark Cox proportional hazards models with linear splines, adjusted for baseline CV risk factors. Results Among 18,823 patients (mean age 72±9 years; 47% men; mean baseline SBP 129±15 mmHg), the mean percentage of TIR during the first year was 42%, corresponding to ≈153 days. Randomization to active therapies increased the percentage of TIR across trials, by 3% (95%CI −1 to 7%) with spironolactone, 7% (5 to 9%) with sacubitril/valsartan, 2% (0 to 4%) with dapagliflozin, and 4% (2 to 5%) with finerenone. There was a significant association between higher percentage of TIR and lower subsequent risk of the composite outcome (P=0.021). This association was primarily driven by a lower risk of HFH (P=0.007), whereas no significant associations were observed with CV death or all-cause death. Sensitivity analyses applying a stricter range (120-130mmHg) and more liberal ranges (100-130mmHg, 120-140mmHg) produced qualitatively similar findings. TIR accounted for only a limited proportion of mediation of the treatment effects of spironolactone, dapagliflozin, and finerenone on the composite outcome (all less than 5%). Conclusions In patients with HFmrEF/HFpEF, BP control during the first year was associated with a lower subsequent risk of HFH, supporting current guideline recommendations.