Systolic Blood Pressure and Systolic Heart Failure Outcomes
What is the association between systolic blood pressure (SBP) <130 mm Hg and major adverse cardiovascular events in a propensity-matched cohort of older patients with heart failure and reduced ejection fraction (HFrEF)?
This was a retrospective observational study of a cohort of HFrEF patients (defined by left ventricular EF [LVEF] ≤40%) in the Medicare-linked OPTIMIZE-HF data set enrolled between March 2003–December 2004. Patients with a >20 mm Hg difference in SBP from admission to discharge were excluded to avoid measurement errors and impact of inpatient events leading to change in SBP. Propensity matching was used to generate a cohort of patients with SBP <130 mm Hg versus SBP ≥130 mm Hg at discharge. Outcomes assessed included all-cause mortality, all-cause readmission, and HF-related readmission at 30 days and 1 year, and at follow-up.
A total of 2,378 patients were included in the matched cohort, with a mean age of 76 ± 10 years, and 45% were women with mean EF of 28 ± 8%. Thirty-day all-cause mortality occurred in 7% of patients with discharge SBP <130 versus 4% of patients with SBP ≥130 mm Hg (hazard ratio [HR], 1.76; 95% confidence interval [CI], 1.24-2.48). This association persisted at 6-year follow-up (HR, 1.15; 95% CI, 1.04-1.26).
Discharge SBP <130 mm Hg was not associated with 30-day all-cause or HF readmission, but was associated with a higher risk for 12-month and 6-year all-cause or HF readmission. In a sensitivity analysis after excluding patients with SBP <110 mm Hg, higher risk for 30-day and 1-year all-cause mortality persisted in the group with discharge SBP between 110-129 mm Hg compared to those with SBP ≥130 mm Hg but not at 2-, 4-, and 6-year follow-up. When stratified by presence of hypertension, the association between SBP <130 mm Hg and 30-day and 1-year all-cause mortality persisted. However, SBP <130 mm Hg was associated with a higher hazard for 6-year all-cause mortality only for the group without hypertension.
In a propensity-matched cohort of older patients with HFrEF, discharge SBP <130 mm Hg was associated with an increased hazard for all-cause mortality at 30 days extending out to 6 years. This association persisted with exclusion of patients with SBP <110 mm Hg at 1 year, but not with longer follow-up.
Extensive observational literature exists describing an association between low SBP at admission and increased mortality in HFrEF patients, as low SBP in these patients is a marker of advanced disease. Furthermore, randomized trials evaluating use of HF therapy also showed that lower SBP was associated with worse outcomes, but those receiving standard HF medications had lower mortality compared to those not receiving them. This highlights that low SBP should not preclude use of guideline-directed medical therapy for HF.
While this is the first study examining the association between discharge SBP and mortality in an elderly HFrEF cohort, it has several notable limitations. Data on use of HF medication doses were lacking and so were data on SBP and medication use at follow-up. Hence, while the authors did use propensity matching, there remain several unmeasured confounders that could not be matched for since the data were not available in the registry used. Also, it is not representative of contemporary HF therapy, as patients were enrolled in the OPTIMIZE-HF registry from 2003 to 2004. Accordingly, these data cannot be interpreted to represent a causal association between low discharge SBP and mortality.
Nonetheless, it is noteworthy that no randomized trial in hypertensive patients with HFrEF exists, particularly among elderly patients, evaluating the ideal BP target. The SPRINT trial also excluded patients with HF. Current guidelines recommend a goal SBP of <130 mm Hg based solely on expert consensus among patients with hypertension and HF. In the absence of evidence guiding clinicians on the ideal BP target, especially in elderly hypertensive HFrEF patients, use of guideline-directed medical therapy should not be withheld; however, the use of other antihypertensives without a proven mortality benefit in HF patients (e.g., calcium channel blockers, alpha receptor blockers) should be re-evaluated.
Keywords: Antihypertensive Agents, Blood Pressure, Calcium Channel Blockers, Geriatrics, Heart Failure, Hypertension, Inpatients, Metabolic Syndrome X, Patient Readmission, Primary Prevention, Stroke Volume, Systole
< Back to Listings