Background
Little is known about associations of calcium channel blockers (CCBs)
with outcomes in patients with heart failure and preserved ejection fraction
(HFpEF).
Methods and Results
Of the 10,570 hospitalized HFpEF patients, ≥65 years, EF
≥40%, in the Organized Program to Initiate Lifesaving
Treatment in Hospitalized Patients with Heart Failure (OPTIMIZE-HF;
2003–2004), linked to Medicare data (through December 31, 2008),
7514 had no prior history of CCB use. Of these, 815 (11%) patients
received new discharge prescriptions for CCBs. Propensity scores for CCB
initiation, calculated for each of the 7514 patients, were used to assemble
a matched cohort of 1620 (810 pairs) patients (mean age, 80 years; mean EF,
56%; 65% women; 10% African American) receiving and
not receiving CCBs, balanced on 114 baseline characteristics. The primary
composite endpoint of all-cause mortality or HF hospitalization occurred in
82% and 81% of patients receiving and not receiving CCBs
(hazard ratio {HR} for CCBs, 1.03; 95% confidence interval {CI},
0.92–1.14). HRs (95% CIs) for all-cause mortality, HF
hospitalization and all-cause hospitalization were 1.05 (0.94–1.18),
1.05 (0.91–1.21), and 1.03 (0.93–1.14), respectively.
Similar associations were observed when we categorized patients into those
receiving amlodipine and non-amlodipine CCBs. Among 7514 pre-match patients,
multivariable-adjusted and propensity-adjusted HRs (95% CI) for
primary composite endpoint were 1.03 (0.95–1.12) and 1.02
(0.94–1.11), respectively.
Conclusions
In hospitalized older HFpEF patients, new discharge prescriptions for
CCBs had no associations with composite or individual endpoints of mortality
or HF hospitalization, regardless of the class of CCBs.