National and VHA guidelines generally recommend a beta-adrenergic receptor blocker be prescribed for stable patients with systolic heart failure. Two beta-blockers on the national VHA formulary have sufficient evidence to be FDA approved for heart failure, carvedilol and controlled-release sustained acting metoprolol succinate (metoprolol SA). Theoretically, their effectiveness for reducing hospitalization and mortality might differ. Carvedilol has been associated with better outcomes compared to short-acting metoprolol tartrate. Even though tens-of-thousands of veterans with heart failure are being treated with carvedilol or metoprolol SA, the doses being used and relative outcomes have not been studied to be assured that beta-blocker therapy is being used optimally.
The study objectives were to describe the utilization of prescriptions for carvedilol and metoprolol SA and compare time to hospitalization or death among veterans with heart failure.
Cohorts of 17,429 and 8,683 veterans with a prior diagnosis of heart failure who began to receive VA prescriptions for carvedilol or metoprolol SA, respectively, in fiscal years 2000 through 2003 were identified using national data compiled by the VA Pharmacy Benefits Management Strategic Healthcare Group (PBM), the VA National Patient Care Database and Medicare data compiled by the VA Information Resource Center (VIReC). Administrative data were processed to define study variables including demographics, comorbidity, prescription medication use, duration and prescribed doses of carvedilol and metoprolol SA therapy, and time to hospitalization or death due to any cause. Time to hospitalization or death was compared using Cox proportional hazards regression within comparable strata defined by the propensity to prescribe carvedilol rather than metoprolol SA estimated by logistic regression on baseline variables. Sensitivity analyses were conducted to assess potential confounding effects of unmeasured prognostic variables.
The median age of the study cohorts was 74-75 years and they had several comorbidities and concurrent prescription medications. The carvedilol cohort appeared to have a higher percentage with severe heart failure. After a median time on therapy over 1 year, the majority of prescibed daily doses of both carvedilol and metoprol SA were a fraction (25% or less) of proven doses. Older age was independently associated with lower daily doses as was a history of pulmonary disease. Approximately 60% of each cohort was admitted to a hospital or died within 1 year. The propensity score adjusted metoprolol SA to carvedilol hazard ratio (HR) for these outcomes was 0.99 (95% confidence interval 0.96 to 1.03). Mortality after 1 year was 12% and 15% in the metoprolol SA and carvedilol cohorts. The HR for mortality was 0.91 (0.85 to 0.96). However, the observation of better survival in the metoprolol SA cohort was sensitive to plausible differences in unmeasured prognostic variables such as systolic blood pressure, ejection fraction, renal function and severity of symptoms.
Approximately 5% of the veterans who get care from the VHA have heart failure which accounts for nearly 12% of all admissions to VHA hospitals. Annual mortality is over 10%. Beta-blockers can improve these outcomes and are being prescribed for tens-of-thousands of veterans with heart failure. This study suggests that quality improvement efforts could focus on strategies for timely dose titration in hopes of increasing the effectiveness of prescribed beta-blockers. However, more needs to be known about the effectiveness of the doses that are being prescribed and why these doses are being prescribed. The observed lack of explainable differences between metoprolol SA and carvedilol in all-cause hospitalization and mortality suggests that these two beta-blockers are equally effective (or perhaps ineffective) as prescribed.
External Links for this Project
- Rector TS, Anand IS, Nelson DB, Ensrud KE. Carvedilol versus controlled-release metoprolol for elderly veterans with heart failure. Journal of the American Geriatrics Society. 2008 Jun 1; 56(6):1021-7. [view]
- Rector TS. Effectiveness of Controlled-Release Metoprolol versus Carvedilol as Prescribed for Veterans with Heart Failure. 2007 Nov 1. [view]
- Rector T, Nelson DB, Anand I. Effectiveness of Controlled-Release Metoprolol (M) versus Carvedilol (C) as Prescribed in the Veterans Population. Presented at: Heart Failure Society of America Annual Scientific Meeting; 2007 Sep 18; Washington, DC. [view]