Author + information
- ↵⁎Reprint requests and correspondence:
Dr. Lori B. Daniels, Department of Medicine, Division of Cardiology, University of California, San Diego, Mail Code 0986, 9350 Campus Point Drive, Suite 1D, La Jolla, California 92037-1300
One of the great success stories in cardiology is the ability of statins to improve the prognosis in patients at risk of a first or subsequent cardiovascular event. This benefit has been shown across a wide spectrum of patient profiles. The efficacy of statin therapy has been challenged only in high-risk patients such as those on dialysis (1,2), older patients with systolic heart failure (3), and patients with chronic heart failure of any cause (4).
Large retrospective studies, observational studies, and studies using surrogate end points had many believing that the beneficial effects of statins would extend to heart failure patients. This optimism was curtailed with publication of the CORONA (Controlled Rosuvastatin Multinational Trial in Heart Failure) study, the first multicenter, randomized, double-blind, placebo-controlled study assessing the effect of statins in congestive heart failure patients (3). This clinical trial included 5,011 patients who were at least 60 years of age, with New York Heart Association functional class II or greater ischemic heart failure and reduced left ventricular ejection fraction; participants were randomized to receive either 10-mg rosuvastatin daily or placebo and were followed for a median of 2.7 years. Although rosuvastatin decreased low-density lipoprotein cholesterol, it conferred no significant benefit on the primary composite outcome (death from cardiovascular disease, nonfatal myocardial infarction, or nonfatal stroke) or on the secondary individual outcomes of death, any coronary event, and cardiovascular death. As elegantly reviewed by Masoudi (5), these negative results could have been due to a lack of benefit of statins in ischemic heart failure, a rosuvastatin-specific shortcoming, or the advanced age and comorbidities of the population. Rosuvastatin did decrease hospitalizations (for worsening heart failure, p = 0.01; for cardiovascular disease, p < 0.001; and for all-cause hospitalizations, p = 0.007).
Subsequently, the GISSI-HF (Gruppo Italiano per lo Studio della Sopravivenza nell'Insufficienza Cardiaca Heart Failure), a randomized trial of 4,574 heart failure patients followed for a median of 3.9 years, extended the findings of CORONA to patients with chronic heart failure of any cause and irrespective of left ventricular ejection fraction. In this trial, rosuvastatin did not significantly reduce the primary outcome of death or of death or cardiovascular hospitalization (4).
Patients with ischemic heart disease typically derive substantial benefit from statin therapy (6). At some point after the development of heart failure, however, cardiac protection seems to wane, and patients reach a point where their cardiovascular disease is too advanced to modify with statin therapy or competing risks negate benefit from statin treatment. It would be clinically useful to hone in on the transition point along the continuum of ischemic heart disease and heart failure severity where statins are no longer effective at improving most outcomes.
Natriuretic peptides are hormones that are secreted by cardiomyocytes in response to increased ventricular stretch (7). Both B-type natriuretic peptide (BNP) and its N-terminal precursor (NT-proBNP) have significant prognostic value in patients with ischemic heart disease and heart failure (8–10). Therefore, natriuretic peptides should be good candidates for attempting to distinguish lower-risk heart failure patients who may benefit from statins from higher-risk heart failure patients in whom statin therapy is likely to be futile.
This possibility was retrospectively evaluated using data from the Heart Protection Study in which 20,536 subjects at high risk of vascular disease were randomly assigned to either simvastatin 40 mg/day or placebo (11). As expected, baseline NT-proBNP levels were strongly and independently predictive of future vascular events. Higher baseline NT-proBNP levels also were associated with a smaller, but still highly statistically significant, relative risk reduction in events with statin therapy (despite similar absolute reductions in low-density lipoprotein cholesterol). However, because participants with higher NT-proBNP levels had a greater absolute risk of events, the absolute benefits of statin therapy were similar at all levels of NT-proBNP. Because the presence of heart failure was not recorded at baseline, no direct conclusions could be drawn about the utility of statins in patients with and without heart failure.
In this issue of the Journal, Cleland et al. (12) report a post hoc analysis of 3,664 patients from the CORONA study in whom NT-proBNP levels were measured before randomization to either rosuvastatin or placebo. Unlike the Heart Protection Study, all CORONA participants had a history of symptomatic heart failure due to ischemic heart disease with reduced left ventricular ejection fraction. Although rosuvastatin therapy failed to significantly improve primary outcomes overall (3), the investigators found that after stratifying by tertile of NT-proBNP, patients in the lowest tertile (NT-proBNP <868 pg/ml) did benefit from rosuvastatin therapy, with a significant reduction in the primary end point (cardiovascular mortality, nonfatal myocardial infarction, or nonfatal stroke) as well as in 2 pre-specified secondary end points: the composite coronary end point and hospitalizations.
As in the Heart Protection Study, low-density lipoprotein cholesterol was lowered similarly in each tertile of NT-proBNP and did not explain the lack of benefit from rosuvastatin in the upper 2 tertiles. Discontinuation of randomized treatment did increase with increasing values of NT-proBNP, but probably not enough to account for the difference.
In addition to discovering a subgroup of ischemic heart failure patients in whom rosuvastatin was beneficial, the investigators also confirmed the strong prognostic value of NT-proBNP, which was a stronger predictor of each pre-specified outcome than any other variable measured in this cohort. That NT-proBNP is so prognostically powerful lends credence to the hypothesis that lower-risk heart failure patients who may still benefit from statin therapy can be identified and that natriuretic peptides may be better suited to this task than any other readily available clinical parameter. However, as plausible as this hypothesis seems, caution must be exercised in interpreting any post hoc subgroup analysis, especially when the analysis is designed to look for benefit in a subset of overall null results (13). It seems that the use of tertiles for stratification of NT-proBNP levels was not pre-specified, and it is unclear how robust these data would be to other cut points. Validation in other studies and prospective trials using natriuretic peptides with pre-specified cut points to guide statin administration will be essential to fully explore this paradigm.
Other factors that could influence the generalizability of these results pertain to the particulars of the population studied in CORONA. Natriuretic peptide levels are known to rise with increasing age, and the mean age in CORONA was 73 years, with approximately 40% of subjects older than the age of 75. This was also a sick population: in addition to a mean left ventricular ejection fraction of approximately 30%, <25% of the patients in this study had NT-proBNP levels <423 pg/ml. Nonischemic heart failure patients were not studied, and whether natriuretic peptides will be useful to guide statin therapy in them remains unknown.
In addition to these cohort-specific caveats, characteristics of natriuretic peptide levels themselves will influence how these results translate into clinical practice. There is significant intraindividual variability in natriuretic peptide levels that may preclude defining any single cut point as the guide for statin administration in heart failure patients; more likely there will be a cut point with a surrounding “gray zone” where clinical judgment will, as always, play a critical role. Finally, if natriuretic peptides are used to guide statin use in lower-risk heart failure patients, what will be the correct approach when a heart failure patient on statin therapy decompensates? In patients with more advanced heart failure who stand to garner less benefit from statin therapy, the potential risks of statin use plus the associated cost of continued therapy might tilt the risk-benefit ratio toward discontinuation of statin therapy.
How should these results affect our clinical decision making? Clinical practice guidelines recommend that statins be prescribed to patients with ischemic heart disease (14), but do not make heart failure a consideration. If these results are confirmed in other studies, natriuretic peptide levels may have a new application in guiding statin decisions for heart failure patients.
This work was supported in part by a grant from the American College of Cardiology/Guidant Foundation to Dr. Daniels. Drs. Daniels and Barrett-Connor have received research support from Biosite, Inc., and Roche Diagnostics, Inc.
↵⁎ Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
- American College of Cardiology Foundation
- Daniels L.B.,
- Maisel A.S.
- Schnabel R.,
- Lubos E.,
- Rupprecht H.J.,
- et al.
- Doust J.A.,
- Pietrzak E.,
- Dobson A.,
- Glasziou P.
- Emberson J.R.,
- Ng L.L.,
- Armitage J.,
- Bowman L.,
- Parish S.,
- Collins R.
- Cleland J.G.F.,
- McMurray J.J.V.,
- Kjekshus J.,
- et al.,
- CORONA Study Group
- Barrett-Connor E.
- Smith S.C. Jr..,
- Allen J.,
- Blair S.N.,
- et al.