Author + information
- Received December 1, 2015
- Revision received April 3, 2016
- Accepted May 2, 2016
- Published online August 1, 2016.
- Jacob P. Kelly, MDa,b,∗ (, )
- Allison Dunning, MScb,
- Phillip J. Schulte, PhDc,
- Mona Fiuzat, PharmDa,b,
- Eric S. Leifer, PhDd,
- Jerome L. Fleg, MDd,
- Lawton S. Cooper, MDd,
- Steven J. Keteyian, PhDe,
- Dalane W. Kitzman, MDf,
- Ileana L. Pina, MDg,
- William E. Kraus, MDa,h,
- David J. Whellan, MDi,
- Christopher M. O'Connor, MDa,b and
- Robert J. Mentz, MDa,b
- aDivision of Cardiology, Department of Medicine, Duke University Medical Center, Durham, North Carolina
- bDuke Clinical Research Institute, Durham, North Carolina
- cDepartment of Health Sciences Research, Mayo Clinic, Rochester, Minnesota
- dDivision of Cardiovascular Science, National Heart, Lung, and Blood Institute, Bethesda, Maryland
- eDivision of Cardiovascular Medicine, Henry Ford Hospital, Detroit, Michigan
- fSection of Cardiology, Department of Internal Medicine, Wake Forest University, Winston-Salem, North Carolina
- gDivision of Cardiology, Department of Medicine, Albert Einstein College of Medicine, New York, New York
- hDuke Molecular Physiology Institute, Durham, North Carolina
- iDivision of Cardiology, Thomas Jefferson University, Philadelphia, Pennsylvania
- ↵∗Reprint requests and correspondence:
Dr. Jacob P. Kelly, Department of Medicine, Duke University Medical Center, 2301 Erwin Road, Durham, North Carolina 27710.
Objectives The aim of this study was to assess for a treatment interaction between statin use and exercise training (ET) response.
Background Recent data suggest that statins may attenuate ET response, but limited data exist in patients with heart failure (HF).
Methods HF-ACTION (Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training) was a randomized trial of 2,331 patients with chronic HF with ejection fraction ≤35% who were randomized to usual care with or without ET. We evaluated whether there was a treatment interaction between statins and ET response for the change in quality of life and aerobic capacity (peak oxygen consumption and 6-min walk distance) from baseline to 3 months. We also assessed for a treatment interaction among atorvastatin, simvastatin, and pravastatin and change in these endpoints with ET. Multiple linear regression analyses were performed for each endpoint, adjusting for baseline covariates.
Results Of 2,331 patients in the HF-ACTION trial, 1,353 (58%) were prescribed statins at baseline. Patients treated with statins were more likely to be older men with ischemic HF etiology but had similar use of renin angiotensin system blockers and beta-blockers. There was no evidence of a treatment interaction between statin use and ET on changes in quality of life or exercise capacity, nor was there evidence of differential association between statin type and ET response for these endpoints (all p values >0.05).
Conclusions In a large chronic HF cohort, there was no evidence of a treatment interaction between statin use and short-term change in aerobic capacity and quality of life with ET. These findings contrast with recent reports of an attenuation in ET response with statins in a different population, highlighting the need for future prospective studies. (Exercise Training Program to Improve Clinical Outcomes in Individuals With Congestive Heart Failure; NCT00047437)
In recently updated lipid guidelines, statin use carries a Class I recommendation for secondary prevention in patients with atherosclerotic cardiovascular disease and for primary prevention in high-risk individuals (1,2). Statin use in heart failure (HF) patients is also common. Statin use has a Class III recommendation for patients with HF without other indications in the most recent U.S. and European HF guidelines (3–5). However, the most common etiology of HF in most developed countries is atherosclerotic cardiovascular disease.
Current guidelines recommend nonpharmacological interventions including aerobic exercise training (ET) and cardiac rehabilitation (CR) in patients with chronic HF with reduced ejection fraction (EF) (4). Although many patients with HF receive statins, and the use of ET interventions are increasingly incorporated into a patient-centered treatment approach, guidelines do not provide specific recommendations for a combined use strategy or appropriate timing for initiating ET and statin therapy (6,7).
Recent data, however, suggest that statins may attenuate the ET response in some populations (8). For example, in overweight or obese patients with metabolic risk factors, statins attenuated the improvement in muscle enzyme mitochondrial levels and cardiorespiratory fitness with ET (8). Consistent with this finding, several small studies have also shown statins to reduce skeletal muscle mitochondrial content and oxidative capacity (9–12). HF-ACTION (Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training) is the largest clinical trial to examine the effect of ET in patients with chronic HF receiving contemporary optimal medical therapy (13). We performed a secondary analysis of HF-ACTION to assess for a treatment interaction between statin use and response to ET with respect to changes in aerobic capacity and health-related quality of life (HR-QoL) from baseline to 3 months.
HF-ACTION was a multicenter, randomized, placebo-controlled trial of usual care with and without an ET regimen in 2,331 patients with stable chronic HF. The study design and results have been previously reported (13,14). The ET intervention included 36 supervised sessions of aerobic exercise over 3 months, followed by home training on a treadmill or stationary cycle for an additional 2 years. The usual care group received self-management educational materials with information on medications, fluid management, and sodium intake and recommendations for 30 min of moderate-intensity physical activity most days of the week as tolerated. Changes in baseline peak oxygen uptake (VO2) and 6-min walk distance (6MWD) were analyzed at 3 months to characterize exercise capacity. HR-QoL was measured using the 23-item self-administered disease-specific Kansas City Cardiomyopathy Questionnaire (KCCQ) at 3 months. Each participating center’s ethics committee or institutional review board approved the study, and all patients gave written informed consent. The present analysis focuses on patients enrolled in HF-ACTION who were on statin therapy at baseline compared with those not on a statin to evaluate for a treatment interaction of statin therapy on ET response.
Patients enrolled in HF-ACTION were required to have stable chronic HF with left ventricular EF ≤35% and New York Heart Association functional class II to IV symptoms while on optimal HF therapy for ≥6 weeks. Patients were excluded if they had major comorbidities or other limitations that would preclude ET, devices that would prevent achievement of target heart rates, or recent (≤6 weeks) or planned (≤6 months) major cardiovascular procedures. For the present analysis, statin use was defined as HMG-CoA reductase inhibitor use at baseline stratified by the specific statin marked on the case report form: atorvastatin, pravastatin, simvastatin, or other. Other lipid-lowering agents were also captured.
Outcomes of interest
The primary outcomes for the present analysis were change in peak VO2 and change in KCCQ from baseline to 3 months, as in previous HF-ACTION analyses, because of the increased missingness at later time points (15,16). The secondary outcome was change in 6MWD from baseline to 3 months.
We performed post-hoc analyses of patients with and without statin treatment at baseline to assess their change in peak VO2, KCCQ, and 6MWD. The baseline characteristics, including demographics, physical and laboratory findings, medical history, and therapies, were summarized as frequencies and percentages for categorical variables and by medians and 25th and 75th percentiles for continuous variables in patients stratified by randomized treatment (ET vs. usual care) and statin use (no statin vs. statin) (Table 1). Comparisons of baseline characteristics across statin use were conducted within each treatment group using Student t tests or Wilcoxon tests for continuous variables and chi-square tests or Fisher exact tests for categorical variables when appropriate.
Linear regression models were used in the analysis to examine the relationship between statin use and change in the individual endpoints from baseline to 3 months. To account for the missing data at 3 months for the outcomes of interest, inverse weighting for missingness was performed. Specifically, subjects in the regression model were weighted by the inverse of their probability of having a nonmissing response similar to recent analyses in HF-ACTION (15,16). This should reduce the bias that would have been present in a complete case analysis by assigning higher weight to observed data subjects who were similar to subjects who were more likely to have missing data (15). Inverse propensity weighting was also used to account for the nonrandom use of statin therapy at baseline (15). The final weight used in the analysis is the inverse of the product of the probability for missingness and the probability for statin use. To investigate for possible confounding variables, the propensity for statin treatment model was selected using backward stepwise variable selection methods (conservatively using p < 0.10 to enter and stay in the model) and further included all variables previously used in multivariable HF-ACTION models (Table 2 footnote) (17). In the weighted linear models, we assessed interactions between randomized treatment and statin use for each outcome. From this model, we report the interaction p value and adjusted effect estimates for statin use within randomized subgroups.
Analysis of variance was used to examine the relationship between type of statin used and the change in exercise capacity and HR-QoL endpoints. Patients were classified into 5 groups: no statin, atorvastatin, simvastatin, pravastatin, and other. If a significant difference in outcomes by statin type was found, multiple comparison (Tukey) analysis was conducted to determine what those differences were.
Adherence to ET (defined as exercise >90 min/week for the first 3 months of the study) was assessed in the subset of patients randomized to ET comparing statin and nonstatin patients. This analysis evaluated for an interaction between statin use and exercise adherence in the ET group. These analyses were weighted for missingness and nonrandom statin use and adjusted for the same covariates as previously mentioned. Statistical significance was assessed using 2-sided p values, with values <0.05 considered statistically significant. All statistical computations were generated using SAS version 9.4 (SAS Institute Inc., Cary, North Carolina).
Statin therapy was used in 1,353 (58%) patients at baseline in HF-ACTION. The baseline characteristics of patients randomized to usual care or ET further dichotomized into those on statin therapy versus those not on statins are presented in Table 1. Patients treated with statins were more likely to be older men with hypertension, an ischemic etiology of HF, peripheral vascular disease, an elevated blood urea nitrogen, and a lower heart rate. Angiotensin-converting enzyme inhibitor/angiotensin receptor blocker use and beta-blocker use were similar between statin and nonstatin patients, whereas mineralocorticoid antagonist use was lower and implanted cardioverter-defibrillator use was higher in the statin-treated patients.
Patients randomized to usual care had higher rates of missing data compared with those randomized to ET at 3 months (Online Table 1). The changes in exercise capacity and HR-QoL from baseline to 3 months stratified by statin use and randomized treatment group are displayed in Tables 2 and 3⇓ (statin type). There was no evidence for a treatment interaction between statin use and ET on changes in quality of life or exercise capacity, nor was there evidence of differential association between statin type and ET response for the endpoints (all p > 0.05). Figure 1 presents statin use and change in peak VO2, 6MWD, and KCCQ from baseline to 3 months as a function of statin use and ET status. Online Figure 1 displays median and interquartile range of the changes in peak VO2, 6MWD, and KCCQ score on the basis of type of statin. There was also no interaction between statin use and exercise adherence with respect to exercise capacity or HR-QoL in patients randomized to ET (interaction p values for peak VO2, 6MWD, and KCCQ of 0.44, 0.80, and 0.60, respectively) (Table 4).
In the largest clinical trial of ET in patients with chronic HF, we evaluated for a treatment interaction between statin use and ET with respect to changes in exercise capacity and HR-QoL. We found that statin use was more prevalent in older men with hypertension, ischemic heart disease, and implanted cardioverter-defibrillators, but these patients had similar use of baseline guideline-directed medical therapy for HF compared with nonstatin patients. We found no evidence that statin treatment attenuated the response to ET in patients with chronic HF and reduced EF. Furthermore, various statins did not have a differential association with changes in quality of life or exercise capacity with ET. Our findings suggest that patients with HF on statin therapy who undergo ET do not have a blunted training response compared with nonstatin patients.
Previous studies suggest that statin use lowers mitochondrial content and oxidative capacity (9–12). Furthermore, atorvastatin impairs exercise-mediated mitochondrial adaptations in skeletal muscle and lowers running capacity in rodents (18,19). Many highly trained athletes are intolerant of statin treatment related to increased muscle enzyme release from skeletal muscle after significant endurance events such as marathons (20–23). However, limited data exist on changes in aerobic capacity in patients on statin therapy (24). A recent study that randomized 37 patients who were overweight or obese and at risk of metabolic syndrome to 12 weeks of aerobic ET with or without simvastatin demonstrated attenuated cardiorespiratory fitness responses and decreased skeletal muscle mitochondrial content in patients treated with simvastatin (8). However, 2 small studies evaluated the effects of 40 mg daily atorvastatin and 80 mg daily simvastatin in healthy subjects on maximal exercise capacity over 8 and 12 weeks of therapy, respectively, demonstrating no effect on maximal oxygen consumption (25,26). A prospective study that evaluated 1,201 patients with coronary artery disease undergoing 12 weeks of CR after an acute cardiac event found no effect of statin therapy on ET response (27). An age/sex-appropriate similar increase in peak oxygen consumption from CR in statin and nonstatin treatment was also found (+3.2 ± 3.7 ml/kg/min and +3.1 ± 3.7 ml/kg/min, respectively) (8,27). The STOMP (Statins on Skeletal Muscle Function and Performance) study randomized 420 healthy patients to evaluate symptoms, muscle strength, and exercise capacity in patients treated with atorvastatin or placebo over 6 months (28). STOMP found that atorvastatin therapy did not decrease endurance, aerobic capacity, or physical activity levels compared with placebo (change in peak VO2: −0.8 ml/kg/min [−1.3 to −0.3 ml/kg/min (post minus baseline)] and −0.8 ml/kg/min [−1.4 to −0.2 ml/kg/min (post minus baseline)], respectively) (28). Although these studies do not include patients with HF, they include patients who were on potent statin therapies for up to 6 months without demonstrating an impaired response to ET. These studies support our findings in the chronic HF population. Interestingly, a study in rats already undergoing ET that then had statin therapy added also found no attenuated response to exercise (29).
In HF-ACTION, ET reduced rehospitalizations and improved quality of life, offering significant benefits on top of guideline medical therapy for HF (13,30,31). With the recent Centers for Medicare and Medicaid expansion of coverage for CR to beneficiaries with HF, many patients are being referred for such care, including a sizeable proportion receiving statins. Our results suggest that there is no apparent blunting of the response to aerobic ET with concomitant statin use. Instead, the expanded CMS coverage for ET should be welcomed as a new therapeutic option to improve exercise capacity and quality of life and to potentially reduce hospitalizations.
This was a retrospective analysis from a clinical trial of ET; thus, cause and effect relationships for statin therapy cannot be directly assessed. Although several small studies have suggested that statin use affects muscle strength, this patient-level data was unavailable for analysis in the HF-ACTION trial. Despite covariate adjustment, other measured and unmeasured variables may have influenced these results. Overadjustment bias may be present, as we do not know the timing of statin initiation relative to variables in the propensity model; thus, future prospective studies are needed to externally validate our findings. Prior statin use was not captured in the HF-ACTION trial. Thus, if a patient discontinued statin therapy due to statin intolerance, whether self-reported or clinically confirmed, prior to enrollment in HF-ACTION, they would be included in a nonstatin group for the current analysis; this would bias our findings toward the null hypothesis given that up to 15% of patients experience statin-related myalgias (32). It is plausible that these patients may be more likely to experience attenuated gains in exercise capacity or quality of life in response to ET when using statins. The inclusion and exclusion criteria of the HF-ACTION trial prevents generalization to the entire HF population, including those with preserved left ventricular EF and the very elderly. A final limitation is the modest increase in aerobic capacity with ET in HF-ACTION, due in part to low training adherence.
The results from this retrospective analysis of the large HF-ACTION trial suggest that statin therapy does not attenuate the improvement in aerobic capacity or quality of life response to ET in patients with chronic HF and reduced EF. Future prospective studies examining the effect of statin therapy on the responses to ET in the HF population are warranted to confirm our findings.
COMPETENCY IN MEDICAL KNOWLEDGE: ET and CR carry Class I and IIA recommendations, respectively, in patients with HF, but are uncommonly prescribed and completed in patients with HF related to various barriers. Our findings suggest that there does not appear to be an attenuated ET response in patients with HF on statin therapy.
TRANSLATIONAL OUTLOOK: Statin use does not blunt the ET response in patients with HF and reduced EF who engage in ET compared with nonstatin users. Future prospective studies are needed to confirm these findings and include patients with HF and preserved EF.
For a supplemental table and figure, please see the online version of this article.
HF-ACTION was funded by the National Institutes of Health. The content of this paper is solely the responsibility of the authors and does not necessarily reflect the views of the National Institutes of Health or the United States Government. Dr. Kelly is supported by grant 5T32 HL 7101-39 from the National Institutes of Health. Dr. Kitzman has received a research grant from Novartis; has served as a consultant for Relypsa, GlaxoSmithKline, Abbvie, Corvia, Merck, and Bayer; and owns stock in Gilead and Relypsa. Dr. Mentz receives research support from the National Institutes of Health (U10HL110312), Amgen, AstraZeneca, Bristol-Myers Squibb, GlaxoSmithKline, Gilead, Novartis, Otsuka, and ResMed; has received honoraria from HeartWare, Janssen, Luitpold Pharmaceuticals, Novartis, ResMed, St. Jude, and Thoratec; and has served on an advisory board for Luitpold Pharmaceuticals, Inc. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. John R. Teerlink, MD, served as Guest Editor for this paper.
- Abbreviations and Acronyms
- cardiac rehabilitation
- exercise training
- heart failure
- health-related quality of life
- Kansas City Cardiomyopathy Questionnaire
- oxygen uptake
- 6-min walk distance
- Received December 1, 2015.
- Revision received April 3, 2016.
- Accepted May 2, 2016.
- American College of Cardiology Foundation
- Stone N.J.,
- Robinson J.G.,
- Lichtenstein A.H.,
- et al.
- Yancy C.W.,
- Jessup M.,
- Bozkurt B.,
- et al.
- McMurray J.J.,
- Adamopoulos S.,
- Anker S.D.,
- et al.
- Mikus C.R.,
- Boyle L.J.,
- Borengasser S.J.,
- et al.
- Dirks A.J.,
- Jones K.M.
- Mentz R.J.,
- Schulte P.J.,
- Fleg J.L.,
- et al.
- O'Connor C.M.,
- Whellan D.J.,
- Wojdyla D.,
- et al.
- Muraki A.,
- Miyashita K.,
- Mitsuishi M.,
- Tamaki M.,
- Tanaka K.,
- Itoh H.
- Bouitbir J.,
- Charles A.L.,
- Rasseneur L.,
- et al.
- Kearns A.K.,
- Bilbie C.L.,
- Clarkson P.M.,
- et al.
- Thompson P.D.,
- Parker B.
- Rengo J.L.,
- Savage P.D.,
- Toth M.J.,
- Ades P.A.
- Parker B.A.,
- Capizzi J.A.,
- Grimaldi A.S.,
- et al.