Cardiorespiratory fitness attenuates risk for major adverse cardiac ...

Cardiorespiratory fitness attenuates risk for major adverse cardiac events in hyperlipidemic men and women independent of statin therapy: The Henry Ford ExercIse Testing Project Rupert K. Hung, BA, a Mouaz H. Al-Mallah, MD, MSc, a,b Mohamud A. Qadi, BS, a Gabriel E. Shaya, BA, a,c Roger S. Blumenthal, MD, a Khurram Nasir, MD, MPH, a,d Clinton A. Brawner, PhD, e Steven J. Keteyian, PhD, e and Michael J. Blaha, MD, MPH a Baltimore, MD; Riyadh, Saudi Arabia; Miami, FL; and Detroit, MI

Aims

We sought to evaluate the effect of cardiorespiratory fitness (CRF) in predicting mortality, myocardial infarction (MI), and revascularization in patients with hyperlipidemia after stratification by gender and statin therapy.

Methods and results

This retrospective cohort study included 33,204 patients with hyperlipidemia (57 ± 12 years old, 56% men, 25% black) who underwent physician-referred treadmill stress testing at the Henry Ford Health System from 1991 to 2009. Patients were stratified by gender, baseline statin therapy, and estimated metabolic equivalents from stress testing. We computed hazard ratios using Cox regression models after adjusting for demographics, cardiac risk factors, comorbidities, pertinent medications, interaction terms, and indication for stress testing.

Results There were 4,851 deaths, 1,962 MIs, and 2,686 revascularizations over a median follow-up of 10.3 years. In men and women not on statin therapy and men and women on statin therapy, each 1-metabolic equivalent increment in CRF was associated with hazard ratios of 0.86 (95% CI 0.85-0.88), 0.83 (95% CI 0.81-0.85), 0.85 (95% CI 0.83-0.87), and 0.84 (95% CI 0.81-0.87) for mortality; 0.93 (95% CI 0.90-0.96), 0.87 (95% CI 0.83-0.91), 0.89 (95% CI 0.86-0.92), and 0.90 (95% CI 0.86-0.95) for MI; and 0.91 (95% CI 0.88-0.93), 0.87 (95% CI 0.83-0.91), 0.89 (95% CI 0.87-0.92), and 0.90 (95% CI 0.86-0.94) for revascularization, respectively. No significant interactions were observed between CRF and statin therapy (P N .23). Conclusion

Higher CRF attenuated risk for mortality, MI, and revascularization independent of gender and statin therapy in patients with hyperlipidemia. These results reinforce the prognostic value of CRF and support greater promotion of CRF in this patient population. (Am Heart J 2015;170:390-399.e6.)

Higher cardiorespiratory fitness (CRF) has been shown to protect against mortality in both healthy individuals 1-4 and in those with chronic diseases such as coronary artery disease From the aJohns Hopkins Ciccarone Center for the Prevention of Heart Disease, Baltimore, MD, b King Abdul-Aziz Cardiac Center, Riyadh, Saudi Arabia, cUniversity of Miami Miller School of Medicine, Miami, FL, dBaptist Health South Florida, Miami, FL, and eHenry Ford Health System, Detroit, MI. None of the authors have any relevant relations to industry to disclose. The present study was conducted in accordance with the Declaration of Helsinki and was approved by the Henry Ford Health System Institutional Review Board. An oral presentation of this study was presented at the American College of Cardiology Scientific Sessions in Washington, DC, on March 30, 2014. Poster presentations of this study were presented at Cardiovascular Disease Prevention International Symposium in Miami, FL, on February 6, 2014, and at the Johns Hopkins Heart and Vascular Institute's Cardiovascular Research Retreat in Baltimore, MD, on May 30, 2014. Submitted January 21, 2015; accepted April 15, 2015. Reprint requests: Michael J. Blaha, MD, MPH, Johns Hopkins Ciccarone Center for the Prevention of Heart Disease, 600 N Wolfe St, Carnegie 565A, Baltimore, MD 21287. E-mail: [email protected] 0002-8703 © 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ahj.2015.04.030

(CAD), 5-11 diabetes mellitus, 12-16 and hypertension, 17-22 in addition to modulating the relationship between obesity and mortality. 23-25 Cardiorespiratory fitness has also been associated with lower risk for mortality in patients with hyperlipidemia, 26-30 but little is known about the prognostic value of CRF on nonfatal cardiovascular outcomes in this patient population. Management of hyperlipidemia uses multiple modalities including lifestyle modifications comprising physical activity, diet and weight loss, and medications such as statin therapies. 31-36 Notably, the high prevalence of obesity and physical inactivity in developed countries are often linked with hyperlipidemia and increased prescription of statin therapy. Furthermore, statins have wellknown effects on skeletal muscle and, in some studies, have been associated with lower levels of physical activity and an attenuation in the response of CRF to exercise training. 35,37-40 However, statin therapy did not diminish the prognostic importance of CRF on mortality

American Heart Journal Volume 170, Number 2

risk in male veterans with hyperlipidemia, 41 suggesting that CRF may continue to be an important consideration even among those on statin therapy. In addition, previous literature has been mixed on the effect of CRF on mortality between men and women1,42-44 and between veteran and nonveteran cohorts. 6,45 Specifically, whether previous findings in male veterans with hyperlipidemia can be generalized to both men and women with hyperlipidemia seen in routine clinical practice remains an important area of uncertainty, especially given anticipated increases in statin therapy in response to the new cholesterol treatment guidelines.46 We sought to assess the prognostic value of CRF on risks for all-cause mortality, myocardial infarction (MI), and revascularization in a multiethnic, clinically referred cohort of patients with hyperlipidemia, stratified by gender and statin therapy.

Methods Study design This study is based on data from the Henry Ford ExercIse Testing Project (The FIT Project), a retrospective cohort study aimed at investigating the long-term implications of exercise capacity on cardiovascular outcomes and total mortality. 47 The FIT Project is unique in its combined use of (1) directly measured exercise data, (2) retrospective collection of medical history and medication treatment data taken at the time of the stress test, (3) retrospective supplementation of supporting clinical data using the electronic medical record (EMR) and administrative databases, and (4) epidemiologic follow-up for all-cause mortality and select nonfatal outcomes via linkage with the death registry and medical claims files, respectively. The FIT Project population is a registry of 69,885 consecutive patients who underwent physician-referred treadmill stress testing at Henry Ford Health System in metropolitan Detroit, MI, between 1991 and 2009. These medical centers are part of a large, vertically integrated organization that provides health care and offers a managed care insurance plan. Data from the exercise test, medical history, and medications were collected by exercise physiologists and nurses and entered at the time of testing into a common clinical reporting tool that directly populated the EMR. Supporting clinical data and follow-up for cardiovascular outcomes were derived from the EMR and administrative databases shared across Henry Ford Health System. The FIT Project was approved by the Henry Ford Health System Institutional Review Board. Study population We initially included all patients from The FIT Project who had established hyperlipidemia at the time of stress testing (n = 34,081). Hyperlipidemia was defined as self-reported diagnosis, an EMR problem list-based diagnosis, low-density

Hung et al 391

lipoprotein cholesterol N160 mg/dL on laboratory testing closest to the date of the stress test, or use of medications for hyperlipidemia. Patients missing covariates of interest were excluded (n = 887), leaving 33,204 patients with hyperlipidemia for analysis. Patients were subsequently categorized according to both gender and baseline statin therapy: men not on statin therapy (n = 10,210), men on statin therapy (n = 8,220), women not on statin therapy (n = 8,650), and women on statin therapy (n = 6,124).

Exercise testing All patients underwent routine, clinically referred, symptom-limited treadmill stress testing following the standard Bruce protocol. 48 For individuals with repeat stress testing, only the results from the first test were considered in the registry. Patients b18 years old at the time of stress testing and patients undergoing modified Bruce and non-Bruce protocol tests were not included in the registry. In accordance with clinical guidelines, 49 treadmill testing was terminated at the discretion of the supervising clinician for reasons that included significant arrhythmias, abnormal hemodynamic responses, diagnostic ST-segment changes, exercise-limiting symptoms such as chest pain or shortness of breath, or if the patient was unwilling or unable to continue. Resting heart rate and blood pressure were taken before stress testing by clinical personnel. Target heart rate was calculated as 85% of the age-predicted maximal heart rate determined by the formula: 220 − age. Cardiorespiratory fitness, expressed in estimated metabolic equivalents (METs), was calculated by the treadmill controller system (Q-Stress; Quinton Instruments, Bothell, WA) based on achieved speed and elevation and was further categorized into 4 groups (b6, 6-10, 10-12, or N12 METs) for select analyses. Medical history and medication use A medical history including age, gender, race, indication for testing, risk factor burden, active medication use, and medical history was obtained by trained nurses and/ or exercise physiologists immediately before the stress test. Race and smoking were defined exclusively by self-report. Obesity was defined by self-report and/or assessment by the clinician historian. Family history of coronary artery disease was defined as compatible history in a first degree relative. Indication for stress testing was extracted from the stress test requisition provided by the referring physician and subsequently categorized into common indications (ischemia evaluation/risk stratification, chest pain, shortness of breath, preoperative evaluation, etc). Medication use and medical history were gathered by self-report at the time of testing and then supplemented by a retrospective verification using the EMR, administrative databases, and/or pharmacy claims files from enrollees in the integrated health plan. A database-verified diagnosis was considered present when the appropriate International

American Heart Journal August 2015

392 Hung et al

Classification of Diseases, Ninth Edition, code was present on ≥3 separate encounters within the health system. Diabetes mellitus and hypertension were defined as a self-reported prior diagnosis, a database-verified diagnosis, and/or use of medications for the respective medical conditions. β-Blockers and angiotensin-converting enzyme (ACE) inhibitors were considered antihypertensive medications for this analysis. Prior atrial fibrillation and congestive heart failure were defined as prior clinical diagnosis of at least paroxysmal atrial fibrillation or systolic or diastolic heart failure, respectively.

Laboratory testing Laboratory results were identified (as available) through a retrospective search of the laboratory databases. For each patient and for each laboratory value, the test performed closest to the date of the stress test was selected for inclusion. Follow-up and event adjudication Patients were followed up for a mean of 10.8 ± 4, 5.8 ± 4, and 5.7 ± 4 years for the occurrence of all-cause mortality, MI, and subsequent revascularization, respectively. Mortality was ascertained in April 2013, after federal law changes in 2011 limited reporting of certain deaths by state agencies. 50 An algorithmic search of the Social Security Death Index Death Master File was completed using social security number, first name, last name, and date of birth data. Myocardial infarctions and subsequent revascularizations were ascertained in May 2010 through linkage with administrative claims files from services delivered by the affiliated group practice and/or reimbursed by the health plan. Linkage was performed using appropriate International Classification of Diseases, Ninth Edition, and Current Procedural Terminology codes for MI, percutaneous coronary intervention (PCI), and coronary artery bypass graft (CABG). To limit biases associated with loss to follow-up, patients were censored for nonmortality outcomes at their last contact with the integrated Henry Ford Health System group practice when ongoing coverage with the health plan could no longer be confirmed. Statistical analysis Groups were compared using χ 2 testing or analysis of variance techniques, as appropriate. The unadjusted cumulative incidence for mortality, MIs, and subsequent revascularizations were derived from Kaplan-Meier estimates at median follow-up times for each outcome, equivalent to 10.3, 5.2, and 5.0 years for mortality, MI, and subsequent revascularization, respectively. Results were stratified by MET category, gender, and baseline statin therapy. We used Cox regression models to calculate hazard ratios (HRs) for mortality, MIs, and subsequent revascularizations with adjustment for age; gender; race; resting heart rate; resting systolic and

diastolic blood pressures; history of hypertension, obesity, smoking, diabetes mellitus, atrial fibrillation, heart failure, CAD, MI, PCI, or CABG; family history of CAD; medications used to treat chronic obstructive pulmonary disease (COPD), hypertension, or hyperlipidemia; aspirin; β-blockers; potential interaction between obesity and statin therapy; and indication for stress testing. In examining the interactive effect between CRF and statin therapy, analyses were additionally adjusted for a propensity score modeled as the propensity to be on statin therapy. The propensity score was derived from logistic regression models adjusting for age; gender; history of diabetes, hypertension, CAD, PCI, or CABG; use of antihypertensive medications; β-blockers; angiotensin receptor blockers; ACE inhibitors; aspirin; and date of stress testing to account for secular trends in the prescription of statin therapy. The propensity score models enable assessment of the survival benefits from the lipid-lowering effects and other pleotropic mechanisms associated with statin therapy. Cox regression models were further used for sensitivity analyses and evaluation of potential interactions between CRF, gender, and statin therapy. Statistical analyses were performed using Stata (version 13.1; 2014; StataCorp, College Station, TX). No extramural funding was used to support this work. The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the manuscript, and its final contents.

Results Baseline characteristics Table I shows baseline characteristics and stress test results for the study population (n = 33,204), which included 14,774 (44%) women and 8,260 (25%) blacks. Online Appendix Supplementary Table I shows the lipid profile for the study population. Online Appendix Supplementary Table II shows the baseline characteristics by MET categories. Mean age of the study population was 57 ± 12 years. At baseline, 14,344 (43%) patients with hyperlipidemia were on statin therapy, of which 8,220 (57%) were men and 6,124 (43%) were women. Compared to women, men in our cohort were more likely to be white and smokers and to have a history of CAD, prior MI, PCI, and CABG (P b .001). Compared to men, women in our cohort were more likely to be black and obese and to have a family history of CAD and undergo stress testing due to chest pain (P b .001). Patients on statin therapy tended to be older, obese, and on other medication regimens and to have a history of diabetes, hypertension, CAD, prior MI, PCI, and CABG (P b .001). Mean METs achieved in the total cohort was 8.7 ± 3 METs. In general, men and patients not on statin therapy had higher CRF than women and those on statin therapy, respectively (P b .001 for both). Patients in the lower MET categories were more likely to be older and women,


Hung et al 393

Table I. Baseline characteristics Men

Demographic data Age (y) Race White Black Other Physiological measurements Resting heart rate (beat/min) Resting systolic blood pressure (mm Hg) Resting diastolic blood pressure (mm Hg) Medical history Obesity (BMI ≥30 kg/m 2) Current smoker Hx hypertension Hypertension medications ACE inhibitor Angiotensin receptor blocker Calcium-channel blocker Hx diabetes mellitus β-Blocker Hx MI Hx CAD Hx PCI Hx CABG Family Hx of CAD Hx atrial fibrillation Hx heart failure Nonstatin lipid-lowering drug COPD medications Aspirin Indication for stress test Chest pain Shortness of breath Established CAD Preoperation Stress test results Mean METs achieved b6 METs 6-10 METs 10-12 METs N12 METs

Women

Total cohort (n = 33,204)

No statin (n = 10,210)

Statin (n = 8220)

No statin (n = 8650)

Statin (n = 6124)

57.0 ± 12

53.7 ± 12

58.8 ± 11

56.5 ± 12

60.6 ± 11

69% 25% 6%

73% 20% 7%

72% 21% 7%

64% 30% 6%

64% 31% 5%

72 ± 13 132 ± 19 81 ± 10

72 ± 12 132 ± 18 83 ± 10

70 ± 13 133 ± 19 81 ± 10

75 ± 12 132 ± 20 80 ± 10

74 ± 13 133 ± 20 79 ± 10

b.001 b.001 b.001

24% 43% 73% 55% 24% 4% 16% 24% 27% 14% 18% 7% 5% 52% 3% 2% 7% 9% 27%

18% 46% 62% 40% 16% 2% 12% 16% 17% 11% 15% 4% 4% 49% 2% 1% 8% 6% 21%

25% 49% 84% 72% 38% 6% 18% 32% 41% 25% 33% 15% 10% 48% 5% 2% 9% 9% 40%

24% 37% 67% 47% 15% 3% 14% 17% 19% 7% 8% 2% 2% 57% 2% 1% 5% 10% 17%

33% 38% 84% 70% 31% 7% 20% 34% 36% 14% 18% 8% 4% 56% 3% 2% 7% 13% 31%

b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001

44% 10% 8% 2%

43% 12% 7% 1%

35% 14% 9% 2%

55% 9% 4% 2%

45% 9% 8% 2%

b.001 b.001 b.001 b.001

8.7 ± 3 17% 30% 35% 18%

9.9 ± 3 10% 20% 36% 34%

8.9 ± 3 14% 29% 37% 20%

8.0 ± 3 21% 35% 36% 8%

7.3 ± 2 27% 42% 27% 4%

b.001 b.001 b.001 b.001 b.001

P⁎

b.001 b.001

Baseline characteristics and stress test results for the study population. Abbreviation: Hx, history. ⁎ P value between statin-gender groups.

were more likely to have comorbidities and be on medications, and less likely to undergo stress testing due to chest pain (P b .001 for all).

Unadjusted survival During follow-up, 4,851 deaths (15%), 1,962 MIs (6%), and 2,686 revascularizations (8%) were observed. Table II shows the crude cumulative incidence of each outcome at the respective median follow-up time. Online Appendix Supplementary Figures 1 to 3 show the associated Kaplan-Meier survival functions for each outcome by MET categories within each statin-gender group. In each

statin-gender group, there were fewer deaths, MIs, and subsequent revascularizations in higher MET categories (P b .001). In general, within each MET category, there were more events in men and those on statin therapy than in women and those not on statin therapy (P b .001).

Adjusted survival The Figure shows the association between CRF and adjusted risk for mortality, MI, and subsequent revascularization. Men and women in increasing MET categories had markedly lower risk for mortality, regardless of statin therapy (P b .001). Higher MET categories were also


394 Hung et al

Table II. Crude cumulative incidences at median follow-up All-cause mortality

b6 METs

6-10 METs

10-12 METs

N12 METs

Men not taking statins Men taking statins Women not taking statins Women taking statins

41% (37%-44%) 41% (38%-44%) 24% (22%-27%) 23% (21%-26%)

16% 19% 7% 9%

(15%-18%) (17%-20%) (6%-8%) (8%-10%)

6% 9% 2% 5%

(5%-7%) (8%-11%) (2%-3%) (4%-6%)

3% 3% 1% 1%

(2%-3%) (2%-4%) (1%-3%) (0%-6%)

Future MI Men not taking statins Men taking statins Women not taking statins Women taking statins

11% (9%-14%) 14% (12%-17%) 7% (6%-8%) 10% (8%-12%)

6% 9% 2% 4%

(5%-7%) (8%-10%) (2%-3%) (3%-5%)

2% 5% 1% 1%

(2%-3%) (4%-6%) (0%-1%) (1%-2%)

1% 2% 1% 2%

(1%-2%) (2%-4%) (0%-2%) (1%-5%)

Subsequent revascularizations Men not taking statins Men taking statins Women not taking statins Women taking statins

15% (13%-17%) 20% (17%-22%) 8% (7%-10%) 12% (10%-14%)

13% 16% 3% 6%

(11%-14%) (14%-17%) (3%-4%) (5%-7%)

5% 10% 1% 3%

(4%-6%) (9%-11%) (1%-2%) (2%-4%)

2% 5% 0% 2%

(2%-3%) (4%-6%) (0%-1%) (1%-7%)

Crude cumulative incidence of all-cause mortality, future MIs, and subsequent revascularizations at median follow-up times of 10.3, 5.2, and 5.0 years, respectively, in increasing METs categories within statin-gender groups. Kaplan-Meier estimates with 95% CIs are shown. Differences between METs categories within statin-gender groups were all significant (P b .001) for all outcomes.

associated with lower risk for MI and subsequent revascularization in the total cohort, but statistical significance was not always reached in the lower MET categories for men and the highest MET category for women (online Appendix Supplementary Table III). Interactions between CRF and statin therapy were nonsignificant (P N .23). Table III shows the individual effect of each 1-MET increment in CRF on outcomes, with stratification by baseline history of CAD. Notably, CAD attenuated the association between CRF and the risk for mortality, MI, and subsequent revascularization. For mortality, greater per-MET risk reductions among those on statin therapy were observed only in men without CAD. Greater per-MET risk reductions in women compared to men were noted only among those with CAD. For MIs and revascularizations, higher METs were protective in all patients except in those with CAD who were not on statin therapy.

Interactive effects between CRF and statin therapy Table IV shows the interactive effect of CRF and statin therapy on mortality risk in both men and women, relative to those with b6 METs and not on statin therapy, with stratification by baseline history of CAD. Higher levels of CRF were associated with lower risk for mortality in both men and women, independent of statin therapy and CAD history. In the total cohort, before considering CRF, statin therapy was associated with a 12% relative reduction in mortality risk (HR 0.88, 95% CI 0.82-0.94), without significant effect modification by gender. With stratification by MET categories, risk reductions associated with statin therapy appeared greatest in men and women with lower levels of CRF without baseline CAD.

Sensitivity analyses Sensitivity analyses were independently conducted (i) excluding those who died within 1 year of their stress test (n = 200), (ii) excluding those who underwent preoperative stress testing (n = 602), (iii) excluding those unable to achieve 85% of their age-predicted maximal heart rate (n = 8,993), (iv) excluding those with hyperlipidemia diagnosed solely through laboratory values (n = 2,852), and (v) with further adjustment for the decade of stress testing to account for temporal changes in management of hyperlipidemia and other comorbidities, with no significant changes to our main findings. Sensitivity analyses on those with body mass index (BMI) data (n = 10,237) showed that BMI attenuated the individual effect of CRF and the interactive effect of CRF and statin therapy in certain subgroups, but without significant changes to the overall findings regarding CRF (online Appendix Supplementary Tables IV and V). Notably, within the BMI sensitivity analyses, CRF was one of the strongest determinants of overall survival (z = −13.2), only behind age in significance (z = 17.7). The contribution of obesity to prognosis was modest in comparison (z = −3.0) compared to that of CRF and other traditional cardiovascular risk factors. Overall, CRF remained a strong predictor of outcomes in all sensitivity analyses performed.

Discussion Higher CRF was associated with markedly lower risk for mortality, MI, and subsequent revascularization in this multiethnic cohort of men and women with hyperlipidemia seen in routine clinical practice. The prognostic value of CRF was not diminished by baseline statin therapy, suggesting continued importance of CRF in all patients with hyperlipidemia.


Hung et al 395

Figure

Effect of CRF on outcomes. Adjusted HRs by METs categories in the total cohort and each statin-gender group. Hazard ratios were adjusted for age; gender; race; resting heart rate and systolic and diastolic blood pressures; history of hypertension, dyslipidemia, obesity, smoking, diabetes mellitus, atrial fibrillation, heart failure, CAD, MI, PCI, or CABG; family history of CAD; medications used to treat COPD, hypertension, or hyperlipidemia; aspirin; clopidogrel; β-blockers; potential interaction between obesity and statin therapy; and indication for stress testing. ⁎P b .05.

Cardiorespiratory fitness and risk for mortality Higher CRF was associated with greater survival in both men and women in our cohort, with slight effect modification from statin therapy observed only in men without CAD. Statin therapy was associated with a 12% relative survival benefit in our cohort—consistent with current literature 36—but after consideration for CRF, improved survival was observed mostly in patients without CAD with lower levels of CRF. Our results contrast to the findings in male veterans with dyslipidemia 41 in which a synergistic effect between higher levels of CRF and statin therapy was reported in all patients.

These discrepancies in findings may be due to differences between veteran and nonveteran populations—namely, different body compositions, disease burdens, medical therapies, medication adherences, and accessibility to health care—and our additional consideration for the secular trend in statin prescriptions. Overall, the importance of higher CRF in both men and women was not attenuated by statin therapy. Survival benefits associated with statin therapy may be modulated by CRF, findings with potential public health implications.


396 Hung et al

Table III. Effect of CRF on mortality and cardiovascular outcomes: per-MET analysis Total cohort

Mortality

MI

Revascularization

Total cohort (n = 33,204) Men not taking statins (n = 10,210) Men taking statins (n = 8220) Women not taking statins (n = 8650) Women taking statins (n = 6124)

0.85 (0.84-0.86) 0.86 (0.84-0.88) 0.85 (0.83-0.87) 0.83 (0.81-0.85) 0.84 (0.81-0.87)

0.91 0.93 0.89 0.87 0.90

(0.89-0.92) (0.90-0.96) (0.86-0.92) (0.83-0.91) (0.86-0.95)

0.90 (0.88-0.91) 0.91 (0.88-0.93) 0.89 (0.87-0.92) 0.87 (0.83-0.91) 0.90 (0.86-0.94)

No established CAD Patients without established CAD (n = 27,191) Men not taking statins (n = 8715) Men taking statins (n = 5529) Women not taking statins (n = 7919) Women taking statins (n = 5028)

0.85 (0.83-0.86) 0.86 (0.84-0.88) 0.82 (0.79-0.85) 0.83 (0.80-0.86) 0.86 (0.82-0.90)

0.87 0.89 0.84 0.82 0.92

(0.84-0.89) (0.85-0.93) (0.79-0.89) (0.77-0.87) (0.84-1.00)

0.88 (0.86-0.90) 0.88 (0.86-0.91) 0.87 (0.84-0.91) 0.85 (0.80-0.89) 0.88 (0.82-0.95)

Established CAD Patients with established CAD (n = 6013) Men not taking statins (n = 1495) Men taking statins (n = 2691) Women not taking statins (n = 731) Women taking statins (n = 1096)

0.87 (0.85-0.88) 0.88 (0.86-0.91) 0.87 (0.84-0.89) 0.84 (0.79-0.88) 0.83 (0.78-0.88)

0.95 0.99 0.92 0.96 0.91

(0.93-0.98) (0.95-1.04) (0.89-0.96) (0.89-1.04) (0.85-0.98)

0.94 (0.91-0.96) 0.97 (0.93-1.01) 0.91 (0.88-0.94) 0.95 (0.88-1.02) 0.93 (0.87-0.99)

Adjusted HRs for mortality, MIs, and revascularizations associated with each 1-MET increment in CRF, with stratification by presence of established CAD at baseline. 95% CIs are shown in parentheses. Hazard ratios were adjusted for age; gender; race; resting heart rate and systolic and diastolic blood pressures; history of hypertension, dyslipidemia, obesity, smoking, diabetes mellitus, atrial fibrillation, heart failure, CAD, MI, PCI, or CABG; family history of CAD; medications used to treat COPD, hypertension, or hyperlipidemia; aspirin; clopidogrel; β-blockers; potential interaction between obesity and statin therapy; and indication for stress testing.

Cardiorespiratory fitness and risk for MI and revascularization Higher CRF was generally associated with lower risk for both MI and revascularization in our cohort, with significant effect modification from statin therapy observed only in patients with CAD. Among patients with CAD, CRF was inversely associated with MIs and revascularizations only in those on statin therapy, suggesting important roles for both higher CRF and statin therapy in patients with hyperlipidemia and CAD. Overall, our results suggest that higher CRF has a protective effect beyond statin therapy against MIs and revascularizations in patients with hyperlipidemia. Implications of low CRF In The FIT Project, patients with the lowest CRF tended to have the highest risk for mortality, MIs, and subsequent revascularizations regardless of baseline statin therapy. Statins are effective medications widely used in the treatment of hyperlipidemia and prevention of CAD and mortality 31,36,51; however, our findings suggest that statin therapy does not reduce the importance of higher CRF in both men and women with hyperlipidemia. Rather, CRF remained the strongest modifiable prognostic indicator for long-term survival in our cohort, surpassed in prognostic significance only by older age. Although obesity is intricately linked to hyperlipidemia, obesity was comparatively only a mild contributor to overall prognosis, in accordance with results from other studies. 23 Cardiorespiratory fitness has been reported to significantly modulate the association between obesity and mortality 23-25; our findings suggest a continued need to incorporate CRF into discussions of the so-called obesity paradox.

As physical activity can increase CRF, 52-56 our findings support continued consideration and promotion of physical activity and fitness-based interventions in all patients with hyperlipidemia, independent of gender and statin therapy. Notably, the greatest differences in mortality risk were observed in patients with the lowest levels of CRF, suggesting that even modest improvements in CRF in this group may lead to substantial benefits in overall survival. Given current literature on statin therapy, myalgia, and decreased physical activity, 35,37-40 the interface between CRF and statin therapy on long-term outcomes remains an important area in need of further study.

Limitations The main limitations of this study are the lack of longitudinal data regarding changes in CRF and comorbidity burden, in addition to date of initiation, duration, dosage, and treatment response among those on statin therapy. Furthermore, whether statins were used or prescribed may depend on factors not fully accounted for in this study, such as personal choice, adequate cholesterol control through lifestyle modifications, statin intolerance, use of other lipid-lowering medications, and extent and severity of underlying CAD. Similarly, the decision to undergo revascularization procedures in the follow-up period may have been confounded by factors such as frailty, morbidity burden, and recommendations by physicians based on the stress test results. Caution should be used when interpreting the interactive effects between CRF, statin therapy, and mortality risk, as factors influencing the propensity to be on statin therapy and


Hung et al 397

Table IV. Interactive effect of CRF and statin therapy on mortality Men No CAD

Statin therapy

Women No statin therapy

Statin therapy

No statin therapy

b6 METs 6-10 METs 10-12 METs N12 METs

1.00 0.48 0.23 0.10

CAD

Statin therapy

No statin therapy

Statin therapy

No statin therapy


1.00 0.52 0.35 0.18

1.05 0.56 0.38 0.19

1.00 (ref) 0.47 (0.36-0.63) 0.31 (0.19-0.49) –

1.04 0.60 0.30 0.17

(ref) (0.39-0.58) (0.19-0.29) (0.07-0.15)

(ref) (0.44-0.62) (0.29-0.44) (0.13-0.27)

1.16 0.56 0.23 0.12

(0.95-1.42) (0.46-0.68) (0.19-0.29) (0.09-0.16)

(0.88-1.25) (0.46-0.68) (0.30-0.48) (0.12-0.30)

1.00 0.47 0.23 0.07

(ref) (0.38-0.58) (0.17-0.32) (0.02-0.30)

1.22 0.50 0.20 0.11

(1.02-1.47) (0.41-0.61) (0.16-0.26) (0.06-0.19)

(0.84-1.27) (0.45-0.81) (0.18-0.50) (0.02-1.24)

Interactive effect of cardiorespiratory fitness and statin therapy on risk for mortality in the total cohort, stratified by gender and baseline CAD. Men and women on statin therapy who achieved b6 METs served as the respective referent groups. 95% CIs are shown in parentheses. Hazard ratios were adjusted for the propensity to be on statin therapy; race; resting heart rate; resting systolic and diastolic blood pressures; history of obesity, smoking, MI, heart failure, or atrial fibrillation; family history of CAD; use of other lipid-lowering agents; calcium-channel blockers; medications used to treat COPD; potential interaction between obesity and statin therapy; and indication for stress testing.

other unmeasured confounders including diet and socioeconomic status may not be fully accounted for in our study as described above. Cardiorespiratory fitness was measured using the Bruce treadmill protocol, which may overestimate true CRF. 57 In addition, statin therapy can impair physical fitness, 38 and many patients on statin therapy report exertional myalgias and fatigue, which may detrimentally affect their ability to undergo stress testing. 37 Cardiorespiratory fitness is also known to be partly determined by genetics 58,59; the extent to which genetics may explain our findings is unknown. As with all cohort studies, a causal relationship between higher CRF and better outcomes cannot be concluded. Despite gender and racial diversity in our cohort, our study may have socioeconomic, geographic, selection, or survivorship bias due to our inclusion/exclusion criteria, for example, asymptomatic patients and those unable to undergo exercise testing would have been less likely to be included. Lastly, HRs were estimated using stress test data that may not accurately represent a patient's true CRF due to presence of symptomatic complaints such as chest pain and shortness of breath. However, as this is often how patients present in clinical practice, we argue that these findings are of clinical relevance.

Conclusion Higher CRF was associated with lower risk for mortality, MIs, and subsequent revascularizations irrespective of gender and statin therapy in this cohort of patients with hyperlipidemia seen in routine clinical practice. The prognostic value of CRF was not diminished with statin therapy, suggesting that fitness remains an important consideration in all patients with hyperlipidemia, including those on statin therapy.

References 1. Kodama S, Saito K, Tanaka S, et al. Cardiorespiratory fitness as a quantitative predictor of all-cause mortality and cardiovascular events in healthy men and women: a meta-analysis. JAMA 2009;301: 2024-35. 2. Barlow CE, Defina LF, Radford NB, et al. Cardiorespiratory fitness and long-term survival in “low-risk” adults. J Am Heart Assoc 2012;1: e001354. 3. Farrell SW, Fitzgerald SJ, McAuley PA, et al. Cardiorespiratory fitness, adiposity, and all-cause mortality in women. Med Sci Sports Exerc 2010;42:2006-12. 4. Kaminsky L, Arena R, Beckie T, et al. The importance of cardiorespiratory fitness in the United States: the need for a national registry: a policy statement from the American Heart Association. Circulation 2013;127: 652-62. 5. Keteyian SJ, Brawner CA, Savage PD, et al. Peak aerobic capacity predicts prognosis in patients with coronary heart disease. Am Heart J 2008;156:292-300. 6. Goel K, Thomas RJ, Squires RW, et al. Combined effect of cardiorespiratory fitness and adiposity on mortality in patients with coronary artery disease. Am Heart J 2011;161:590-7. 7. Martin BJ, Arena R, Haykowsky M, et al. Cardiovascular fitness and mortality after contemporary cardiac rehabilitation. Mayo Clin Proc 2013;88:455-63. 8. Vanhees L, Fagard R, Thijs L, et al. Prognostic significance of peak exercise capacity in patients with coronary artery disease. J Am Coll Cardiol 1994;23:358-63. 9. Kavanagh T, Mertens DJ, Hamm LF, et al. Peak oxygen intake and cardiac mortality in women referred for cardiac rehabilitation. J Am Coll Cardiol 2003;42:2139-43. 10. Kavanagh T, Mertens DJ, Hamm LF, et al. Prediction of long-term prognosis in 12 169 men referred for cardiac rehabilitation. Circulation 2002;106:666-71. 11. Hung RK, Al-Mallah MH, McEvoy JW, et al. Prognostic value of exercise capacity in patients with coronary artery disease: The FIT (Henry Ford ExercIse Testing) Project. Mayo Clin Proc 2014;89: 1644-54. 12. McAuley PA, Myers JN, Abella JP, et al. Exercise capacity and body mass as predictors of mortality among male veterans with type 2 diabetes. Diabetes Care 2007;30:1539-43.

398 Hung et al

13. Church TS, LaMonte MJ, Barlow CE, et al. Cardiorespiratory fitness and body mass index as predictors of cardiovascular disease mortality among men with diabetes. Arch Intern Med 2005;165: 2114-20. 14. Lyerly GW, Sui X, Lavie CJ, et al. The association between cardiorespiratory fitness and risk of all-cause mortality among women with impaired fasting glucose or undiagnosed diabetes mellitus. Mayo Clin Proc 2009;84:780-6. 15. Nylén ES, Faselis C, Kheirbek R, et al. Statins modulate the mortality risk associated with obesity and cardiorespiratory fitness in diabetics. J Clin Endocrinol Metab 2013;98:3394-401. 16. McAuley PA, Artero EG, Sui X, et al. Fitness, fatness, and survival in adults with prediabetes. Diabetes Care 2014;37:529-36. 17. Kokkinos P, Faselis C, Myers J, et al. Statin therapy, fitness, and mortality risk in middle-aged hypertensive male veterans. Am J Hypertens 2014;27:422-30. 18. Evenson KR, Stevens J, Thomas R, et al. Effect of cardiorespiratory fitness on mortality among hypertensive and normotensive women and men. Epidemiology 2004;15:565-72. 19. Faselis C, Doumas M, Panagiotakos D, et al. Body mass index, exercise capacity, and mortality risk in male veterans with hypertension. Am J Hypertens 2012;25:444-50. 20. McAuley PA, Sui X, Church TS, et al. The joint effects of cardiorespiratory fitness and adiposity on mortality risk in men with hypertension. Am J Hypertens 2009;22:1062-9. 21. Kokkinos P, Manolis A, Pittaras A, et al. Exercise capacity and mortality in hypertensive men with and without additional risk factors. Hypertension 2009;53:494-9. 22. Church TS, Kampert JB, Gibbons LW, et al. Usefulness of cardiorespiratory fitness as a predictor of all-cause and cardiovascular disease mortality in men with systemic hypertension. Am J Cardiol 2001;88:651-6. 23. Barry VW, Baruth M, Beets MW, et al. Fitness vs. fatness on all-cause mortality: a meta-analysis. Prog Cardiovasc Dis 2014;56:382-90. 24. Lavie CJ, McAuley PA, Church TS, et al. Obesity and cardiovascular diseases: implications regarding fitness, fatness, and severity in the obesity paradox. J Am Coll Cardiol 2014:1345-54. 25. McAuley PA, Beavers KM. Contribution of cardiorespiratory fitness to the obesity paradox. Prog Cardiovasc Dis 2014;56:434-40. 26. Myers J, Prakash M, Froelicher V, et al. Exercise capacity and mortality among men referred for exercise testing. N Engl J Med 2002;346:793-801. 27. Farrell SW, Finley CE, Grundy SM. Cardiorespiratory fitness, LDL cholesterol, and CHD mortality in men. Med Sci Sports Exerc 2012;44:2132-7. 28. Blair SN, Kampert JB, Kohl HW, et al. Influences of cardiorespiratory fitness and other precursors on cardiovascular disease and all-cause mortality in men and women. JAMA 1996;276:205-10. 29. Stevens J, Cai J, Evenson KR, et al. Fitness and fatness as predictors of mortality from all causes and from cardiovascular disease in men and women in the Lipid Research Clinics Study. Am J Epidemiol 2002;156:832-41. 30. Farrell SW, Kampert JB, Kohl HW, et al. Influences of cardiorespiratory fitness levels and other predictors on cardiovascular disease mortality in men. Med Sci Sports Exerc 1998;30:899-905. 31. Taylor F, Huffman MD, Macedo AF, et al. Statins for the primary prevention of cardiovascular disease. Cochrane Database Syst Rev 2013;1:CD004816. 32. National Cholesterol Education Program. Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation 2002;106:3143-421. 33. Cholesterol Treatment Trialists. The effects of lowering LDL cholesterol with statin therapy in people at low risk of vascular disease: meta-analysis of individual data from 27 randomised trials. Lancet 2012;380:581-90.


34. Mannu GS, Zaman MJS, Gupta A, et al. Evidence of lifestyle modification in the management of hypercholesterolemia. Curr Cardiol Rev 2013;9:2-14. 35. Desai C, Martin S, Blumenthal R. Non-cardiovascular effects associated with statins. BMJ 2014;349:g3743. 36. Baigent C, Keech A, Kearney PM, et al, Cholesterol Treatment Trialists Collaboration. Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet 2005;366:1267-78. 37. Evans MA. Effects of statins on energy and fatigue with exertion: results from a randomized controlled trial. Arch Intern Med 2012:1. 38. Mikus CR, Boyle LJ, Borengasser SJ, et al. Simvastatin impairs exercise training adaptations. J Am Coll Cardiol 2013;62:709-14. 39. Parker BA, Capizzi JA, Grimaldi AS, et al. Effect of statins on skeletal muscle function. Circulation 2013;127:96-103. 40. Lee D, Markwardt S, Goeres L, et al. Statins and physical activity in older men: the osteoporotic fractures in men study. JAMA Intern Med 2014;174(8):1263-70. 41. Kokkinos PF, Faselis C, Myers J, et al. Interactive effects of fitness and statin treatment on mortality risk in veterans with dyslipidaemia: a cohort study. Lancet 2013;381:394-9. 42. Gulati M, Pandey DK, Arnsdorf MF, et al. Exercise capacity and the risk of death in women: the St. James Women Take Heart Project. Circulation 2003;108:1554-9. 43. Parker BA, Kalasky MJ, Proctor DN. Evidence for sex differences in cardiovascular aging and adaptive responses to physical activity. Eur J Appl Physiol 2010:235-46. 44. Sui X, LaMonte MJ, Blair SN. Cardiorespiratory fitness as a predictor of nonfatal cardiovascular events in asymptomatic women and men. Am J Epidemiol 2007;165:1413-23. 45. McAuley P, Myers J, Abella J, et al. Body mass, fitness and survival in veteran patients: another obesity paradox? Am J Med 2007;120: 518-24. 46. Stone NJ, Robinson JG, Lichtenstein AH, et al. 2013 ACC/AHA Guideline on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2014;63:2889-934. 47. The FIT Project Investigators. Rationale and design of the Henry Ford Exercise Testing Project (The FIT Project). Clin Cardiol 2014;37:456-61. 48. Bruce RA, Kusumi F, Hosmer D. Maximal oxygen intake and nomographic assessment of functional aerobic impairment in cardiovascular disease. Am Heart J 1973;85:546-62. 49. Gibbons R, Balady G, Bricker JT, et al. ACC/AHA 2002 guideline update for exercise testing: summary article. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1997 Exercise Testing Guidelines). J Am Coll Cardiol 2002;40:1531-40. 50. Social Security Death Master File. Change in Public Death Master File Records. http://www.ssdmf.com/import-change-dmf.pdf. [Accessed July 23, 2014]. 51. Sacks FM, Moyé LA, Davis BR, et al. Relationship between plasma LDL concentrations during treatment with pravastatin and recurrent coronary events in the Cholesterol and Recurrent Events trial. Circulation 1998;97:1446-52. 52. Church TS, Earnest CP, Skinner JS, et al. Effects of different doses of physical activity on cardiorespiratory fitness among sedentary, overweight or obese postmenopausal women with elevated blood pressure: a randomized controlled trial. JAMA 2007;297:2081-91. 53. Dunn AL, Marcus BH, Kampert JB, et al. Comparison of lifestyle and structured interventions to increase physical activity and cardiorespiratory fitness: a randomized trial. JAMA 1999;281(4):327-34. 54. Garber C, Blissmer B, Deschenes M, et al. American College of Sports Medicine position stand. Quantity and quality of


exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: guidance for prescribing exercise. Med Sci Sport Exerc 2011;43:1334-59. 55. Thompson P, Arena R, Riebe D, et al. ACSM's new preparticipation health screening recommendations from ACSM's guidelines for exercise testing and prescription, ninth edition. Curr Sport Med Rep 2013;12:215-7. 56. Rengo J, Savage P, Toth M, et al. Statin therapy does not attenuate exercise training response in cardiac rehabilitation. J Am Coll Cardiol 2014;63:2050-1.

Hung et al 399

57. Pinkstaff S, Peberdy MA, Kontos MC, et al. Overestimation of aerobic capacity with the Bruce treadmill protocol in patients being assessed for suspected myocardial ischemia. J Cardiopulm Rehabil Prev 2011;31: 254-60. 58. Bouchard C, Daw EW, Rice T, et al. Familial resemblance for VO2max in the sedentary state: the HERITAGE family study. Med Sci Sports Exerc 1998;30:252-8. 59. Bouchard C, Sarzynski MA, Rice TK, et al. Genomic predictors of the maximal O2 uptake response to standardized exercise training programs. J Appl Physiol 2011;110:1160-70.


Hung et al 399.e1

Appendix

Supplementary Table I. Lipid profile and extended medical history Men Total cohort (n = 33,204)

No statin (n = 10,210)

Statin (n = 8220)

No statin (n = 8650)

Statin (n = 6124)

P⁎

217 ± 47 49 ± 15 139 ± 41 32 ± 16 145 (IQR 101-209)

227 ± 41 44 ± 12 147 ± 36 34 ± 16 155 (IQR 108-230)

193 ± 48 43 ± 12 126 ± 43 33 ± 16 149 (IQR 105-217)

235 ± 39 56 ± 16 147 ± 36 30 ± 15 131 (IQR 92-189)

209 ± 47 54 ± 15 134 ± 45 31 ± 15 141 (IQR 101-199)

b.001 b.001 b.001 b.001 b.001

Lipid profile Total cholesterol (mg/dL) (n = 31,077) HDL cholesterol (mg/dL) (n = 29,860) LDL cholesterol (mg/dL) (n = 30,520) VLDL cholesterol (mg/dL) (n = 29,429) Triglycerides (mg/dL) (n = 30,634) †

Women

Lipid profile for the study population. Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; VLDL, very-low-density lipoprotein; IQR, interquartile range. ⁎ P value between statin-gender groups. † Triglyceride data are provided as median and interquartile range.

Supplementary Table II. Baseline characteristics by MET categories

Demographic data Age (y) Female Race: White Black Other Physiological measurements Resting heart rate (beat/min) Resting systolic blood pressure (mm Hg) Resting diastolic blood pressure (mm Hg) Medical history Obesity (BMI ≥30 kg/m 2) Current smoker Hx hypertension Hypertension medications ACE inhibitor Angiotensin receptor blocker Calcium-channel blocker Hx diabetes mellitus β-Blocker Hx MI Hx CAD Hx PCI Hx CABG Family Hx of CAD Hx atrial fibrillation Hx heart failure Nonstatin lipid-lowering drug COPD medications Aspirin Indication for stress test Chest pain Shortness of breath Established CAD Preoperation

Total cohort (n = 33,204)

b6 METs

6-10 METs

10-12 METs

N12 METs

P⁎

57.0 ± 12 44%

65.4 ± 11 61%

60.3 ± 11 56%

54.2 ± 10 41%

48.6 ± 9 15%

b.001 b.001 b.001

69% 25% 6%

60% 36% 4%

65% 29% 5%

71% 21% 8%

77% 14% 8%

72 ± 13 132 ± 19 81 ± 10

75 ± 14 139 ± 21 81 ± 11

74 ± 13 135 ± 19 81 ± 10

72 ± 12 130 ± 18 81 ± 10

69 ± 11 127 ± 16 81 ± 10

b.001 b.001 .31

24% 43% 73% 55% 24% 4% 16% 24% 27% 14% 18% 7% 5% 52% 3% 2% 7% 9% 27%

24% 40% 90% 77% 35% 5% 25% 37% 38% 28% 35% 11% 12% 43% 6% 6% 7% 12% 34%

31% 44% 81% 66% 29% 5% 19% 30% 32% 15% 20% 8% 6% 51% 3% 1% 8% 10% 30%

25% 44% 68% 49% 20% 4% 13% 19% 24% 10% 13% 6% 3% 55% 2% 0% 8% 8% 24%

11% 41% 53% 30% 12% 2% 7% 10% 15% 6% 9% 4% 2% 57% 1% 0% 6% 6% 19%

b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001

44% 10% 8% 2%

34% 12% 19% 5%

44% 9% 9% 2%

49% 8% 5% 1%

46% 10% 3% 1%

b.001 b.001 b.001 b.001

Baseline characteristics and stress test results for the study population. Abbreviation: Hx, history. ⁎ P value between MET categories.

399.e2 Hung et al


Supplementary Figure 1

Kaplan-Meier survival function for all-cause mortality. Kaplan-Meier survival function for all-cause mortality by MET categories in the total cohort and within each statin-gender group.


Hung et al 399.e3


Kaplan-Meier survival function for myocardial infarction. Kaplan-Meier survival function for myocardial infarction by MET categories in the total cohort and within each statin-gender group.

399.e4 Hung et al



Kaplan-Meier survival function for subsequent revascularization. Kaplan-Meier survival function for subsequent revascularization by MET categories in the total cohort and within each statin-gender group.


Hung et al 399.e5

Supplementary Table III. Effect of cardiorespiratory fitness on outcomes All-cause mortality

b6 METs

6-10 METs

10-12 METs

N12 METs

Total cohort Men not taking statins Men taking statins Women not taking statins Women taking statins

1.00 1.00 1.00 1.00 1.00

(reference) (reference) (reference) (reference) (reference)

0.58 0.61 0.59 0.54 0.56

(0.54-0.62) (0.53-0.69) (0.52-0.67) (0.47-0.62) (0.47-0.67)

0.37 0.37 0.40 0.31 0.39

(0.34-0.40) (0.32-0.43) (0.34-0.47) (0.25-0.38) (0.30-0.52)

0.25 0.26 0.23 0.22 0.14

(0.21-0.29) (0.21-0.32) (0.17-0.30) (0.13-0.39) (0.03-0.57)

Future MI Total cohort Men not taking statins Men taking statins Women not taking statins Women taking statins

1.00 1.00 1.00 1.00 1.00


0.75 0.82 0.84 0.63 0.66

(0.67-0.84) (0.66-1.01) (0.68-1.03) (0.49-0.81) (0.52-0.86)

0.57 0.75 0.57 0.39 0.52

(0.50-0.66) (0.59-0.96) (0.45-0.72) (0.27-0.57) (0.35-0.77)

0.36 0.43 0.33 0.28 0.58

(0.29-0.45) (0.31-0.60) (0.23-0.48) (0.12-0.67) (0.21-1.63)

Subsequent revascularizations Total cohort Men not taking statins Men taking statins Women not taking statins Women taking statins

1.00 1.00 1.00 1.00 1.00


0.81 1.02 0.85 0.70 0.69

(0.73-0.89) (0.84-1.23) (0.72-1.01) (0.55-0.89) (0.55-0.87)

0.56 0.65 0.60 0.41 0.48

(0.49-0.62) (0.53-0.81) (0.49-0.73) (0.29-0.58) (0.34-0.67)

0.34 0.41 0.34 0.13 0.38

(0.29-0.41) (0.31-0.54) (0.26-0.44) (0.05-0.37) (0.15-0.95)

Adjusted HRs by MET categories in the total cohort and each statin-gender group. 95% CIs are shown in parentheses. Hazard ratios were adjusted for age; gender; race; resting heart rate and systolic and diastolic blood pressures; history of hypertension, dyslipidemia, obesity, smoking, diabetes mellitus, atrial fibrillation, heart failure, CAD, MI, PCI, or CABG; family history of CAD; medications used to treat COPD, hypertension, or hyperlipidemia; aspirin; clopidogrel; β-blockers; potential interaction between obesity and statin therapy; and indication for stress testing.

Supplementary Table IV. Effect of cardiorespiratory fitness on outcomes: BMI sensitivity analysis (n = 10,237) All-cause mortality

b6 METs

6-10 METs

10-12 METs

N12 METs

Men not taking statins Men taking statins Women not taking statins Women taking statins

1.00 1.00 1.00 1.00

(reference) (reference) (reference) (reference)

0.58 (0.47-0.72) 0.64 (0.52-0.78) 0.63 (0.50-0.80) 0.57 (0.44-0.75)

0.39 (0.30-0.51) 0.42 (0.32-0.54) 0.36 (0.25-0.53) 0.40 (0.25-0.63)

0.30 (0.20-0.43) 0.28 (0.17-0.45) 0.45 (0.20-0.99) 0.17 (0.02-1.23)

Future MI Men not taking statins Men taking statins Women not taking statins Women taking statins

1.00 1.00 1.00 1.00


0.71 (0.52-0.97) 0.81 (0.61-1.08) 0.72 (0.51-1.02) 0.63 (0.44-0.91)

0.82 (0.59-1.16) 0.61 (0.43-0.85) 0.35 (0.20-0.63) 0.54 (0.30-0.97)

0.51 (0.32-0.82) 0.38 (0.22-0.65) 0.46 (0.16-1.36) 0.65 (0.15-2.83)

Subsequent revascularizations Men not taking statins Men taking statins Women not taking statins Women taking statins

1.00 1.00 1.00 1.00


1.02 (0.77-1.36) 0.85 (0.66-1.11) 0.65 (0.46-0.91) 0.70 (0.51-0.97)

0.79 (0.58-1.08) 0.80 (0.60-1.06) 0.44 (0.27-0.71) 0.42 (0.25-0.71)

0.46 (0.31-0.69) 0.54 (0.36-0.81) 0.27 (0.08-0.92) 0.50 (0.15-1.69)

Adjusted HRs by MET categories in each statin-gender group in the subset of patients with known BMI. 95% CIs are shown in parentheses. Hazard ratios were adjusted for age; gender; race; resting heart rate and systolic and diastolic blood pressures; history of hypertension, dyslipidemia, obesity, smoking, diabetes mellitus, atrial fibrillation, heart failure, CAD, MI, PCI, or CABG; family history of CAD; medications used to treat COPD, hypertension, or hyperlipidemia; aspirin; clopidogrel; β-blockers; potential interaction between obesity and statin therapy; and indication for stress testing.


399.e6 Hung et al

Supplementary Table V. Interactive effect of cardiorespiratory fitness and statin therapy on outcomes: BMI sensitivity analysis (n = 10,237) Men Mortality

Statin therapy

Women No statin therapy

Statin therapy

No statin therapy


1.00 0.52 0.30 0.16

(ref) (0.43-0.64) (0.23-0.38) (0.10-0.25)

1.10 0.56 0.29 0.17

(0.88-1.36) (0.45-0.70) (0.22-0.38) (0.12-0.24)

1.00 0.51 0.31 0.09

(ref) (0.39-0.66) (0.20-0.47) (0.01-0.65)

1.14 0.59 0.24 0.21

(0.91-1.43) (0.45-0.77) (0.16-0.34) (0.10-0.46)

MI b6 METs 6-10 METs 10-12 METs N12 METs

1.00 0.81 0.56 0.36

(ref) (0.61-1.07) (0.41-0.77) (0.22-0.60)

0.89 0.56 0.51 0.26

(0.65-1.21) (0.41-0.77) (0.36-0.71) (0.17-0.41)

1.00 0.67 0.54 0.55

(ref) (0.47-0.94) (0.31-0.93) (0.13-2.26)

1.07 0.64 0.27 0.28

(0.77-1.47) (0.44-0.94) (0.16-0.47) (0.10-0.81)

Subsequent revascularization b6 METs 6-10 METs 10-12 METs N12 METs

1.00 0.90 0.85 0.56

(ref) (0.70-1.17) (0.65-1.11) (0.39-0.82)

0.92 0.92 0.63 0.35

(0.68-1.26) (0.69-1.21) (0.47-0.85) (0.24-0.50)

1.00 0.76 0.46 0.49

(ref) (0.56-1.03) (0.29-0.75) (0.15-1.57)

0.98 0.60 0.34 0.18

(0.71-1.34) (0.43-0.86) (0.21-0.54) (0.06-0.59)

Interactive effect of cardiorespiratory fitness and statin therapy on risk for mortality, MI, and subsequent revascularization in the subset of men and women with known BMI. Patients on statin therapy who achieved b6 METs serve as the referent group. 95% CIs are shown in parentheses. Hazard ratios were adjusted for the propensity to be on statin therapy; race; resting heart rate; resting systolic and diastolic blood pressures; history of obesity, smoking, MI, heart failure, or atrial fibrillation; family history of CAD; use of other lipid-lowering agents, calcium-channel blockers, medications used to treat COPD; potential interaction between obesity and statin therapy; and indication for stress testing.