Statins for primary prevention in physically active individuals: Do the risks outweigh the benefits?

G Model JSAMS-1452; No. of Pages 6

ARTICLE IN PRESS Journal of Science and Medicine in Sport xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Journal of Science and Medicine in Sport journal homepage: www.elsevier.com/locate/jsams

Original research

Statins for primary prevention in physically active individuals: Do the risks outweigh the benefits?夽 Ishak A. Mansi a,b,c,∗ , Jenny L. English d , Michael J. Morris e , Song Zhang c , Eric M. Mortensen a,b,c , Ethan A. Halm b,c a

Department of Medicine, VA North Texas Health Care System, United States Departments of Internal Medicine, University of Texas Southwestern Medical Center, United States c Departments of Clinical Sciences, University of Texas Southwestern Medical Center, United States d Patient Administration Systems & Biostatistics Activity (PASBA), MEDCOM Head Quarter, JBSA Fort Sam, United States e Pulmonary/Critical Care Service, Department of Medicine, San Antonio Military Medical Center, JBSA Fort Sam, United States b

a r t i c l e

i n f o

Article history: Received 21 September 2016 Received in revised form 16 November 2016 Accepted 1 December 2016 Available online xxx Keywords: Physical activity Statin Primary prevention Diabetes Diabetic complications

a b s t r a c t Objectives: There are little data on the potential benefits and adverse events of statins among physically fit individuals. Our objective was to examine the associations of statin use with beneficial cardiovascular outcomes and adverse events in active duty military (a surrogate for high level of physical fitness). Design: This is a retrospective propensity score-matched cohort study of healthy active duty military (fiscal years [FY] 2002–2011). Methods: Statin-users received statins during FY 2005 as their only prescription medication. FY 2002–2004 was used to describe baseline characteristics; and FY 2006–2011were used to capture outcomes. Study outcomes included major acute cardiovascular events (MACE), diabetes mellitus and its complications, kidney diseases, musculoskeletal diseases, obesity, and malignancy. Results: We propensity score matched 837 statin-users to 2488 nonusers. During follow-up, 1.6% statinusers and 1.5% nonusers were diagnosed with MACE (odds ratio [OR] 1.05, 95% confidence interval [CI] 0.55–1.98), 12.5% of statin-users and 5.8% of nonusers were diagnosed with diabetes (OR 2.34, 95% CI 1.79–3.04), and 1.7% statin- users and 0.7% nonusers were diagnosed with diabetes with complication (OR 2.47, 95% CI 1.21–5.04). There were no differences in rates of other adverse events. Conclusions: Among healthy physically active individuals, statin use was associated with doubled the odds of diabetes and diabetic complications without countervailing cardiovascular benefits. Published by Elsevier Ltd on behalf of Sports Medicine Australia.

1. Introduction Physical activity and cardiorespiratory fitness (CRF) have strong inverse relationships to cardiovascular (CV) morbidity and mortality.1 Regardless of the anatomical extent of coronary artery disease, patients with good functional capacity (>10 Metabolic Equivalents [METS]) had lower CV mortality.2 However, physical activity is not included in CV risk calculators that are utilized in estimating the necessity of statin therapy for primary prevention of CV disease.3 Statins effectively lower CV morbidity and mortality but are not without adverse events; specifically, statins are associated with increased risk of diabetes,4 diabetic complications,5 obesity,6–8 and musculoskeletal diseases,9,10 which may inversely affect physical activity. Hence, the overall impact of prescribing statin therapy for

夽 This study was not posted or presented previously. ∗ Corresponding author. E-mail address: [email protected] (I.A. Mansi).

primary prevention for individuals who regularly perform intense physical activity can be different from those who live a sedentary lifestyle.1 There is no statin primary prevention clinical trial that examined the overall effects of statins in physically active individuals who would be expected to benefit less from use of statins for primary prevention. In a previous study including enrollees of Tricare (The US Military Healthcare System), which included active duty military, Veterans, and their families, our group has reported that short-term statin therapy was not associated with reduction in cardiovascular morbidity but was associated with increased risk of adverse events.11 Scarce data exist that exclusively examined the effects of statins in healthy physically active individuals. The objective of this study is to examine the associations of statin use with short- and long-term beneficial CV outcomes and adverse effects in a cohort of active duty military. We considered active duty status as a strong proxy for physical activity with good CRF because of the mandatory biannual military physical fitness test. The military physical fitness

http://dx.doi.org/10.1016/j.jsams.2016.12.075 1440-2440/Published by Elsevier Ltd on behalf of Sports Medicine Australia.

Please cite this article in press as: Mansi IA, et al. Statins for primary prevention in physically active individuals: Do the risks outweigh the benefits? J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2016.12.075

G Model JSAMS-1452; No. of Pages 6

ARTICLE IN PRESS I.A. Mansi et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx

2

Table 1 Short-term outcomes of statin-users in comparison to similar nonusers during FY2006 (1st year of follow-up period).

Major acute cardiovascular events Acute myocardial infarction Cardiac arrest and ventricular fibrillation Acute cerebrovascular disease Peripheral and visceral atherosclerosis Diabetes mellitus Diabetes mellitus with complications Acute and unspecified renal failure Chronic kidney disease Osteoarthritis and arthropathies Dislocation/strain/sprain Malignancy

Statin-users N (%) 837

Nonusers N (%) 2488

OR (95% CI)

p-value

2 (0.2) 1 (0.1) 1 (0.1) 0 (0.0) 0 (0.0) 8 (1.0) 0 (0.0) 1 (0.1) 0 (0.0) 65 (7.8) 38 (4.5) 6 (0.7)

4 (0.2) 1 (0.0) 0 (0.0) 3 (0.1) 1 (0.0) 5 (0.2) 2 (0.1) 0 (0.0) 3 (0.1) 166 (6.7) 122 (4.9) 10 (0.4)

1.49 (0.27–8.14) n/a n/a n/a n/a 4.79 (1.56–14.69) n/a n/a n/a 1.18 (0.87–1.59) 0.92 (0.64–1.34) 1.79 (0.65–4.94)

0.65 – – – – 0.006 – – – 0.28 0.67 0.26

n/a = due to the small number of events in treatment arms calculation of OR was not meaningful.

test includes measured push-ups, sit-ups, and a timed run; hence, it is a good marker of CRF.12 2. Methods The Institutional Review Boards at Brooke Army Medical Center and the VA North Texas Health System approved this study. We extracted national TRICARE data from the Military Health System (MHS) Data Repository (MDR), as detailed in a previous publication.11 The MDR includes patient demographic information, outpatient and inpatient medical encounters within MHS and outside of MHS, laboratory data performed within MHS, and the Pharmacy Data Transaction Service (PDTS). PDTS tracks medication utilization regardless of dispensing pharmacy location or affiliation. According to our agreement with TRICARE, all data were de-identified, to include rounding dates of medical encounters to the nearest quarter of the year. The study period comprised the fiscal years (FY) 2002–FY2011 (10/1/2001–9/30/2011). The period from FY 2002 to 2004 was used to describe baseline characteristics (Baseline period); FY 2005 was used to identify patients as statin-users or nonusers (exposure period); and the period from FY 2006 to 2011 was used to capture outcomes (follow-up period). We included all active duty military at age 35 years or older, who had at least one medical encounter at baseline and were still enrolled in the system at FY 2011. To form a healthy cohort, we: (1) included patients who had a medical encounter in FY 2005 but were not prescribed any medication during the exposure period (FY 2005) except for potentially a statin; (2) excluded patients with preexisting CV diseases at baseline according to the Agency for Health Research and Quality Clinical Classifications Software (AHRQ-CCS): coronary artery diseases, cerebrovascular diseases, and peripheral vascular diseases. Statin-users initiated statins during FY 2005 and continued to use statins for a cumulative duration ≥60 days during FY 2005, but not thereafter. We restricted statin use to FY 2005 only to minimize confounding by indication (as detailed later). Statin-users were considered new users if they did not receive statins in FY 2002–2004. We required that statin-users did not receive other prescription medications in FY 2005 to ensure that no other drugs may have contributed to the outcomes nor caused drug interactions as shown in previous studies.13 Statin nonusers never received a statin throughout the study. We limited nonusers to those who had medical encounters in the first 6 months of FY 2005 because the number of these patients was very large. An outcome event was defined as the occurrence of International Classification of Diseases, 9th Revision, Clinical Modification [ICD-9-CM] codes during the follow-up period. The outcomes were a priori diagnosis groups as defined by AHRQ-CCS categories14 : (1) major acute CV events (MACE), which comprised acute myocardial

infarction, cardiac arrest and ventricular fibrillation, acute cerebrovascular disease, and peripheral and visceral atherosclerosis; (2) diabetes mellitus; (3) diabetes mellitus with complication; (4) acute and unspecified renal failure; (5) chronic kidney disease; (6) osteoarthritis and arthropathies; (7) dislocation, strain, sprain; and (8) malignancy (Appendix A).15 Patients’ comorbidities were identified using ICD-9-CM codes and their Charlson Comorbidity Index was calculated using Deyo’s method.15 Baseline characteristics of groups were compared using ␹2 for categorical variables and Student’s t-test for continuous variables. We created a propensity score to match statin-users and nonusers at a ratio of 1:3 using 50 variables comprising: demographics, personal history, comorbidities, occurrence of outcomes of interest at baseline, undergoing procedures,14 and healthcare utilization (Supplementary Table S1 in the online version at DOI: 10.1016/j.jsams.2016.12.075). We used logistic regression to create the propensity score and perform nearest number matching with a caliper of 0.01 as previously described.11 We captured outcomes at three different intervals: (1) short-term outcomes: FY 2006 only (Table 1); (2) intermediateterm outcomes: FY 2006–2009 (Supplementary Table S2 in the online version at DOI: 10.1016/j.jsams.2016.12.075); and long-term outcomes: FY 2006–2011 (Table 2). To increase specificity of chronic diseases diagnoses, we required that they had been diagnosed in ≥2 separate encounters as previously published11 ; however, we accepted 1 encounter in defining acute diseases such as cardiac arrest or in conditions known to have low sensitivity in identification using ICD-9-CM codes such as obesity (Appendix B). To minimize confounding by indication in determining MACE, we adopted several measures. First, we included healthy population who received statins as the only prescription medication. Patients with CV disease are likely to receive anti-platelet agents, beta-blockers, and other medications. Second, we included statin-users who used statins in FY 2005 only since studies have shown that patients who adhered to statins for shorter periods only were less likely to have diseases that compel long-term statin prescription.16 Third, we did not include the exposure period (FY2005) in our outcomes; excluding a time period equivalent to the expected period for a drug to attain efficacy can mitigate the effects of confounding by indication.17 Fourth, we counted the incidence of undergoing invasive cardiac procedures during baseline to serve as a measure for our success in minimizing confounding by indication. If confounding by indication is present, statin-users were expected to have more invasive procedures. For primary analysis, we used conditional logistic regression to calculate odds ratios (OR) and 95% confidence intervals (95% CI) in the propensity score- matched cohort in each of short-term, intermediate-term, and long-term follow-up. For Secondary analyses, we performed several analyses adjusting for variables such as vital signs and laboratory values measured

Please cite this article in press as: Mansi IA, et al. Statins for primary prevention in physically active individuals: Do the risks outweigh the benefits? J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2016.12.075

G Model

ARTICLE IN PRESS

JSAMS-1452; No. of Pages 6

I.A. Mansi et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx

3

Table 2 Long-term outcomes of statin-users in comparison to similar nonusers during follow-up FY 2006-2011. Statin-users N (%) N = 837

Nonusers N (%) N = 2488

Primary analysis OR (95% CI)

p-value

Secondary analysis Adjusted OR (95% CI)

p-value

13 (1.6)

37 (1.5)

1.05 (0.55–1.98)

0.89

1.71 (0.80–3.64)a 1.23 (0.49–3.06)b 1.16 (0.46–2.95)c

0.17 0.66 0.75

6 (0.7) 1 (0.1) 4 (0.5) 2 (0.2)

8 (0.3) 5 (0.2) 13 (0.5) 12 (0.5)

2.24 (0.77–6.47) 0.59 (0.07–5.09) 0.91 (0.30–2.81) 0.49 (0.11–2.21)

0.14 0.64 0.88 0.36

Diabetes mellitus

105 (12.5)

144 (5.8)

2.34 (1.79–3.04)

<0.001

Diabetes mellitus with complications

14 (1.7)

17 (0.7)

2.47 (1.21–5.04)

0.01

2.86 (2.05–3.99)a 2.86 (2.06–3.98)d 2.40 (0.88–6.55)a 2.59 (0.98–6.83)d

<0.001 <0.001 0.09 0.05

Acute and unspecified renal failure Chronic kidney disease Osteoarthritis and arthropathies Dislocation/strain/sprain Malignancy

2 (0.2) 5 (0.6) 363 (43.4) 247 (29.5) 30 (3.6)

19 (0.8) 17 (0.7) 1054 (42.4) 746 (30.0) 85 (3.4)

0.31 (0.07–1.34) 0.87 (0.32–2.38) 1.04 (0.89–1.22) 0.99 (0.82–1.16) 1.05 (0.69–1.61)

0.12 0.79 0.61 0.80 0.82

0.79 (0.12–5.17)c 0.98 (0.81–1.18)e 0.85 (0.69–1.04)e 1.21 (0.72–2.04)e

0.80 0.80 0.11 0.46

Major acute cardiovascular events

Acute myocardial infarction Cardiac arrest and ventricular fibrillation Acute cerebrovascular disease Peripheral and visceral atherosclerosis

a

Adjusted for: Systolic blood pressure, diastolic blood pressure, body mass index at FY 2011, and propensity score; results were available for 2102 patients only. Adjusted for: Systolic blood pressure, diastolic blood pressure, body mass index, HDL- Cholesterol, LDL-Cholesterol; results were available for 1103 patients only. c Adjusted for: Presence of diabetes mellitus during follow-up period, systolic blood pressure, diastolic blood pressure, body mass index, HDL- Cholesterol, LDL-Cholesterol, and propensity score; results were available for 1103 patients only. d Adjusted for: body mass index at FY 2011, and propensity score; results were available for 2102 patients only. e Adjusted for: Presence of diabetes mellitus during follow-up period, systolic blood pressure, diastolic blood pressure, body mass index, and propensity score; results were available for 2102 patients only. b

at end of the follow-up period (FY 2011) to account for any confounders that might have been differentially introduced during the follow-up period.18 For example, clinicians may have prescribed statins because they noted in those patients (who would be statinusers) a tendency toward sedentary life style or unhealthy eating habits; therefore, we adjusted for body mass index, blood pressure, and other parameters at FY 2011 (Table 2). We also performed Cox-proportional hazard regression analysis and determined hazard ratios (HR) of each outcome among statin-users in comparison to nonusers (Table 3). In each analysis, the outcome was the time interval from the beginning of the follow-up period until the first occurrence of the outcome of interest. Statistical significance was defined as two-tailed p-values <0.05. Statistical analyses were performed using SPSS version 21 (IBM, Armonk, NY). 3. Results Cohort assembly is depicted in Fig. S1 of Supplementary material in the online version at DOI: 10.1016/j.jsams.2016.12.075. Out of 877 statin-users and 86,097 nonusers who met study criteria, we successfully matched 837 statin-users and 2488 nonusers (ratio of 1:3) with no statistically significant differences in all baseline characteristics (Supplementary Table S1 in the online version at DOI: 10.1016/j.jsams.2016.12.075). The cohort was healthy at baseline with a mean age of 45 year-old and a mean Charlson comorbidity score of <0.1; 97% were men. Only one patient underwent cardiac catheterization and none had a revascularization procedure. Patients were followed-up for 19,443 person-years. Statin-users used statins for a mean period of 234 days (standard deviation [SD] ± 97 days) and a median of 270 days; 75% of the prescriptions were for simvastatin, 19% for atorvastatin, 4% for pravastatin and lovastatin, and 2% for rosuvastatin. At some point during the study period, 19% used low-intensity statins, 82% used moderate- intensity statins, and 8% used high-intensity statins. Statin intensity was defined according to the 2013 American College of Cardiology/American Heart Association (ACC/AHA) cholesterol guidelines with modification to include simvastatin 80 mg in the high-intensity group as we described in previous publication.11 Tables 1 and 2 depict short-term (FY 2006), and long-term outcomes (FY 2006–2011). Table S2 of Supplementary material in the online version at DOI: 10.1016/j.jsams.2016.12.075 depicts

Table 3 Long-term hazard ratios of outcomes in statin-users in comparison to similar nonusers.

Major acute cardiovascular events Diabetes mellitus Diabetes mellitus with complication Acute kidney disease Chronic kidney disease Osteoarthritis and arthropathies Dislocation/strain/sprain

Hazard ratio (95% CI)

p-value

1.04 (0.56, 1.97) 2.26 (1.77, 2.90) 2.46 (1.21, 4.99) 0.31 (0.07, 1.34) 0.87 (0.32, 2.37) 1.06 (0.94, 1.19) 0.99 (0.86, 1.14)

0.89 <0.001 0.01 0.12 0.79 0.38 0.89

intermediate-term outcomes (FY 2006–FY 2009). Statin-users in comparison to nonusers had increased risk of diabetes that consistently continued throughout follow-up. At FY 2011, 5.8% of nonusers and 12.5% of statin-users were diagnosed with diabetes (OR 2.34, 95% CI 1.79–3.04). The risk for diabetes with complications was also higher among statin-users. In long-term follow-up, 0.7% of nonusers, 1.7% of statin-users were diagnosed with diabetes with complication (OR 2.47, 95% CI 1.21–5.04). There was no difference in the risk of musculoskeletal diseases, renal diseases, or malignancy between statin-users and nonusers. At end of long-term follow-up, only 1.5% of nonusers and 1.6% of statin-users were diagnosed with MACE (OR 1.05, 95% CI 0.55–1.98). Overall, incidence of MACE was low (2.63 events per 1000 person-years in nonusers and 2.58 in statin-users, p = 0.9). Secondary analyses showed that the risk of diabetes persisted despite various adjustments (Table 2). Similarly, adjusting for these variables and lipid values during the follow-up did not change our findings regarding MACE. Of note, one patient of the nonusers died during FY2011, but none of the statin-users. Time-to-event analyses showed a similar pattern of increased hazard of diabetes and diabetes complications in statin-users with no difference in MACE (Table 3 and Supplementary Fig. S2 in the online version at DOI: 10.1016/j.jsams.2016.12.075). 4. Discussion This study of healthy active duty military, a strong surrogate for physically active population with good CRF, demonstrated that statin use for primary prevention was associated with approximately doubled odds of developing diabetes mellitus and diabetic complications. Both statin-users and nonusers in this physically active healthy population had very low rates of MACE (2.63 events

Please cite this article in press as: Mansi IA, et al. Statins for primary prevention in physically active individuals: Do the risks outweigh the benefits? J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2016.12.075

G Model JSAMS-1452; No. of Pages 6 4

ARTICLE IN PRESS I.A. Mansi et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx

per 1000 person-years in nonusers and 2.58 in statin- users), in contrast to a rate of 3.99 events per 1000 person-years in the general population19 ; hence, statin beneficial cardiovascular effects could not be demonstrated. Active duty military status is a strong proxy for physical activity with good CRF because of the military requirements for passing a biannual basic military physical fitness test. The basic military physical fitness test provides a measure of upper and lower body muscular endurance and overall physical fitness.20 Whereas the details of the test may vary with the branch of the service (Army, Air Force, or Navy), it generally consists of timed push-ups and situps and a timed run that must be completed within a prespecified time interval. For example, the Army Physical Fitness Test must be completed within 2 h, with a maximum of 20 min of rest between its three events of push-ups, sit-ups, and a 2-mile run, which have to be performed in this specific sequence. It has a maximum score of 300 and minimum passing score of 180.21 In order for a 45-yearold male soldier (mean age of subjects included in this study) to pass the test, he should be able to perform at least 30 push-ups in two minutes and 32 sit- ups in two minutes, and complete a 2 mile-run in <18.43 min.21 In a previous study from the same Tricare population using similar inclusion and exclusion criteria to this study but included Veterans and family members not only active duty military, we reported that the overall rate of MACE was higher than the rate reported in this study (approximately 2.1% in our previous study vs. 1.5% in this study).11 Of note, although the overall rate of diabetes was much higher in the previous study (12.5%) in comparison to the current study that included active duty military only (7.5%), the odds ratio of diabetes in statin-users in comparison to nonusers was higher in the current study (OR 2.34) than in the previous study that included general population (OR 1.93). Similarly, although the overall rate of diabetes with complication was higher in the previous study (1.7%) in comparison to the current study (0.9%), the odds ratio of diabetes with complication was higher in active duty military (OR 2.47) than in the previous study (OR 2.15).11 This observational study helps fill a gap in the literature because there are no published randomized trials that examined the longterm effects of statin use in physically active individuals. Our findings are clinically significant because none of the CV risk score calculators consider CRF in its calculations, therefore, some of those healthy and active individuals will unnecessary receive statins.3 Additionally, our study also found that the risks of adverse events persist long after statins were discontinued, which have been also reported in some studies noting that both statins CV benefits and adverse events persisted for 3–5 years after the treatment had been stopped.9,11,22 The increased risk of diabetes among statin-users is well described in clinical trials,23 meta- analysis,4 and observational studies.5,24 In the JUPITER trial (the Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin), higher incidence of diabetes was noted among statin-users despite its short median follow-up of 1.9 years.23 We have previously reported the association of statin use with increased risk of diabetic complications.5,11 Additionally, several studies including a prospective observational study,10 a cross-sectional study,6 and a clinical trial,7 indicate that statin use is associated with higher risk of obesity. On the other hand, several studies have shown that structured aerobic exercise have favorable CV benefits similar to medications.1,25,26 A recent study noted that individuals with high physical fitness (>10 METS) in comparison to those with lowest fitness had lowest risk of mortality regardless of whether statins were taken or not indicating that statins offered little or no benefit in those at high CRF.27 In a meta- analysis (33 studies including 102,980 participants), each MET unit of maximal aerobic capacity

was associated with 13% reduction in total mortality1 ; such a reduction compare favorably to statins.25 In another meta-analysis, there were no detectable differences between exercise and drug interventions in the secondary prevention of coronary heart disease.28 Whereas from the perspective of their beneficial effects, exercise and statins may be comparable, statins increase the risk of diabetes and insulin resistance,29 but exercise decreases the risk of diabetes and insulin resistance.25 This study suggests that we may need to adjust our approach and priorities to primary prevention for various populations. Recently, the incidence of acute coronary events has been declining in developed countries,30 whereas the prevalence of diabetes and obesity had soared.31 For example, the estimated prevalence of diabetes mellitus in the US in 2011–2012 was 12–14%31 and the estimated relative risk increase of incident diabetes with statin use is 10–12%7 ; this would translate to an absolute number of additional 1.2–1.4 incident diabetes for every 100 statin-users. Therefore, the benefit-risk ratio of statin therapy for healthier contemporary populations such as those included in this study may be very different from those demonstrated in older clinical trials employing sicker populations. Indeed, in this study, the absolute number of MACE events was too small to demonstrate CV benefits. Even considering that statins would result in a relative risk reduction of CV diseases of ≥25% according to clinical trials,32 the absolute number of MACE events would be too small to demonstrate a statistically significant benefit. Our study has several limitations including its retrospective observational design that may suffer from unrecognized confounders. It may be tempting to surmise that our findings are due to unidentified confounders, however, one would expect to see an increase in risk of other diseases such as kidney diseases, malignancy, and osteoarthritis, not only those outcomes that have biological plausibility and are reported in recent studies. Although limiting statin-users to those who used statins in FY 2005 helps mitigating confounding by indication, it limits the generalizability of our study to individuals who used statins for shorter periods only. Using ICD-9-CM codes may not reflect disease severity, however, we do not know of any reason for differential bias between statin-users and nonusers. Furthermore, adjusting for actual values of blood pressure, body mass index, blood glucose, hemoglobin A1c, and serum lipids at end of follow-up did not significantly change our results; yet, we only had laboratory and vital signs data at the end of follow-up for approximately one-third of the study patients. Another limitation for the study is that we could not ascertain that all subjects in the study have successfully passed their military physical fitness test. Whereas it is expected that all active duty military pass the test, exemptions due to illness, injury, or other causes may exist; the proportions of such exceptions is not expected to be high, specifically among this study healthy population. 5. Conclusion Short term statin use in a healthy, physically active population was associated with higher risks of diabetes and diabetic complications without any of the hoped for CV benefits. Our findings raise concerns about the risk-benefit ratio of statins among physically active, healthy individuals. Future refinements of CV risk calculators and national guidelines regarding statins for primary prevention should consider factoring in physical activity. Practical implications - This study examined the risks of major acute cardiovascular events, diabetes mellitus and its complications, kidney diseases, musculoskeletal diseases, and malignancy in a cohort of healthy physically active participants of statin-users and nonusers.

Please cite this article in press as: Mansi IA, et al. Statins for primary prevention in physically active individuals: Do the risks outweigh the benefits? J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2016.12.075

G Model

ARTICLE IN PRESS

JSAMS-1452; No. of Pages 6

I.A. Mansi et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx

- Statin-users used statins for primary prevention and statins were their only prescription medication. Nonusers received medical care but did not receive any medications. - We matched statin users and nonusers baseline characteristics on demographics, personal history, comorbidities, occurrence of outcomes of interest at baseline, undergoing procedures, and healthcare utilization. We, thereafter, examined the risks of outcomes during the follow-up period. - Among healthy physically active participants, statin use was associated with doubled the odds of diabetes and diabetic complications without countervailing cardiovascular benefits. - Physically active healthy individuals with good cardiorespiratory fitness may not benefit from statin therapy for primary prevention. Further study is urgently needed to examine the benefit-risk ratio of statin therapy in this category of population. Additionally, specific cardiovascular risk calculators that incorporate intensive physical activity as one of its substrate need to be developed to guide statin therapy for such population. Funding source No funding was provided for the conduct of this study. This material is the result of work supported with resources and the use of facilities at the VA North Texas Health Care System and the University of Texas Southwestern. Drs. Halm’s effort was funded in part by the AHRQ R24 HS022418 and NCATS U54 RFA-TR-12-006, and Dr. Mortensen’s by AHRQ R24 HS022418. Disclaimer The views expressed herein are those of the authors and do not reflect the official policy or position of the Department of the Army, Department of Defense, VA Administration, or the US Government. The authors are employees of the US government. This work was prepared as part of their official duties and, as such, there is no copyright to be transferred. Acknowledgement This material is the result of work supported with resources and the use of facilities at the VA North Texas Health Care System and the University of Texas Southwestern. No funding was provided for the conduct of this study; Drs. Halm’s effort was funded in part by the AHRQ R24 HS022418 and NCATS U54 RFA-TR-12-006, and Dr. Mortensen’s by AHRQ R24 HS022418. Appendix A. Definitions of outcomes. Outcome

Definition

AHRQ-CCS 49 Diabetes mellitus without complication AHRQ-CCS 50 Diabetes mellitus with complication Major acute cardiovascular events (MACE) comprised of: AHRQ-CCS category100 Acute myocardial infarction AHRQ-CCS category 107 Cardiac arrest and ventricular fibrillation Acute cerebrovascular disease AHRQ-CCS category 109 AHRQ-CCS category 114 Peripheral and visceral atherosclerosis AHRQ-CCS category 157 Acute and unspecified renal failure AHRQ-CCS category 158 Chronic kidney disease AHRQ-CCS categories 203 and 204 Osteoarthritis and arthropathies Dislocation/strain/sprain AHRQ-CCS categories 225 and 232 ICD-9-CM codes of any malignancy Malignancy excluding skin cancer other than melanoma as defined by Deyo et al.18

AHRQ-CCS = the Agency of Health Research and Quality Clinical Classification Software.17

5

Appendix B. Number of separate encounters with a diagnosis that are required to define the occurrence of the diagnosis. Disease group (definition)

Number of encounters

Immunization and infectious disease screening (CCS 10) Thyroid diseases (CCS 48) Diabetes mellitus without complication (CCS 49) Diabetes mellitus with complication (CCS 50) Gout (CCS 54) Deficiency anemia (CCS 59) Headache (CCS 84) Cataract (CCS 86) Valvular heart diseases (CCS 96) Pericarditis, myocarditis, endocarditis (CCS 97) Hypertension (CCS 98) Hypertension with complication (CCS 99) Acute myocardial infarction (CCS 100) Non-specific chest pain (CCS 102) Cor pulmonale (CCS 103) Nonspecific heart diseases (CCS 104) Conduction disorders (CCS 105) Cardiac dysrhythmias (CCS 106) Cardiac arrest and ventricular fibrillation (CCS 107) Acute cerebrovascular disease (CCS 109) Peripheral and visceral atherosclerosis (CCS 114) Aortic; peripheral; and visceral artery aneurysms (CCS 115) Chronic obstructive pulmonary disease and bronchiectasis (CCS 127) Asthma (CCS 128) Gastroduodenal ulcer (CCS 139) Gastritis and duodenitis (CCS 140) Gastrointestinal hemorrhage (CCS 153) Nephritis and nephrosis (CCS 156) Acute and unspecified renal failure (CCS 157) Chronic kidney disease (CCS 158) Osteoarthritis and other non-traumatic joint disorders (CCS 203 and 204) Spondylosis, intervertebral disc disorders, other back problems (CCS 205) Osteoporosis (CCS 206) Pathological fracture (CCS 207) Trauma-related joint disorders and dislocations; sprains and strains (CCS 225 and 232) Syncope (CCS 245) Rehabilitation care, fitting of prostheses, and adjustment of devices (CCS 254) Alcohol abuse/dependence (CCS 660) Illicit drugs use (CCS 661) Family history of cardiovascular disease (V171, V1749, V174, V1741, and V173) Obesity (ICD-9-CM codes: 2780, 27800, 27801, 27802, 27803, 2781, 2788, and 7831) Smoking (ICD-9-CM codes: 3051 and V1582)

1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 2 1 1 1 1 1

Musculoskeletal procedures (CCS procedural codes categories) Procedures involving joints, muscles and tendons (149, 151, 155, 156, 159, 160, 162, 163, and 164) Partial or total hip or knee replacement (152, 153, 154) Procedures involving spinal vertebrae (3, 5, and 158) Reduction of fracture and dislocation (ICD- 9-CM codes: 797, 7970, 7971, 7972, 7973, 7974, 7975, 7976, 7977, 7978, 7979, 798, 7980, 7981, 7982, 7983, 7984, 7985, 7986, 7987, 7988, 7989)

1 1 1

Cardiovascular procedures (CCS procedural codes categories) Electrocardiography (202) Echocardiography (193) Stress test (201) Cardiac catheterization (47) Percutaneous coronary intervention (45 and 46) Coronary artery bypass graft surgery (44) Pacemaker/defibrillator implantation (48) Peripheral arterial revascularization procedures (51, 55, and multilevel code 7.18)

1 1 1 1 1 1 1 1

1

CCS = Clinical Classification Software Category of the Agency of Health Research and Quality.17

Please cite this article in press as: Mansi IA, et al. Statins for primary prevention in physically active individuals: Do the risks outweigh the benefits? J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2016.12.075

G Model JSAMS-1452; No. of Pages 6

ARTICLE IN PRESS I.A. Mansi et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx

6

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(19):2024–2035. 2. Mark DB, Hlatky MA, Harrell Jr FE et al. Exercise treadmill score for predicting prognosis in coronary artery disease. Ann Intern Med 1987; 106(6):793–800. 3. Ridker PM, Cook NR. Comparing cardiovascular risk prediction scores. Ann Intern Med 2015; 162(4):313–314. 4. Sattar N, Preiss D, Murray HM et al. Statins and risk of incident diabetes: a collaborative meta-analysis of randomised statin trials. Lancet 2010; 375(9716):735–742. 5. Mansi I, Frei CR, Wang CP et al. Statins and new-onset diabetes mellitus and diabetic complications: a retrospective cohort study of US healthy adults. J Gen Intern Med 2015; 30(11):1599–1610. 6. Sugiyama T, Tsugawa Y, Tseng CH et al. Different time trends of caloric and fat intake between statin users and nonusers among US adults: gluttony in the time of statins? JAMA Intern Med 2014; 174(7):1038–1045. 7. Maki KC, Ridker PM, Brown WV et al. An assessment by the statin diabetes safety task force: 2014 update. J Clin Lipidol 2014; 8(3 Suppl):S17–S29. 8. Swerdlow DI, Preiss D, Kuchenbaecker KB et al. HMG-coenzyme A reductase inhibition, type 2 diabetes, and bodyweight: evidence from genetic analysis and randomised trials. Lancet 2015; 385(9965):351–361. 9. Hippisley-Cox J, Coupland C. Unintended effects of statins in men and women in England and Wales: population based cohort study using the QResearch database. BMJ 2010; 340:c2197. 10. Lee DS, 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):1263–1270. 11. Mansi IA, English J, Zhang S et al. Long-term outcomes of short-term statin use in healthy adults: a retrospective cohort study. Drug Saf 2016; 39(6):543–559. 12. Army Regulation 40-501; Standards of Medical Fitness, Washington, DC, Headquarters, Medical Services, Department of the Army, 2007. 13. Li DQ, Kim R, McArthur E et al. Risk of adverse events among older adults following co-prescription of clarithromycin and statins not metabolized by cytochrome P450 3A4. CMAJ 2015; 187(3):174–180. 14. Elixhauser A, Steiner C, Palmer L. Clinical classifications software (CCS) for ICD9-CM. Databases and related tools from the healthcare cost and utilization project (HCUP), 2012. 15. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992; 45(6):613–619. 16. Mann DM, Woodward M, Muntner P et al. Predictors of nonadherence to statins: a systematic review and meta-analysis. Ann Pharmacother 2010; 44(9):1410–1421.

17. Danaei G, Rodriguez LA, Cantero OF et al. Observational data for comparative effectiveness research: an emulation of randomised trials of statins and primary prevention of coronary heart disease. Stat Methods Med Res 2013; 22(1):70–96. 18. Manson JE, Shufelt CL, Robins JM. The potential for postrandomization confounding in randomized clinical trials. JAMA 2016; 315(21):2273–2274. 19. Yusuf S, Rangarajan S, Teo K et al. Cardiovascular risk and events in 17 low-, middle-, and high-income countries. N Engl J Med 2014; 371(9):818–827. 20. Field manual 7-22: Army physical readiness training, Washington, DC, Department of the Army, 2012. Available at: http://www.globalsecurity.org/military / library/policy/army/fm/7-22/fm7-22.pdf. [Accessed 20 October 2016]. 21. Army physical fitness test. Available at: https://www.marist.edu/studentlife/ rot c/pdfs/apft.pdf. [Accessed 20 October 2016]. 22. Ford I, Murray H, Packard CJ et al. Long-term follow-up of the west of Scotland coronary prevention study. N Engl J Med 2007; 357(15):1477–1486. 23. Ridker PM, Danielson E, Fonseca FA et al. Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein. N Engl J Med 2008; 359(21):2195–2207. 24. Culver AL, Ockene IS, Balasubramanian R et al. Statin use and risk of diabetes mellitus in postmenopausal women in the Women’s Health Initiative. Arch Intern Med 2012; 172(2):144–152. 25. Franklin BA, Lavie CJ. Impact of statins on physical activity and fitness: ally or adversary? Mayo Clin Proc 2016; 90(10):1314–1319. 26. 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 2016; 381(9864):394–399. 27. Myers J, Kokkinos P, de Araújo C. Coronary artery disease prevention: should exercise, statins, or both, be prescribed? Rev DERC 2014; 20(4):102–105. 28. Naci H, Ioannidis JPA. Comparative effectiveness of exercise and drug interventions on mortality outcomes: metaepidemiological study. Br J Sports Med 2015; 49(21):1414–1422. 29. Mitchell P, Marette A. Statin-induced insulin resistance through inflammasome activation: sailing between Scylla and Charybdis. Diabetes 2016; 63(11):3569–3571. 30. Fuster S, Kelly BB, editors. Institute of medicine (US) committee on preventing the global epidemic of cardiovascular disease: meeting the challenges in developing countries, in promoting cardiovascular health in the developing world: a critical challenge to achieve global health. Washington (DC); 2010. Available from: http:// www.ncbi.nlm.nih.gov/books/NBK4 5693/. 31. Menke A, Casagrande S, Geiss L et al. Prevalence of and trends in diabetes among adults in the United States, 1988-2012. JAMA 2015; 314(10):1021–1029. 32. Heart Protection Study Collaborative G, Bulbulia R, Bowman L et al. Effects on 11-year mortality and morbidity of lowering LDL cholesterol with simvastatin for about 5 years in 20,536 high-risk individuals: a randomised controlled trial. Lancet 2011; 378(9808):2013–2020.

Please cite this article in press as: Mansi IA, et al. Statins for primary prevention in physically active individuals: Do the risks outweigh the benefits? J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2016.12.075