low lifetime
Content
Low lifetime recreational activity is a risk factor for peripheral arterial disease
Andrew M. Wilson, MBBS, PhD,a Amir H. Sadrzadeh-Rafie, MD,b Jonathan Myers, PhD,b Themistocles Assimes, MD, PhD,b Kevin T. Nead, BS,b Mamie Higgins, BS,b Andre Gabriel, MD,c Jeffrey Olin, DO,c and John P. Cooke, MD, PhD,b Melbourne, Australia; Stanford, Calif; and New York, NY
Background: The relationship between lifetime physical activity and the risk of developing peripheral arterial disease (PAD) is not known. Methods: We studied 1381 patients referred for elective coronary angiography in a point prevalence analysis. PAD was defined as ankle-brachial index (ABI) <0.9 at the time or a history of revascularization of the lower extremities regardless of ABI measure. We used a validated physical activity questionnaire to retrospectively measure each patient’s lifetime recreational activity (LRA). Multivariate and logistic regression analyses were used to assess the independent association of LRA to ABI and the presence of PAD. Results: PAD was present in 19% (n 258) of all subjects. Subjects reporting no regular LRA had greater diastolic blood pressure and were more likely to be female. They had lower average ABI, and a higher proportion had PAD (25.6%). In a regression model, including traditional risk factors and LRA, multivariate analysis showed that age (P < .001), female gender (P < .001), systolic blood pressure (P .014), fasting glucose (P < .001), serum triglycerides (P .02), and cumulative pack years (P < .001) were independent negative predictors of ABI, and LRA was a positive predictor of ABI (P < .001). History of sedentary lifestyle independently increased the odds ratio for PAD (odds ratio, 0.46; 95% confidence interval, 1.01-2.10) when assessed by logistic regression. Intriguingly, there is a correlation between physical activity and gender, such that women with low LRA are at greatest risk. Conclusion: Recalled LRA is positively correlated to ABI and associated with PAD. Whereas the mechanism for this effect is not clear, LRA may be a useful clinical screening tool for PAD risk, and strategies to increase adult recreational activity may reduce the burden of PAD later in life. ( J Vasc Surg 2011; : .)
Peripheral arterial disease of the lower extremities (PAD) is a common disorder affecting 8 to 12 million individuals in the United States. PAD causes limb pain with exertion, and reduces functional capacity and quality of life1 and is frequently associated with atherosclerotic disease in other vascular beds.2 Patients with PAD are at increased risk from myocardial infarction, cerebrovascular events, aortic aneurysm rupture, and death from cardiovascular causes, as well as ischemic ulceration and amputation.3 In patients with PAD, exercise therapy improves physical functioning.4 Conversely, lack of activity is a risk for disease progression, loss of function,5 and adverse cardiovascular
From the Department of Medicine, University of Melbourne, Melbournea; Stanford University Medical Center, Stanfordb; and Mount Sinai Medical Center, New York.c This study was supported in part by grants from the National Heart, Lung and Blood Institute (RO 1 HL-075774 and K12 HL087746; JPC), and NIH grant M01 RR 00070 (General Clinical Research Center, Stanford University School of Medicine). Competition of interest: none. Additional material for this article may be found online at www.jvascsurg. org. Reprint requests: John P. Cooke, MD, PhD, Professor of Medicine, Stanford University Medical Center, 300 Pasteur Dr, Falk Cardiovascular Research Center, Stanford, CA 94305-5406 (e-mail: john.cooke@ stanford.edu). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest. 0741-5214/$36.00 Copyright © 2011 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2011.02.052
disease outcome.6 It is known that exercise has a favorable effect on endothelial function, in part via beneficial effects on nitric oxide (NO) metabolism.7,8 Lack of leisure time physical activity within the year prior to examination is associated with a lower ABI.9 This may be due to the patients’ inability to participate in moderate and high intensity activities due to claudication. However, it is unclear whether the typically sedentary nature of the PAD patient is merely a manifestation of the disease, or may also have an etiologic role. There are no data available regarding the role of lifetime physical activity in modulating the development of PAD. To this end, we aimed to assess the relative impact of lifetime recreational activity (LRA) on the prevalence of PAD in a group of patients referred for coronary angiography. METHODS Participants. The National Heart, Lung and Blood Institute-funded study, “Genetic Determinants of Peripheral Arterial Disease,” is an observational study to delineate the determinants of atherosclerotic plaque distribution. Subjects entered into the study are extensively characterized to include ethnicity, lifestyle, physical function, medical and family history, traditional risk factors, and proteomic and genomic information. Subjects were recruited from patients undergoing elective coronary arteriograms at Stanford University and Mt. Sinai Medical Centers. A point prevalence analysis was performed for 1381 subjects (mean [ SD] age, 65 11 years) who had been recruited between April 2004 and January 2008. Indications for cath1
2 Wilson et al
JOURNAL OF VASCULAR SURGERY 2011
Table I. Subject demographics and lifetime recreational activity
No LRA (n 417) Age, years Gender, % F Body mass index (kg/m2) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) High-density lipoprotein (mg/dL) Triglycerides (mg/dL) Low-density lipoprotein (mg/dL) Glucose (mg/dL) Pack years smoking Log C-reactive protein (mg/dL) 66 11 51.8 29.5 6.4 149 26 77 12 42 13 117 90 80 31 122 54 17 27 0.3 0.6 (n Q2 241) (n Q3 241) (n Q4 241) (n Q5 241) P value .147 .001 .175 .116 .001 .129 .494 .088 .087 .999 .107
241]).
66 10 39.4 29 5.7 145 22 73 14 40 12 110 62 75 31 124 101 17 23 0.2 0.5
66 11 25.7 28.8 5.9 145 20 74 13 40 11 114 73 75 29 117 43 16 23 0.2 0.6
66 11 27.8 28.6 5.1 149 22 75 14 40 12 106 69 80 34 112 35 16 23 0.2 0.6
64 11 20.3 28.7 6.5 147 23 75 12 42 12 113 85 77 28 114 43 16 24 0.2 0.5
LRA, Lifetime recreational activity; Q, quintile. Variables are compared based on the quintiles of LRA (one group with no LRA [n Analysis of covariance was used to generate the P value with age adjustment.
417] and four quartiles of those who reported any activity [n
eterization included angina, shortness of breath, abnormal stress test, or known coronary artery disease (CAD). Patients admitted for emergent catheterizations, or screening catheterizations prior to organ transplants were excluded from the study. Patients 40 years old and those with language barriers, a history of radiation treatment, and known chronic infectious diseases such as HIV and hepatitis B and C were also excluded. Study assessments. Information about demographic factors, medications, personal and familial history of cardiovascular disease, walking impairment questionnaire, and social history were obtained using structured questionnaires administered by study personnel. Prior to the coronary angiogram, posterior tibial, dorsalis pedis, and brachial artery systolic pressures were measured using a 5 MHz Doppler ultrasound. One arm was used for the brachial pressures due to the presence of an intravenous line in the other arm. Ankle-brachial indexes (ABI) were calculated by dividing the higher ankle pressure of each leg over the brachial pressure. Each patient was then classified as having PAD (ABI 0.9 in either leg) or not having PAD (ABI 0.9 in both legs). Patients with ABI 1.3 or noncompressible arteries were not included in the continuous analysis. Patients with normal ABI ( 0.9) but with a history of PAD-related surgery, angioplasty, or stenting were also classified as having PAD, but the data for ABI were not included in the continuous analysis. Past history of PAD procedure was included in the logistic regression, however. Finally, 30 mL of blood was collected from each patient through a venous or arterial femoral sheath, or from an intravenous line in the arm. The blood specimens were centrifuged at 3000 RPM for 20 minutes at 4 degrees Celsius. Aliquots of EDTA plasma and serum were stored at 75 degrees Celsius. Laboratory measurements were obtained from the Stanford and Mt. Sinai hospital clinical laboratories, and high sensitivity C-reactive protein (hsCRP) was measured for a subset of the cohort (n 580) at the Boston Children’s Hospital (Dr Nader Rifai). Each angiogram was assessed by a board-certified interventional cardiologist blinded to subject details.
Assessment of physical activity. Lifetime physical activity patterns were quantified using a questionnaire10 modeled after the Harvard Alumni physical activity questionnaire of Paffenbarger and colleagues. The questionnaire was administered by a trained research assistant. The questionnaire responses were entered into a Microsoft Access program (created by JM; Microsoft, Redmond, Wash). The program was used to record the metabolic costs of recreational activities, and to express the results as energy expenditure in kcal/week.11 Energy costs of activities were estimated from the compendium of physical activities developed by Ainsworth and colleagues.12 Energy cost of stairs climbed per week was calculated using the estimation of Basset et al.13 For reference, one flight of stairs climbed was considered 10 steps, and 12 blocks was considered one mile. Energy expenditure was expressed in terms of lifetime adulthood recreational activity, and recreational activity was also expressed separately as energy expended during the year prior to undergoing angiogram (recent activity). We focused on recreational activity as it has been shown to have a much more consistent correlation to outcomes in previous studies.14 Questionnaires examining lifetime exposure to activity have good reproducibility.15 The study was approved by the Institutional Review Boards at Stanford University and Mt. Sinai, and registered at http:// clinicaltrials.gov (NCT 00380185). Written informed consent was obtained from each patient. The authors had full access to and take full responsibility for the integrity of the data, and all authors have read and agreed to the manuscript as written. Statistical analysis. Data are reported as mean standard deviation unless otherwise indicated. All data were examined for normality, and non-parametric tests were used where appropriate. To compare means of demographic data in Table I, analysis of covariance was used, correcting for age. For analysis of hsCRP, the data were log transformed and also corrected for age, body mass index (BMI), and gender. Glomerular filtration rate (GFR) was estimated using the modification of diet in renal disease formula.16 For correlation between dependents with mul-
JOURNAL OF VASCULAR SURGERY Volume , Number
Wilson et al 3
Table II. Peripheral arterial disease and lifetime recreational activity
No LRA (n 417) Ankle-brachial index Peripheral arterial disease (%) 0.94 0.21 25.6 (n 1.02 17 Q2 241) 0.18 (n Q3 241) (n Q4 241) (n Q5 241) P value .001 .001
1.03 0.18 15.8
1.02 0.18 16.2
1.02 0.17 13.7
LRA, Lifetime recreational activity; Q, quintile. Ankle-brachial index values are compared based on the categories of no lifetime recreational activity (ie, No LRA group) vs those subjects with any activity (ie, Q2-Q5). Peripheral arterial disease prevalence in the No LRA group was significantly different from that in the Q5 group.
Table III. Spearman correlation for lifetime activity groups
Spearman coefficient Age (years) Gender (% female) Body mass index Systolic blood pressure Diastolic blood pressure Triglycerides High-density lipoprotein cholesterol Low-density lipoprotein cholesterol Glucose Ankle brachial index Peripheral arterial disease (%) Pack years smoking Log C-reactive protein 0.048 0.220 0.047 0.013 0.044 0.043 0.026 0.028 0.035 0.167 0.109 0.016 0.125 P value .074 .001 .081 .624 .101 .120 .353 .323 .203 .001 .001 .550 .009
Table IV. Multivariate regression for ankle-brachial index
coefficient Age Female gender Body mass index Systolic blood pressure Diastolic blood pressure Estimated glomular filtration rate Fasting glucose Triglycerides High-density lipoprotein cholesterol Low-density lipoprotein cholesterol Pack years smoking LRA groups 0.160 0.159 0.046 0.082 0.002 0.033 0.137 0.068 0.050 0.027 0.181 0.097 P value .001 .001 .101 .014 .949 .254 .001 .020 .097 .352 .001 .001
tiple predictors, multivariate and logistic regression models were used with all variables prespecified prior to performance of the analysis. Due to the skewed distribution of LRA per week resulting from the large number of subjects reporting no significant recreational activity, subjects were divided into five groups; one with almost entirely sedentary and no activity (n 417), and quartiles of activity (n 241) each. The correspondent values for no activity, mild, moderate, high, and vigorous LRA per week were 0, 573, 573 to 1190, 1191 to 2256, and 2256 kcal/week, respectively. SPSS software (version 15.0; SPSS Inc, Chicago, Ill) was used to perform all statistical calculations. RESULTS Table I describes selected baseline characteristics for groups according to the LRA. Notably, 417 patients (30%) in our study population reported an absence of any LRA. There were no significant differences between activity groups in terms of age, BMI, systolic blood pressure, lipids, glucose, pack years smoking, or hsCRP. Subjects reporting no regular LRA had greater levels of diastolic blood pressure and were more likely to be female (P .001). A higher percentage of women were entirely sedentary (ie, reporting the absence of any LRA [P .001]). PAD was present in 258 subjects (19%). We observed that the least active subjects (those reporting no activity) had a significantly lower ABI than the category of those reporting any level of LRA (0.94 0.21 vs 1.02 0.17; P .001). Furthermore, subjects with the greatest LRA had lower prevalence of PAD compared with sedentary subjects
LRA, Lifetime recreational activity. LRA Groups, lifetime recreational activity divided into five groups: one group with no LRA (n 417) and four quartiles of those who reported any activity (n 241).
(13.7% vs 25.6%; P .001). The reduction in the prevalence of PAD between the groups was not linear; the largest reduction occurred between the group with essentially no recreational activity and the next least active group, with smaller differences observed between the other activity groups (Table II). Table III shows the Spearman correlation between LRA groups and demographic and clinical characteristics. There is a weak negative correlation between LRA and gender ( 0.220; P .001) and a weak inverse correlation between LRA and log hsCRP ( 0.125; P .009). LRA was a significant correlate of PAD prevalence ( 0.109; P .001) and ABI ( 0.167; P .001). As previous studies have documented, the independent predictors of ABI include age, systolic blood pressure, fasting glucose, triglycerides, and history of tobacco use (Table IV). Of interest, we find that female gender and history of no LRA are also independent predictors of ABI (P .001). Similar findings were observed after assessment of independent predictors for the categorical variable of PAD by logistic regression (Table V). Based on standardized beta coefficients, the rank order of independent predictors was higher fasting glucose, female gender, tobacco use, older age, history of no LRA, lower GFR, triglycerides, and lower BMI. Subjects with history of sedentary lifestyle had a 1.46-fold (95% confidence interval [CI], 1.0112.103) greater odds ratio (OR) for PAD than subjects with
4 Wilson et al
JOURNAL OF VASCULAR SURGERY 2011
Table V. Logistic regression for diagnosis of peripheral arterial disease
Variable Age Female gender Body mass index Systolic blood pressure Diastolic blood pressure Estimated glomular filtration rate Fasting glucose Triglycerides High-density lipoprotein cholesterol Low-density lipoprotein cholesterol Pack years smoking History of lifetime recreational activity vs no activity Standardized B coefficient 0.433 0.516 0.192 0.098 0.089 0.287 0.557 0.253 0.152 0.113 0.443 0.377 Odds ratio (95% confidence interval) 1.041 (1.022-1.061) 1.676 (1.135-2.474) 0.968 (0.939-0.999) 1.004 (0.996-1.013) 0.993 (0.978-1.009) 0.989 (0.982-0.996) 1.009 (1.006-1.013) 1.003 (1.000-1.006) 0.988 (0.972-1.004) 1.004 (0.998-1.009) 1.018 (1.012-1.024) 0.686 (0.475-0.990) P value .001 .009 .041 .333 .388 .002 .001 .029 .132 .209 .001 .044
physical activity (P .044; the OR for sedentary lifestyle is simply the reciprocal of the OR created by comparing the prevalence of PAD in the four LRA categories of activity combined, to that in the LRA category of no activity). In terms of the relationship between current symptoms and recalled activity, there was no significant correlation between current symptoms of leg pain when walking and reported LRA ( 0.009; P .752). Additionally, there was no association between current leg pain and recent activity ( 0.024; P .377). DISCUSSION Our study is the first to reveal that LRA is associated with prevalent PAD. The findings are more remarkable in that this effect was observed in a population already at high risk for atherosclerosis (eg, individuals undergoing diagnostic coronary angiography). We found that the main risk appeared to be in the group with no reported LRA (ie, the most sedentary). This suggests that even modest recreational activity may reduce the prevalence of PAD. Various epidemiologic studies have suggested that increased adulthood recreational physical activity is protective against CAD and its risk factors, as well as all-cause mortality.17-19 For example, the Copenhagen City Heart study revealed that long-term moderate or high intensity physical activity was associated with a reduction in mortality from coronary heart disease, cancer, and all causes.20 Our study adds important new information regarding the association of recreational activity and the prevalence of PAD. Although it is well-known that patients with PAD are sedentary, is this a cause or an effect of the disease? For example, in a small twin study, Carmelli et al21 found that twins with PAD were more likely to have low physical activity and to be smokers than their non-PAD siblings. However, in this and most other studies, there is little or no information about the physical activity of these individuals prior to the development of PAD. In the Edinburgh Artery Study, which was a community based cross-sectional study, a history of strenuous activity between 35 and 45 years of age was associated with a reduction in the prevalence of PAD,22 but only in male smokers. Our study substantially extends these findings since we examined cumulative life-
time exposure to leisure activity in a large population at high risk of atherosclerosis. We have shown that low LRA is associated with prevalent PAD and low ABI in a group of patients referred for coronary angiography. This association was not simply due to an increase in traditional risk factors since it was independent in multivariate models. The association was most pronounced in those who gave a history of being sedentary throughout life. We observed a weak inverse correlation between hsCRP and lifetime physical activity groups in our high-risk population (patients presenting for coronary angiography). This inverse relationship between physical activity and markers of inflammation also has been observed in lower risk populations.23-25 An association also exists between parameters of cardio-respiratory fitness and CRP.26 Furthermore, hsCRP levels are extremely low in ultra marathon runners independent of adiposity,27 and CRP levels can be reduced by exercise training in sedentary adult subjects.28 A similar association has been shown between plasma levels of inflammatory markers or D-dimers and current physical activity (measured by accelerometer) in PAD patients.29 The other independent risk factors for PAD in this study were increasing age, systolic blood pressure, packyears smoking, and features of insulin resistance and diabetes such as hyperglycemia and elevated triglycerides. This is consistent with other previous studies suggesting that cigarette smoking and diabetes are significant risk factors for PAD.30,31 In fact, in our study, we saw these associations even in a population with a high burden of CAD and risk factors. Intriguingly, we find that female gender, in addition to a history of no LRA, is an independent risk factor for PAD. This association is not simply due to an increase in traditional risk factors since the association is independent in multivariable regression models. We have also found that there is a correlation between physical activity and gender, such that women with low lifetime recreational activity are at greatest risk of PAD. Previous studies have documented gender differences in the effect of PAD on quality of life. By comparison to men, women with PAD have a higher prevalence of leg pain and greater walking impairment.39 Furthermore, health-related quality of life, particularly physical
JOURNAL OF VASCULAR SURGERY Volume , Number
Wilson et al 5
functioning, is greatly compromised in women with PAD.32 When considering recreational activity as a binomial variable, the increase in risk of PAD in our study population due to a sedentary lifestyle was approximately 1.5-fold when including all other risk factors. The mechanism by which sedentary state may predispose to PAD cannot be determined from this study, but there are several physiological possibilities. Exercise may enhance endothelial function, by upregulating vasoprotective pathways such as heme oxygenase, superoxide dismutase, and nitric oxide synthase, and by downregulating the expression of proteins mediating vascular inflammation and thrombosis.7,8 In patients with PAD, regular activity has been associated with improvements in endogenous fibrinolysis,33 improved microcirculatory function,34 and reduced mortality.6 It has also been shown that patients with PAD who walk regularly experience less functional decline over time.6 Women who exercise strenuously seem protected against age-related increase in central arterial stiffness.35 In addition, physical training in mice increases the release of endothelial progenitor cells, inhibits neointima formation after balloon injury, and promotes angiogenesis.36 There may also be specific benefits in those at highest risk of PAD. For example, in a small study of physical activity in smokers, Anton et al37 found that regular activity was associated with higher femoral artery blood flow and higher arterial conductance. The Ankle Brachial Index Collaboration has provided data indicating that an ABI of 1.4 is a better cutpoint than 1.3 for predicting cardiovascular events and mortality in a general population.38 The Ankle Brachial Index Collaboration performed a data metaanalysis of relevant studies of prospective cohorts of participants from the general population. However, our subjects are from a population of patients undergoing cardiac catheterization. These patients are more likely than those drawn from a general population to have risk factors that increase the prevalence of fibrosis, medial calcification, and atherosclerosis. In this population of subjects, we reasoned that, by using an upper limit of 1.4 as a normal ABI, the risk of misclassifying patients might be increased. By using an upper limit for normality of 1.3, we only excluded 13 patients (1% of the total study population) that would have been included with the more liberal cutoff value. Study limitations. The Harvard Alumni Questionnaire, which is the basis for our assessment of LRA, is a validated instrument that has provided strong support for the association between physical activity patterns and cardiovascular morbidity and mortality.10 However, as with any questionnaire, the responses depend on patient recollection. It is possible that patient recall of lifetime physical activity may be biased by their recent symptoms. However, most patients with PAD are asymptomatic or have atypical symptoms.39,40 There is discordance between leg symptoms and clinical evidence of PAD. Indeed, in our study there was no significant correlation between exerciseinduced leg pain and LRA or recent activity. This suggests that the results observed were not due to recall bias.
Our questionnaire assesses overall adulthood activity patterns and does not quantify activity by recent or remote decades. Accordingly, we cannot perform sensitivity analyses excluding LRA data from recent decades to see if the relationship of PAD is maintained with LRA in earlier decades. In addition, we have only established a correlation between LRA and the prevalence of PAD. It is possible that other factors (other than the traditional risk factors for which we accounted) associated with a sedentary lifetime were causal in the later onset of PAD. We studied a population of subjects that is already at high risk of arterial disease, a group enriched in patients with cardiovascular risk factors. As a result, the range of these variables may be more narrow than in the general population, which would tend to reduce the strength of the observed associations. This may mean that the association between lifetime recreational activity and PAD/ABI may be stronger in the general population. CONCLUSIONS Low levels of LRA were associated with low ABI and with the diagnosis of PAD, independently of traditional risk factors, in patients undergoing cardiac catheterization. Subjects in the least active group were more likely to be female and had higher diastolic blood pressure. The risk of PAD was increased by 1.5-fold in those with no history of recreational activity, corrected for other cardiovascular disease risk factors. Strategies to increase recreational activity may reduce the burden of PAD later in life. For health care providers, a history of no lifetime recreational activity may heighten the clinical index of suspicion for PAD. AUTHOR CONTRIBUTIONS Conception and design: TA, JO, JC Analysis and interpretation: AW Data collection: MH, AG Writing the article: AW, AR, JC Critical revision of the article: JM, TA, JC Final approval of the article: AW, AR, JM, TA, MH, AG, JO, JC, KN Statistical analysis: AW, AR, KN Obtained funding: JO, JC Overall responsibility: JO, JC
REFERENCES
1. McDermott MM, Liu K, Greenland P, Guralnik JM, Criqui MH, Chan C, et al. Functional decline in peripheral arterial disease: associations with the ankle brachial index and leg symptoms. JAMA 2004;292: 453-61. 2. Hirsch AT, Criqui MH, Treat-Jacobson D, Regensteiner JG, Creager MA, Olin JW, et al. Peripheral arterial disease detection, awareness, and treatment in primary care. JAMA 2001;286:1317-24. 3. Newman AB, Shemanski L, Manolio TA, Cushman M, Mittelmark M, Polak JF, et al. Ankle-arm index as a predictor of cardiovascular disease and mortality in the Cardiovascular Health Study. The Cardiovascular Health Study Group. Arterioscler Thromb Vasc Biol 1999;19:538-45. 4. Hiatt WR, Hirsch AT, Regensteiner JG, Brass EP. Clinical trials for claudication. Assessment of exercise performance, functional status, and clinical end points. Vascular Clinical Trialists. Circulation 1995;92: 614-21.
6 Wilson et al
JOURNAL OF VASCULAR SURGERY 2011
5. McDermott MM, Liu K, Ferrucci L, Criqui MH, Greenland P, Guralnik JM, et al. Physical performance in peripheral arterial disease: a slower rate of decline in patients who walk more. Ann Intern Med 2006;144: 10-20. 6. Garg PK, Tian L, Criqui MH, Liu K, Ferrucci L, Guralnik JM, et al. Physical activity during daily life and mortality in patients with peripheral arterial disease. Circulation 2006;114:242-8. 7. Cooke JP. Flow, NO, and atherogenesis. Proc Natl Acad Sci U S A 2003;100: 768-70. 8. McAllister RM, Laughlin MH. Vascular nitric oxide: effects of physical activity, importance for health. Essays Biochem 2006;42:119-31. 9. Gardner AW, Clancy RJ. The relationship between ankle-brachial index and leisure-time physical activity in patients with intermittent claudication. Angiology 2006;57:539-45. 10. Myers J, Kaykha A, George S, Abella J, Zaheer N, Lear S, et al. Fitness vs. activity patterns in predicting mortality in men. Am J Med 2004; 117:912-8. 11. Myers J, Gullestad L, Bellin D, Ross H, Vagelos R, Fowler M, et al. Physical activity patterns and exercise performance in cardiac transplant recipients. J Cardiopulm Rehabil 2003;23:100-6. 12. Ainsworth BE, Haskell WL, Leon AS, Jacobs DR Jr, Montoye HJ, Sallis JF, et al. Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports Exerc 1993;25:71-80. 13. Bassett DR, Vachon JA, Kirkland AO, Howley ET, Duncan GE, Johnson KR, et al. Energy cost of stair climbing and descending on the college alumnus questionnaire. Med Sci Sports Exerc 1997;29:1250-4. 14. Myers J, Kaykha A, George S, Abella J, Zaheer N, Lear S, et al. Fitness versus physical activity patterns in predicting mortality in men. Am J Med 2004;117:912-8. 15. Mandic S, Myers JN, Oliveira RB, Abella JP, Froelicher VF. Characterizing differences in mortality at the low end of the fitness spectrum. Med Sci Sports Exerc 2009;41:1573-9. 16. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461-70. 17. Myers J, Prakash M, Froelicher VF, Kalisetti D, Atwood JE. Exercise capacity and all-cause mortality in patients referred for exercise testing. New Engl J Med 2002;346:793-801. 18. Kokkinos P, Myers J, Faselis C, Panagiotakos DB, Doumas M, Pittaras A, et al. Exercise capacity and mortality in older men: a 20-year follow-up study. Circulation 2010;122:790-7. 19. Kokkinos P, Myers J, Pittaras A, Kokkinos JP, Narayan P, Manolis A, et al. Exercise capacity and mortality in black and white men. Circulation 2008;117:614-22. 20. Schnohr P, Lange P, Scharling H, Jensen JS. Long-term physical activity in leisure time and mortality from coronary heart disease, stroke, respiratory diseases, and cancer. The Copenhagen City Heart Study. Eur J Cardiovasc Prev Rehabil 2006;13:173-9. 21. Carmelli D, Fabsitz RR, Swan GE, Reed T, Miller B, Wolf PA. Contribution of genetic and environmental influences to ankle-brachial blood pressure index in the NHLBI Twin Study. Lung: National Heart, and Blood Institute. Am J Epidemiol 2000;151:452-8. 22. Housley E, Leng GC, Donnan PT, Fowkes FG. Physical activity and risk of peripheral arterial disease in the general population: Edinburgh Artery Study. J Epidemiol Comm Health 1993;47:475-80. 23. Hjelstuen A, Anderssen SA, Holme I, Seljeflot I, Klemsdal TO. Markers of inflammation are inversely related to physical activity and fitness in sedentary men with treated hypertension. Am J Hypertens 2006;19: 669-75; discussion:676-7. 24. Elosua R, Bartali B, Ordovas JM, Corsi AM, Lauretani F, Ferrucci L, et al. Association between physical activity, physical performance, and
25.
26.
27.
28.
29.
30. 31. 32.
33.
34.
35.
36.
37.
38.
39.
40.
inflammatory biomarkers in an elderly population: the InCHIANTI study. J Gerontol A Biol Sci Med Sci 2005;60:760-7. Pitsavos C, Panagiotakos DB, Chrysohoou C, Kavouras S, Stefanadis C. The associations between physical activity, inflammation, and coagulation markers, in people with metabolic syndrome: the ATTICA study. Eur J Cardiovasc Prev Rehabil 2005;12:151-8. McGavock JM, Mandic S, Vonder Muhll I, Lewanczuk RZ, Quinney HA, Taylor DA, et al. Low cardiorespiratory fitness is associated with elevated C-reactive protein levels in women with type 2 diabetes. Diabetes Care 2004;27:320-5. Tomaszewski M, Charchar FJ, Przybycin M, Crawford L, Wallace AM, Gosek K, et al. Strikingly low circulating CRP concentrations in ultramarathon runners independent of markers of adiposity: how low can you go? Arterioscler Thromb Vasc Biol 2003;23:1640-4. Obisesan TO, Leeuwenburgh C, Phillips T, Ferrell RE, Phares DA, Prior SJ, et al. C-reactive protein genotypes affect baseline, but not exercise training-induced changes, in C-reactive protein levels. Arterioscler Thromb Vasc Biol 2004;24:1874-9. McDermott MM, Greenland P, Guralnik JM, Ferrucci L, Green D, Liu K, et al. Inflammatory markers, D-dimer, pro-thrombotic factors, and physical activity levels in patients with peripheral arterial disease. Vasc Med 2004;9:107-15. Criqui MH. Peripheral arterial disease– epidemiological aspects. Vasc Med 2001;6(3 Suppl):3-7. Hiatt WR. Medical treatment of peripheral arterial disease and claudication. N Engl J Med 2001;344:1608-21. McDermott MM, Green D, Greenland P, Liu K, Criqui MH, Chan C, et al. Relation of levels of hemostatic factors and inflammatory markers to the ankle brachial index. Am J Cardiol 2003;92:194-9. Gardner AW, Killewich LA. Association between physical activity and endogenous fibrinolysis in peripheral arterial disease: a cross-sectional study. Angiology 2002;53:367-74. Gardner AW, Killewich LA, Katzel LI, Womack CJ, Montgomery PS, Otis RB, et al. Relationship between free-living daily physical activity and peripheral circulation in patients with intermittent claudication. Angiology 1999;50:289-97. Tanaka H, DeSouza CA, Seals DR. Absence of age-related increase in central arterial stiffness in physically active women. Arterioscler Thromb Vasc Biol 1998;18:127-32. Laufs U, Werner N, Link A, Endres M, Wassmann S, Jürgens K, et al. Physical training increases endothelial progenitor cells, inhibits neointima formation, and enhances angiogenesis. Circulation 2004;109: 220-6. Anton MM, Cortez-Cooper MY, DeVan AE, Neidre DB, Cook JN, Tanaka H, et al. Cigarette smoking, regular exercise, and peripheral blood flow. Atherosclerosis 2006;185:201-5. Fowkes FG, Murray GD, Butcher I, Heald CL, Lee RJ, Chambless LE, et al; Ankle Brachial Index Collaboration. Ankle brachial index combined with Framingham Risk Score to predict cardiovascular events and mortality: a meta-analysis. JAMA 2008;300:197-208. McDermott MM, Greenland P, Liu K, Guralnik JM, Criqui MH, Dolan NC, et al. Leg symptoms in peripheral arterial disease: associated clinical characteristics and functional impairment. JAMA 2001;286:1599-606. McDermott MM, Greenland P, Liu K, Guralnik JM, Celic L, Criqui MH, et al. The ankle brachial index is associated with leg function and physical activity: the Walking and Leg Circulation Study. Ann Intern Med 2002;136:873-83.
Submitted Sep 30, 2010; accepted Feb 20, 2011.
Additional material for this article may be found online at www.jvascsurg.org.
JOURNAL OF VASCULAR SURGERY Volume , Number
Wilson et al 6.e1
Appendix (online only). Lifetime physical activity questionnaire.
6.e2 Wilson et al
JOURNAL OF VASCULAR SURGERY 2011
JOURNAL OF VASCULAR SURGERY Volume , Number
Wilson et al 6.e3
6.e4 Wilson et al
JOURNAL OF VASCULAR SURGERY 2011
Andrew M. Wilson, MBBS, PhD,a Amir H. Sadrzadeh-Rafie, MD,b Jonathan Myers, PhD,b Themistocles Assimes, MD, PhD,b Kevin T. Nead, BS,b Mamie Higgins, BS,b Andre Gabriel, MD,c Jeffrey Olin, DO,c and John P. Cooke, MD, PhD,b Melbourne, Australia; Stanford, Calif; and New York, NY
Background: The relationship between lifetime physical activity and the risk of developing peripheral arterial disease (PAD) is not known. Methods: We studied 1381 patients referred for elective coronary angiography in a point prevalence analysis. PAD was defined as ankle-brachial index (ABI) <0.9 at the time or a history of revascularization of the lower extremities regardless of ABI measure. We used a validated physical activity questionnaire to retrospectively measure each patient’s lifetime recreational activity (LRA). Multivariate and logistic regression analyses were used to assess the independent association of LRA to ABI and the presence of PAD. Results: PAD was present in 19% (n 258) of all subjects. Subjects reporting no regular LRA had greater diastolic blood pressure and were more likely to be female. They had lower average ABI, and a higher proportion had PAD (25.6%). In a regression model, including traditional risk factors and LRA, multivariate analysis showed that age (P < .001), female gender (P < .001), systolic blood pressure (P .014), fasting glucose (P < .001), serum triglycerides (P .02), and cumulative pack years (P < .001) were independent negative predictors of ABI, and LRA was a positive predictor of ABI (P < .001). History of sedentary lifestyle independently increased the odds ratio for PAD (odds ratio, 0.46; 95% confidence interval, 1.01-2.10) when assessed by logistic regression. Intriguingly, there is a correlation between physical activity and gender, such that women with low LRA are at greatest risk. Conclusion: Recalled LRA is positively correlated to ABI and associated with PAD. Whereas the mechanism for this effect is not clear, LRA may be a useful clinical screening tool for PAD risk, and strategies to increase adult recreational activity may reduce the burden of PAD later in life. ( J Vasc Surg 2011; : .)
Peripheral arterial disease of the lower extremities (PAD) is a common disorder affecting 8 to 12 million individuals in the United States. PAD causes limb pain with exertion, and reduces functional capacity and quality of life1 and is frequently associated with atherosclerotic disease in other vascular beds.2 Patients with PAD are at increased risk from myocardial infarction, cerebrovascular events, aortic aneurysm rupture, and death from cardiovascular causes, as well as ischemic ulceration and amputation.3 In patients with PAD, exercise therapy improves physical functioning.4 Conversely, lack of activity is a risk for disease progression, loss of function,5 and adverse cardiovascular
From the Department of Medicine, University of Melbourne, Melbournea; Stanford University Medical Center, Stanfordb; and Mount Sinai Medical Center, New York.c This study was supported in part by grants from the National Heart, Lung and Blood Institute (RO 1 HL-075774 and K12 HL087746; JPC), and NIH grant M01 RR 00070 (General Clinical Research Center, Stanford University School of Medicine). Competition of interest: none. Additional material for this article may be found online at www.jvascsurg. org. Reprint requests: John P. Cooke, MD, PhD, Professor of Medicine, Stanford University Medical Center, 300 Pasteur Dr, Falk Cardiovascular Research Center, Stanford, CA 94305-5406 (e-mail: john.cooke@ stanford.edu). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest. 0741-5214/$36.00 Copyright © 2011 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2011.02.052
disease outcome.6 It is known that exercise has a favorable effect on endothelial function, in part via beneficial effects on nitric oxide (NO) metabolism.7,8 Lack of leisure time physical activity within the year prior to examination is associated with a lower ABI.9 This may be due to the patients’ inability to participate in moderate and high intensity activities due to claudication. However, it is unclear whether the typically sedentary nature of the PAD patient is merely a manifestation of the disease, or may also have an etiologic role. There are no data available regarding the role of lifetime physical activity in modulating the development of PAD. To this end, we aimed to assess the relative impact of lifetime recreational activity (LRA) on the prevalence of PAD in a group of patients referred for coronary angiography. METHODS Participants. The National Heart, Lung and Blood Institute-funded study, “Genetic Determinants of Peripheral Arterial Disease,” is an observational study to delineate the determinants of atherosclerotic plaque distribution. Subjects entered into the study are extensively characterized to include ethnicity, lifestyle, physical function, medical and family history, traditional risk factors, and proteomic and genomic information. Subjects were recruited from patients undergoing elective coronary arteriograms at Stanford University and Mt. Sinai Medical Centers. A point prevalence analysis was performed for 1381 subjects (mean [ SD] age, 65 11 years) who had been recruited between April 2004 and January 2008. Indications for cath1
2 Wilson et al
JOURNAL OF VASCULAR SURGERY 2011
Table I. Subject demographics and lifetime recreational activity
No LRA (n 417) Age, years Gender, % F Body mass index (kg/m2) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) High-density lipoprotein (mg/dL) Triglycerides (mg/dL) Low-density lipoprotein (mg/dL) Glucose (mg/dL) Pack years smoking Log C-reactive protein (mg/dL) 66 11 51.8 29.5 6.4 149 26 77 12 42 13 117 90 80 31 122 54 17 27 0.3 0.6 (n Q2 241) (n Q3 241) (n Q4 241) (n Q5 241) P value .147 .001 .175 .116 .001 .129 .494 .088 .087 .999 .107
241]).
66 10 39.4 29 5.7 145 22 73 14 40 12 110 62 75 31 124 101 17 23 0.2 0.5
66 11 25.7 28.8 5.9 145 20 74 13 40 11 114 73 75 29 117 43 16 23 0.2 0.6
66 11 27.8 28.6 5.1 149 22 75 14 40 12 106 69 80 34 112 35 16 23 0.2 0.6
64 11 20.3 28.7 6.5 147 23 75 12 42 12 113 85 77 28 114 43 16 24 0.2 0.5
LRA, Lifetime recreational activity; Q, quintile. Variables are compared based on the quintiles of LRA (one group with no LRA [n Analysis of covariance was used to generate the P value with age adjustment.
417] and four quartiles of those who reported any activity [n
eterization included angina, shortness of breath, abnormal stress test, or known coronary artery disease (CAD). Patients admitted for emergent catheterizations, or screening catheterizations prior to organ transplants were excluded from the study. Patients 40 years old and those with language barriers, a history of radiation treatment, and known chronic infectious diseases such as HIV and hepatitis B and C were also excluded. Study assessments. Information about demographic factors, medications, personal and familial history of cardiovascular disease, walking impairment questionnaire, and social history were obtained using structured questionnaires administered by study personnel. Prior to the coronary angiogram, posterior tibial, dorsalis pedis, and brachial artery systolic pressures were measured using a 5 MHz Doppler ultrasound. One arm was used for the brachial pressures due to the presence of an intravenous line in the other arm. Ankle-brachial indexes (ABI) were calculated by dividing the higher ankle pressure of each leg over the brachial pressure. Each patient was then classified as having PAD (ABI 0.9 in either leg) or not having PAD (ABI 0.9 in both legs). Patients with ABI 1.3 or noncompressible arteries were not included in the continuous analysis. Patients with normal ABI ( 0.9) but with a history of PAD-related surgery, angioplasty, or stenting were also classified as having PAD, but the data for ABI were not included in the continuous analysis. Past history of PAD procedure was included in the logistic regression, however. Finally, 30 mL of blood was collected from each patient through a venous or arterial femoral sheath, or from an intravenous line in the arm. The blood specimens were centrifuged at 3000 RPM for 20 minutes at 4 degrees Celsius. Aliquots of EDTA plasma and serum were stored at 75 degrees Celsius. Laboratory measurements were obtained from the Stanford and Mt. Sinai hospital clinical laboratories, and high sensitivity C-reactive protein (hsCRP) was measured for a subset of the cohort (n 580) at the Boston Children’s Hospital (Dr Nader Rifai). Each angiogram was assessed by a board-certified interventional cardiologist blinded to subject details.
Assessment of physical activity. Lifetime physical activity patterns were quantified using a questionnaire10 modeled after the Harvard Alumni physical activity questionnaire of Paffenbarger and colleagues. The questionnaire was administered by a trained research assistant. The questionnaire responses were entered into a Microsoft Access program (created by JM; Microsoft, Redmond, Wash). The program was used to record the metabolic costs of recreational activities, and to express the results as energy expenditure in kcal/week.11 Energy costs of activities were estimated from the compendium of physical activities developed by Ainsworth and colleagues.12 Energy cost of stairs climbed per week was calculated using the estimation of Basset et al.13 For reference, one flight of stairs climbed was considered 10 steps, and 12 blocks was considered one mile. Energy expenditure was expressed in terms of lifetime adulthood recreational activity, and recreational activity was also expressed separately as energy expended during the year prior to undergoing angiogram (recent activity). We focused on recreational activity as it has been shown to have a much more consistent correlation to outcomes in previous studies.14 Questionnaires examining lifetime exposure to activity have good reproducibility.15 The study was approved by the Institutional Review Boards at Stanford University and Mt. Sinai, and registered at http:// clinicaltrials.gov (NCT 00380185). Written informed consent was obtained from each patient. The authors had full access to and take full responsibility for the integrity of the data, and all authors have read and agreed to the manuscript as written. Statistical analysis. Data are reported as mean standard deviation unless otherwise indicated. All data were examined for normality, and non-parametric tests were used where appropriate. To compare means of demographic data in Table I, analysis of covariance was used, correcting for age. For analysis of hsCRP, the data were log transformed and also corrected for age, body mass index (BMI), and gender. Glomerular filtration rate (GFR) was estimated using the modification of diet in renal disease formula.16 For correlation between dependents with mul-
JOURNAL OF VASCULAR SURGERY Volume , Number
Wilson et al 3
Table II. Peripheral arterial disease and lifetime recreational activity
No LRA (n 417) Ankle-brachial index Peripheral arterial disease (%) 0.94 0.21 25.6 (n 1.02 17 Q2 241) 0.18 (n Q3 241) (n Q4 241) (n Q5 241) P value .001 .001
1.03 0.18 15.8
1.02 0.18 16.2
1.02 0.17 13.7
LRA, Lifetime recreational activity; Q, quintile. Ankle-brachial index values are compared based on the categories of no lifetime recreational activity (ie, No LRA group) vs those subjects with any activity (ie, Q2-Q5). Peripheral arterial disease prevalence in the No LRA group was significantly different from that in the Q5 group.
Table III. Spearman correlation for lifetime activity groups
Spearman coefficient Age (years) Gender (% female) Body mass index Systolic blood pressure Diastolic blood pressure Triglycerides High-density lipoprotein cholesterol Low-density lipoprotein cholesterol Glucose Ankle brachial index Peripheral arterial disease (%) Pack years smoking Log C-reactive protein 0.048 0.220 0.047 0.013 0.044 0.043 0.026 0.028 0.035 0.167 0.109 0.016 0.125 P value .074 .001 .081 .624 .101 .120 .353 .323 .203 .001 .001 .550 .009
Table IV. Multivariate regression for ankle-brachial index
coefficient Age Female gender Body mass index Systolic blood pressure Diastolic blood pressure Estimated glomular filtration rate Fasting glucose Triglycerides High-density lipoprotein cholesterol Low-density lipoprotein cholesterol Pack years smoking LRA groups 0.160 0.159 0.046 0.082 0.002 0.033 0.137 0.068 0.050 0.027 0.181 0.097 P value .001 .001 .101 .014 .949 .254 .001 .020 .097 .352 .001 .001
tiple predictors, multivariate and logistic regression models were used with all variables prespecified prior to performance of the analysis. Due to the skewed distribution of LRA per week resulting from the large number of subjects reporting no significant recreational activity, subjects were divided into five groups; one with almost entirely sedentary and no activity (n 417), and quartiles of activity (n 241) each. The correspondent values for no activity, mild, moderate, high, and vigorous LRA per week were 0, 573, 573 to 1190, 1191 to 2256, and 2256 kcal/week, respectively. SPSS software (version 15.0; SPSS Inc, Chicago, Ill) was used to perform all statistical calculations. RESULTS Table I describes selected baseline characteristics for groups according to the LRA. Notably, 417 patients (30%) in our study population reported an absence of any LRA. There were no significant differences between activity groups in terms of age, BMI, systolic blood pressure, lipids, glucose, pack years smoking, or hsCRP. Subjects reporting no regular LRA had greater levels of diastolic blood pressure and were more likely to be female (P .001). A higher percentage of women were entirely sedentary (ie, reporting the absence of any LRA [P .001]). PAD was present in 258 subjects (19%). We observed that the least active subjects (those reporting no activity) had a significantly lower ABI than the category of those reporting any level of LRA (0.94 0.21 vs 1.02 0.17; P .001). Furthermore, subjects with the greatest LRA had lower prevalence of PAD compared with sedentary subjects
LRA, Lifetime recreational activity. LRA Groups, lifetime recreational activity divided into five groups: one group with no LRA (n 417) and four quartiles of those who reported any activity (n 241).
(13.7% vs 25.6%; P .001). The reduction in the prevalence of PAD between the groups was not linear; the largest reduction occurred between the group with essentially no recreational activity and the next least active group, with smaller differences observed between the other activity groups (Table II). Table III shows the Spearman correlation between LRA groups and demographic and clinical characteristics. There is a weak negative correlation between LRA and gender ( 0.220; P .001) and a weak inverse correlation between LRA and log hsCRP ( 0.125; P .009). LRA was a significant correlate of PAD prevalence ( 0.109; P .001) and ABI ( 0.167; P .001). As previous studies have documented, the independent predictors of ABI include age, systolic blood pressure, fasting glucose, triglycerides, and history of tobacco use (Table IV). Of interest, we find that female gender and history of no LRA are also independent predictors of ABI (P .001). Similar findings were observed after assessment of independent predictors for the categorical variable of PAD by logistic regression (Table V). Based on standardized beta coefficients, the rank order of independent predictors was higher fasting glucose, female gender, tobacco use, older age, history of no LRA, lower GFR, triglycerides, and lower BMI. Subjects with history of sedentary lifestyle had a 1.46-fold (95% confidence interval [CI], 1.0112.103) greater odds ratio (OR) for PAD than subjects with
4 Wilson et al
JOURNAL OF VASCULAR SURGERY 2011
Table V. Logistic regression for diagnosis of peripheral arterial disease
Variable Age Female gender Body mass index Systolic blood pressure Diastolic blood pressure Estimated glomular filtration rate Fasting glucose Triglycerides High-density lipoprotein cholesterol Low-density lipoprotein cholesterol Pack years smoking History of lifetime recreational activity vs no activity Standardized B coefficient 0.433 0.516 0.192 0.098 0.089 0.287 0.557 0.253 0.152 0.113 0.443 0.377 Odds ratio (95% confidence interval) 1.041 (1.022-1.061) 1.676 (1.135-2.474) 0.968 (0.939-0.999) 1.004 (0.996-1.013) 0.993 (0.978-1.009) 0.989 (0.982-0.996) 1.009 (1.006-1.013) 1.003 (1.000-1.006) 0.988 (0.972-1.004) 1.004 (0.998-1.009) 1.018 (1.012-1.024) 0.686 (0.475-0.990) P value .001 .009 .041 .333 .388 .002 .001 .029 .132 .209 .001 .044
physical activity (P .044; the OR for sedentary lifestyle is simply the reciprocal of the OR created by comparing the prevalence of PAD in the four LRA categories of activity combined, to that in the LRA category of no activity). In terms of the relationship between current symptoms and recalled activity, there was no significant correlation between current symptoms of leg pain when walking and reported LRA ( 0.009; P .752). Additionally, there was no association between current leg pain and recent activity ( 0.024; P .377). DISCUSSION Our study is the first to reveal that LRA is associated with prevalent PAD. The findings are more remarkable in that this effect was observed in a population already at high risk for atherosclerosis (eg, individuals undergoing diagnostic coronary angiography). We found that the main risk appeared to be in the group with no reported LRA (ie, the most sedentary). This suggests that even modest recreational activity may reduce the prevalence of PAD. Various epidemiologic studies have suggested that increased adulthood recreational physical activity is protective against CAD and its risk factors, as well as all-cause mortality.17-19 For example, the Copenhagen City Heart study revealed that long-term moderate or high intensity physical activity was associated with a reduction in mortality from coronary heart disease, cancer, and all causes.20 Our study adds important new information regarding the association of recreational activity and the prevalence of PAD. Although it is well-known that patients with PAD are sedentary, is this a cause or an effect of the disease? For example, in a small twin study, Carmelli et al21 found that twins with PAD were more likely to have low physical activity and to be smokers than their non-PAD siblings. However, in this and most other studies, there is little or no information about the physical activity of these individuals prior to the development of PAD. In the Edinburgh Artery Study, which was a community based cross-sectional study, a history of strenuous activity between 35 and 45 years of age was associated with a reduction in the prevalence of PAD,22 but only in male smokers. Our study substantially extends these findings since we examined cumulative life-
time exposure to leisure activity in a large population at high risk of atherosclerosis. We have shown that low LRA is associated with prevalent PAD and low ABI in a group of patients referred for coronary angiography. This association was not simply due to an increase in traditional risk factors since it was independent in multivariate models. The association was most pronounced in those who gave a history of being sedentary throughout life. We observed a weak inverse correlation between hsCRP and lifetime physical activity groups in our high-risk population (patients presenting for coronary angiography). This inverse relationship between physical activity and markers of inflammation also has been observed in lower risk populations.23-25 An association also exists between parameters of cardio-respiratory fitness and CRP.26 Furthermore, hsCRP levels are extremely low in ultra marathon runners independent of adiposity,27 and CRP levels can be reduced by exercise training in sedentary adult subjects.28 A similar association has been shown between plasma levels of inflammatory markers or D-dimers and current physical activity (measured by accelerometer) in PAD patients.29 The other independent risk factors for PAD in this study were increasing age, systolic blood pressure, packyears smoking, and features of insulin resistance and diabetes such as hyperglycemia and elevated triglycerides. This is consistent with other previous studies suggesting that cigarette smoking and diabetes are significant risk factors for PAD.30,31 In fact, in our study, we saw these associations even in a population with a high burden of CAD and risk factors. Intriguingly, we find that female gender, in addition to a history of no LRA, is an independent risk factor for PAD. This association is not simply due to an increase in traditional risk factors since the association is independent in multivariable regression models. We have also found that there is a correlation between physical activity and gender, such that women with low lifetime recreational activity are at greatest risk of PAD. Previous studies have documented gender differences in the effect of PAD on quality of life. By comparison to men, women with PAD have a higher prevalence of leg pain and greater walking impairment.39 Furthermore, health-related quality of life, particularly physical
JOURNAL OF VASCULAR SURGERY Volume , Number
Wilson et al 5
functioning, is greatly compromised in women with PAD.32 When considering recreational activity as a binomial variable, the increase in risk of PAD in our study population due to a sedentary lifestyle was approximately 1.5-fold when including all other risk factors. The mechanism by which sedentary state may predispose to PAD cannot be determined from this study, but there are several physiological possibilities. Exercise may enhance endothelial function, by upregulating vasoprotective pathways such as heme oxygenase, superoxide dismutase, and nitric oxide synthase, and by downregulating the expression of proteins mediating vascular inflammation and thrombosis.7,8 In patients with PAD, regular activity has been associated with improvements in endogenous fibrinolysis,33 improved microcirculatory function,34 and reduced mortality.6 It has also been shown that patients with PAD who walk regularly experience less functional decline over time.6 Women who exercise strenuously seem protected against age-related increase in central arterial stiffness.35 In addition, physical training in mice increases the release of endothelial progenitor cells, inhibits neointima formation after balloon injury, and promotes angiogenesis.36 There may also be specific benefits in those at highest risk of PAD. For example, in a small study of physical activity in smokers, Anton et al37 found that regular activity was associated with higher femoral artery blood flow and higher arterial conductance. The Ankle Brachial Index Collaboration has provided data indicating that an ABI of 1.4 is a better cutpoint than 1.3 for predicting cardiovascular events and mortality in a general population.38 The Ankle Brachial Index Collaboration performed a data metaanalysis of relevant studies of prospective cohorts of participants from the general population. However, our subjects are from a population of patients undergoing cardiac catheterization. These patients are more likely than those drawn from a general population to have risk factors that increase the prevalence of fibrosis, medial calcification, and atherosclerosis. In this population of subjects, we reasoned that, by using an upper limit of 1.4 as a normal ABI, the risk of misclassifying patients might be increased. By using an upper limit for normality of 1.3, we only excluded 13 patients (1% of the total study population) that would have been included with the more liberal cutoff value. Study limitations. The Harvard Alumni Questionnaire, which is the basis for our assessment of LRA, is a validated instrument that has provided strong support for the association between physical activity patterns and cardiovascular morbidity and mortality.10 However, as with any questionnaire, the responses depend on patient recollection. It is possible that patient recall of lifetime physical activity may be biased by their recent symptoms. However, most patients with PAD are asymptomatic or have atypical symptoms.39,40 There is discordance between leg symptoms and clinical evidence of PAD. Indeed, in our study there was no significant correlation between exerciseinduced leg pain and LRA or recent activity. This suggests that the results observed were not due to recall bias.
Our questionnaire assesses overall adulthood activity patterns and does not quantify activity by recent or remote decades. Accordingly, we cannot perform sensitivity analyses excluding LRA data from recent decades to see if the relationship of PAD is maintained with LRA in earlier decades. In addition, we have only established a correlation between LRA and the prevalence of PAD. It is possible that other factors (other than the traditional risk factors for which we accounted) associated with a sedentary lifetime were causal in the later onset of PAD. We studied a population of subjects that is already at high risk of arterial disease, a group enriched in patients with cardiovascular risk factors. As a result, the range of these variables may be more narrow than in the general population, which would tend to reduce the strength of the observed associations. This may mean that the association between lifetime recreational activity and PAD/ABI may be stronger in the general population. CONCLUSIONS Low levels of LRA were associated with low ABI and with the diagnosis of PAD, independently of traditional risk factors, in patients undergoing cardiac catheterization. Subjects in the least active group were more likely to be female and had higher diastolic blood pressure. The risk of PAD was increased by 1.5-fold in those with no history of recreational activity, corrected for other cardiovascular disease risk factors. Strategies to increase recreational activity may reduce the burden of PAD later in life. For health care providers, a history of no lifetime recreational activity may heighten the clinical index of suspicion for PAD. AUTHOR CONTRIBUTIONS Conception and design: TA, JO, JC Analysis and interpretation: AW Data collection: MH, AG Writing the article: AW, AR, JC Critical revision of the article: JM, TA, JC Final approval of the article: AW, AR, JM, TA, MH, AG, JO, JC, KN Statistical analysis: AW, AR, KN Obtained funding: JO, JC Overall responsibility: JO, JC
REFERENCES
1. McDermott MM, Liu K, Greenland P, Guralnik JM, Criqui MH, Chan C, et al. Functional decline in peripheral arterial disease: associations with the ankle brachial index and leg symptoms. JAMA 2004;292: 453-61. 2. Hirsch AT, Criqui MH, Treat-Jacobson D, Regensteiner JG, Creager MA, Olin JW, et al. Peripheral arterial disease detection, awareness, and treatment in primary care. JAMA 2001;286:1317-24. 3. Newman AB, Shemanski L, Manolio TA, Cushman M, Mittelmark M, Polak JF, et al. Ankle-arm index as a predictor of cardiovascular disease and mortality in the Cardiovascular Health Study. The Cardiovascular Health Study Group. Arterioscler Thromb Vasc Biol 1999;19:538-45. 4. Hiatt WR, Hirsch AT, Regensteiner JG, Brass EP. Clinical trials for claudication. Assessment of exercise performance, functional status, and clinical end points. Vascular Clinical Trialists. Circulation 1995;92: 614-21.
6 Wilson et al
JOURNAL OF VASCULAR SURGERY 2011
5. McDermott MM, Liu K, Ferrucci L, Criqui MH, Greenland P, Guralnik JM, et al. Physical performance in peripheral arterial disease: a slower rate of decline in patients who walk more. Ann Intern Med 2006;144: 10-20. 6. Garg PK, Tian L, Criqui MH, Liu K, Ferrucci L, Guralnik JM, et al. Physical activity during daily life and mortality in patients with peripheral arterial disease. Circulation 2006;114:242-8. 7. Cooke JP. Flow, NO, and atherogenesis. Proc Natl Acad Sci U S A 2003;100: 768-70. 8. McAllister RM, Laughlin MH. Vascular nitric oxide: effects of physical activity, importance for health. Essays Biochem 2006;42:119-31. 9. Gardner AW, Clancy RJ. The relationship between ankle-brachial index and leisure-time physical activity in patients with intermittent claudication. Angiology 2006;57:539-45. 10. Myers J, Kaykha A, George S, Abella J, Zaheer N, Lear S, et al. Fitness vs. activity patterns in predicting mortality in men. Am J Med 2004; 117:912-8. 11. Myers J, Gullestad L, Bellin D, Ross H, Vagelos R, Fowler M, et al. Physical activity patterns and exercise performance in cardiac transplant recipients. J Cardiopulm Rehabil 2003;23:100-6. 12. Ainsworth BE, Haskell WL, Leon AS, Jacobs DR Jr, Montoye HJ, Sallis JF, et al. Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports Exerc 1993;25:71-80. 13. Bassett DR, Vachon JA, Kirkland AO, Howley ET, Duncan GE, Johnson KR, et al. Energy cost of stair climbing and descending on the college alumnus questionnaire. Med Sci Sports Exerc 1997;29:1250-4. 14. Myers J, Kaykha A, George S, Abella J, Zaheer N, Lear S, et al. Fitness versus physical activity patterns in predicting mortality in men. Am J Med 2004;117:912-8. 15. Mandic S, Myers JN, Oliveira RB, Abella JP, Froelicher VF. Characterizing differences in mortality at the low end of the fitness spectrum. Med Sci Sports Exerc 2009;41:1573-9. 16. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461-70. 17. Myers J, Prakash M, Froelicher VF, Kalisetti D, Atwood JE. Exercise capacity and all-cause mortality in patients referred for exercise testing. New Engl J Med 2002;346:793-801. 18. Kokkinos P, Myers J, Faselis C, Panagiotakos DB, Doumas M, Pittaras A, et al. Exercise capacity and mortality in older men: a 20-year follow-up study. Circulation 2010;122:790-7. 19. Kokkinos P, Myers J, Pittaras A, Kokkinos JP, Narayan P, Manolis A, et al. Exercise capacity and mortality in black and white men. Circulation 2008;117:614-22. 20. Schnohr P, Lange P, Scharling H, Jensen JS. Long-term physical activity in leisure time and mortality from coronary heart disease, stroke, respiratory diseases, and cancer. The Copenhagen City Heart Study. Eur J Cardiovasc Prev Rehabil 2006;13:173-9. 21. Carmelli D, Fabsitz RR, Swan GE, Reed T, Miller B, Wolf PA. Contribution of genetic and environmental influences to ankle-brachial blood pressure index in the NHLBI Twin Study. Lung: National Heart, and Blood Institute. Am J Epidemiol 2000;151:452-8. 22. Housley E, Leng GC, Donnan PT, Fowkes FG. Physical activity and risk of peripheral arterial disease in the general population: Edinburgh Artery Study. J Epidemiol Comm Health 1993;47:475-80. 23. Hjelstuen A, Anderssen SA, Holme I, Seljeflot I, Klemsdal TO. Markers of inflammation are inversely related to physical activity and fitness in sedentary men with treated hypertension. Am J Hypertens 2006;19: 669-75; discussion:676-7. 24. Elosua R, Bartali B, Ordovas JM, Corsi AM, Lauretani F, Ferrucci L, et al. Association between physical activity, physical performance, and
25.
26.
27.
28.
29.
30. 31. 32.
33.
34.
35.
36.
37.
38.
39.
40.
inflammatory biomarkers in an elderly population: the InCHIANTI study. J Gerontol A Biol Sci Med Sci 2005;60:760-7. Pitsavos C, Panagiotakos DB, Chrysohoou C, Kavouras S, Stefanadis C. The associations between physical activity, inflammation, and coagulation markers, in people with metabolic syndrome: the ATTICA study. Eur J Cardiovasc Prev Rehabil 2005;12:151-8. McGavock JM, Mandic S, Vonder Muhll I, Lewanczuk RZ, Quinney HA, Taylor DA, et al. Low cardiorespiratory fitness is associated with elevated C-reactive protein levels in women with type 2 diabetes. Diabetes Care 2004;27:320-5. Tomaszewski M, Charchar FJ, Przybycin M, Crawford L, Wallace AM, Gosek K, et al. Strikingly low circulating CRP concentrations in ultramarathon runners independent of markers of adiposity: how low can you go? Arterioscler Thromb Vasc Biol 2003;23:1640-4. Obisesan TO, Leeuwenburgh C, Phillips T, Ferrell RE, Phares DA, Prior SJ, et al. C-reactive protein genotypes affect baseline, but not exercise training-induced changes, in C-reactive protein levels. Arterioscler Thromb Vasc Biol 2004;24:1874-9. McDermott MM, Greenland P, Guralnik JM, Ferrucci L, Green D, Liu K, et al. Inflammatory markers, D-dimer, pro-thrombotic factors, and physical activity levels in patients with peripheral arterial disease. Vasc Med 2004;9:107-15. Criqui MH. Peripheral arterial disease– epidemiological aspects. Vasc Med 2001;6(3 Suppl):3-7. Hiatt WR. Medical treatment of peripheral arterial disease and claudication. N Engl J Med 2001;344:1608-21. McDermott MM, Green D, Greenland P, Liu K, Criqui MH, Chan C, et al. Relation of levels of hemostatic factors and inflammatory markers to the ankle brachial index. Am J Cardiol 2003;92:194-9. Gardner AW, Killewich LA. Association between physical activity and endogenous fibrinolysis in peripheral arterial disease: a cross-sectional study. Angiology 2002;53:367-74. Gardner AW, Killewich LA, Katzel LI, Womack CJ, Montgomery PS, Otis RB, et al. Relationship between free-living daily physical activity and peripheral circulation in patients with intermittent claudication. Angiology 1999;50:289-97. Tanaka H, DeSouza CA, Seals DR. Absence of age-related increase in central arterial stiffness in physically active women. Arterioscler Thromb Vasc Biol 1998;18:127-32. Laufs U, Werner N, Link A, Endres M, Wassmann S, Jürgens K, et al. Physical training increases endothelial progenitor cells, inhibits neointima formation, and enhances angiogenesis. Circulation 2004;109: 220-6. Anton MM, Cortez-Cooper MY, DeVan AE, Neidre DB, Cook JN, Tanaka H, et al. Cigarette smoking, regular exercise, and peripheral blood flow. Atherosclerosis 2006;185:201-5. Fowkes FG, Murray GD, Butcher I, Heald CL, Lee RJ, Chambless LE, et al; Ankle Brachial Index Collaboration. Ankle brachial index combined with Framingham Risk Score to predict cardiovascular events and mortality: a meta-analysis. JAMA 2008;300:197-208. McDermott MM, Greenland P, Liu K, Guralnik JM, Criqui MH, Dolan NC, et al. Leg symptoms in peripheral arterial disease: associated clinical characteristics and functional impairment. JAMA 2001;286:1599-606. McDermott MM, Greenland P, Liu K, Guralnik JM, Celic L, Criqui MH, et al. The ankle brachial index is associated with leg function and physical activity: the Walking and Leg Circulation Study. Ann Intern Med 2002;136:873-83.
Submitted Sep 30, 2010; accepted Feb 20, 2011.
Additional material for this article may be found online at www.jvascsurg.org.
JOURNAL OF VASCULAR SURGERY Volume , Number
Wilson et al 6.e1
Appendix (online only). Lifetime physical activity questionnaire.
6.e2 Wilson et al
JOURNAL OF VASCULAR SURGERY 2011
JOURNAL OF VASCULAR SURGERY Volume , Number
Wilson et al 6.e3
6.e4 Wilson et al
JOURNAL OF VASCULAR SURGERY 2011
