ORIGINAL INVESTIGATION. Exertional Leg Symptoms Other Than Intermittent Claudication Are Common in Peripheral Arterial Disease

ORIGINAL INVESTIGATION Exertional Leg Symptoms Other Than Intermittent Claudication Are Common in Peripheral Arterial Disease Mary McGrae McDermott, MD; Shruti Mehta, BA; Philip Greenland, MD Background: Epidemiological data show that most community-dwelling men and women with lower-extremity peripheral arterial disease (PAD) do not have typical symptoms of intermittent claudication. We compared the prevalence of intermittent claudication, leg symptoms other than intermittent claudication, and absence of exertional leg symptoms between patients with PAD identified from a blood flow laboratory (group 1), patients with PAD in a general medicine practice (group 2), and control patients without PAD (group 3). Methods: Numbers of participants in groups 1, 2, and 3 were 137, 26, and 105, respectively. Patients with previously diagnosed PAD were excluded from groups 2 and 3. All participants underwent ankle-brachial index measurement and were administered the San Diego claudication questionnaire to assess leg symptoms. Results: Within groups 1, 2, and 3, prevalences of intermittent claudication were 28.5% (n = 39), 3.8% (n = 1), and 3.8% (n = 4), respectively. Prevalences of exertional leg symptoms other than intermittent claudication were 56.2% (n = 77), 42.3% (n = 11), and 19.0% (n = 20), respectively. Absence of exertional leg symptoms was reported by 15.3% (n = 21), 53.8% (n = 14), and 77.1% (n = 81), respectively. Among patients with PAD, older age, male sex, diabetes mellitus, and group 2 vs group 1 status were associated independently with absence of exertional leg symptoms in multivariable regression analysis. Lower ankle-brachial index levels and group 1 vs group 2 status were associated with intermittent claudication. Conclusions: Clinical manifestations of PAD are diverse, particularly among patients identified by anklebrachial index screening. Exertional leg symptoms other than intermittent claudication are common in PAD. Patients with PAD who are older, male, diabetic, or identified with ankle-brachial index screening in a primary care setting are more likely to have asymptomatic PAD. Arch Intern Med. 1999;159:387-392 From the Division of General Internal Medicine (Dr McDermott), Department of Medicine (Ms Mehta), and Department of Preventive Medicine (Drs McDermott and Greenland), Northwestern University Medical School, Chicago, Ill. LOWER-EXTREMITY peripheral arterial disease (PAD) can be accurately, noninvasively diagnosed with the ankle-brachial index (ABI), a ratio of Doppler-recorded systolic pressures in the lower and upper extremities. 1-5 Intermittent claudication has been considered the most characteristic manifestation of PAD and is defined as ischemic calf pain precipitated by walking and relieved within 10 minutes of rest. 6-9 However, in community-based epidemiological studies, most people with a low ABI consistent with PAD do not have classic intermittent claudication. 10-12 It is unclear whether these men and women with low ABI unaccompanied by claudication are asymptomatic or have leg symptoms other than claudication. Based on noninvasive testing with the ABI, a previous study showed that nearly 25% of men and women aged 55 years and older in general medicine practices have PAD. 13 A clearer understanding of the spectrum of leg symptoms associated with PAD may aid clinicians in recognizing and diagnosing PAD. Among patients with coronary artery disease, atypical symptoms and asymptomatic disease are frequent. 14-16 In a similar manner, atypical symptoms may also occur in PAD, but this issue has not been well studied. We studied the prevalence of classic intermittent claudication, exertional leg symptoms other than intermittent claudication, and absence of exertional leg symptoms among patients with PAD identified from a blood flow laboratory, patients with PAD identified with ABI This article is also available on our Web site: www.ama-assn.org/internal. 387

SUBJECTS AND METHODS SUBJECT IDENTIFICATION The study protocol was approved by the institutional review board. This study was part of a prospective study of lower-extremity functioning and PAD at Northwestern University Medical Center, Chicago, Ill. Consecutive patients aged 55 years and older diagnosed as having PAD at the study institution s blood flow laboratory between January 1, 1996, and approximately November 1, 1996, were identified by means of computerized lists and enrolled between January 1, 1996, and July 31, 1997. Men and women aged 55 years and older with upcoming appointments in general internal medicine were identified and randomly selected by means of a computer program between January 1, 1996, and March 31, 1998. Identified patients were mailed an informational letter describing the study, telephoned within 2 to 4 weeks after the mailing, and invited to return to the medical center for a study visit. Beginning in August 1997, patients with ABI of 0.90 or greater identified from general medicine were randomly selected for complete data collection. All patients with ABI less than 0.90 in general medicine who met inclusion criteria were enrolled. The prevalence of ABI less than 0.90 among men and women aged 55 years and older in general medicine without a history of PAD was 12%. To increase the prevalence of ABI less than 0.90, patients aged 60 years and older were enrolled from general medicine during the last 5 months of data collection. EXCLUSION CRITERIA We excluded patients with a Mini-Mental Status Examination score less than 18 of 30. 19 Nursing home residents, wheelchair-bound patients, and patients with foot or leg amputations were excluded, because these individuals have uniquely impaired functioning. Patients with lower-extremity ulcers were excluded, because blood pressure cuffs cannot be placed over ulcers for ABI measurement. Non English-speaking patients were excluded, because study investigators were not fluent in non-english languages, and patients with a life expectancy less than 6 months were excluded, because of their limited likelihood of follow-up. Patients with ABI of 1.50 or greater were excluded, because their arteries are often incompressible, preventing accurate assessment of lowerextremity arterial perfusion. Among patients identified from general medicine, those with a history of PAD were excluded. History of PAD was defined as (1) previous lowerextremity revascularization, (2) patient report of a physicianrendered diagnosis of intermittent claudication or poor leg circulation, or (3) medical record documentation of PAD. ABI MEASUREMENT The ABI was measured on all patients by previously described methods. 10 Patients were categorized according to their ABI level and recruitment site. Group 1 included patients identified from the blood flow laboratory with ABI less than 0.90. Group 2 consisted of patients identified from general medicine with ABI less than 0.90. Group 3 consisted of patients identified from general medicine with ABI of 0.90 or greater and less than 1.50. DEFINITIONS OF COMORBID DISEASE Medical record review and a primary care physician questionnaire were used to identify comorbidities, according to definitions and methods in the Women s Health and Aging Study. 20 Diabetes mellitus required at least 1 of the following: (1) use of insulin or an oral hypoglycemic agent, (2) medical record documented hemoglobin A 1c level greater than 10%, or (3) diabetes mellitus documented in the primary care physician questionnaire. Hypercholesterolemia was defined by use of lipid-lowering drugs or medical record documentation of a total cholesterol level greater than 6.45 mmol/l (250 mg/dl). Cigarette smoking was based on patient report. Regular exercise was defined as aerobic activity for the purpose of exercise performed at least 3 times weekly for at least 30 minutes each time. LEG SYMPTOM DEFINITIONS Patients leg symptoms were classified into categories by means of the San Diego claudication questionnaire, based on previous work by Criqui et al 21 (Table 1). The claudication screening in a general internal medicine practice, and patients without PAD at an academic medical center. Among patients with PAD, we identified clinical characteristics associated with (1) absence of exertional leg symptoms and (2) presence of intermittent claudication. On the basis of studies of patients with coronary artery disease, we hypothesized that older patients and patients with diabetes mellitus would be more likely to be asymptomatic. 17,18 We further hypothesized that lower ABIs and current cigarette smoking would be associated with intermittent claudication. RESULTS Of 485 consecutive patients aged 55 years and older with PAD identified in the blood flow laboratory during 1996, 74 met 1 or more exclusion criteria. In addition, 23 were deceased, 61 could not be reached, 82 refused participation, 37 were limited in transportation, 42 could not be scheduled during the enrollment period, and 29 did not keep study appointments, leaving a total of 137 patients with PAD in group 1. Of 411 randomly selected men and women aged 55 years and older with pending appointments in general internal medicine, 56 met exclusion criteria. Of the 56 excluded, 22 had a previous history of PAD. In addition, 109 refused participation, 12 had incorrect or disconnected phone numbers, 1 was deceased, and 33 did not keep study appointments, leaving 200 subjects. Of these, 69 had a normal ABI and were not randomly selected for further study, and 26 (13%) had an ABI less than 0.90 and were categorized into group 2. The remaining 105 patients with ABI of 0.90 or greater and less than 1.50 were categorized into group 3. 388

questionnaire was administered before ABI measurement. Study investigators were blinded to the presence vs absence of PAD when they administered the questionnaire to general medicine patients. Questionnaire administrators were not blinded to the presence vs absence of PAD among patients recruited from the blood flow laboratory. Leg symptom categories and their definitions were as follows 21 : (1) no exertional leg pain, defined as the absence of exertional leg pain, numbness, or discomfort; (2) atypical exertional leg pain, defined as (a) exertional calf symptoms that do not begin at rest but are otherwise not consistent with Rose intermittent claudication or (b) exertional leg symptoms that do not begin at rest and do not include the calves; (3) pain at rest: exertional leg symptoms that also begin at rest; and (4) Rose intermittent claudication: exertional calf symptoms that do not begin at rest, worsen when walking uphill or hurrying, and resolve within 10 minutes of rest. This definition of Rose claudication questionnaire is 91% sensitive and 67.5% specific for clinician-diagnosed claudication. 6 STATISTICAL ANALYSES We used 2 tests of association to compare the prevalence of clinical characteristics and leg symptoms between groups 1, 2, and 3. Because no patients had different types of leg symptoms in each leg, 1 P value was calculated to compare leg symptoms between the 3 groups. Logistic regression was used to identify independent predictors of absence of exertional leg symptoms and intermittent claudication among patients with PAD, controlling for sex, age, race, exercise, current smoking, hypercholesterolemia, diabetes mellitus, past lower-extremity revascularization, pentoxifylline use, ABI, and group 2 vs group 1 status. Because pentoxifylline is the only medication consistently associated with a significant effect on claudication symptoms, 22 we adjusted for pentoxifylline but not other medications in regression analyses. Additional logistic regression analyses were performed to determine whether exercise frequency or exercise duration was independently associated with leg symptoms. CHARACTERISTICS OF STUDY PATIENTS Table 2 compares clinical characteristics and the prevalence of leg symptoms between groups 1, 2, and 3. Compared with group 1, group 2 patients were less likely to have classic symptoms of intermittent claudication and more likely to have absence of exertional leg symptoms. REGRESSION ANALYSES Table 3 shows results of multivariable logistic regression analyses relating clinical characteristics to absence of exertional leg symptoms and presence of intermittent claudication, respectively, among PAD patients. Older age, male sex, diabetes mellitus, and group 2 vs group 1 status were associated independently with absence of exertional leg symptoms. Lower ABI and group 1 vs Table 1. San Diego Claudication Questionnaire* 1. Do you get pain, discomfort, or numbness in your legs when you walk? 2. Does this pain ever begin when you are standing still or sitting? 3. In what part of the leg or buttock do you feel it? (A) Includes calf/calves 1 1...Yes 2 2...No (B) Includes thigh/thighs 1 1...Yes 2 2...No (C) Includes buttock/buttocks 1 1...Yes 2 2...No 4. Do you get it when you walk uphill or hurry? 3 3...Never walk uphill or hurry 5. Do you get it if you walk at an ordinary pace on the level? 6. Does the pain ever disappear while you are still walking? 7. What do you do if you get this pain while you are walking? 1 1...Stop or slow down 2 2...Carry on 8. What happens to it if you stand still? 1 1...Lessens or relieved 2 2...Unchanged 9. How soon? 1 1...10 minutes or less 2 2...More than 10 minutes *Coding: (1) no exertional leg symptoms:1=2;(2)rest pain:1=1and 2 = 1; (3) non-rose exertional leg symptoms:1=1and2=2and3a=2and (3b=1or3c=1);alternatively,1=1and2=2and3a=2andnot Rose; and (4) Rose intermittent claudication:1=1and2=2and3a=1and(4=1or [4=3and5=1])and6=2and7=1and8=1and9=1. group 2 status were associated independently with intermittent claudication. In additional regression analyses, increasing exercise duration was associated significantly with intermittent claudication symptoms (odds ratio, 1.33; P =.03). There were no other independent relationships between exercise frequency or duration and intermittent claudication or absence of exertional leg symptoms. COMMENT Intermittent claudication is considered the most classic manifestation of PAD and is described as the earliest manifestation or most common symptom in PAD. 8,12,23-25 However, among white women participating in the Study of Osteoporotic Fractures, only 18% of 389

Table 2. Characteristics and Leg Symptoms Among Patients With Peripheral Arterial Disease and Controls* Characteristic Group 1 (n = 137) Group 2 (n = 26) Group 3 (n = 105) Trend P Age, mean ± SD, y 71.9 ± 9.8 69.0 ± 6.5 68.0 ± 7.2.002 Male sex, No. (%) 80 (58.4) 8 (30.8) 47 (44.8).01 Ankle-brachial index, mean ± SD 0.55 ± 0.18 0.71 ± 0.26 1.06 ± 0.09.001 African American race, No. (%) 19 (13.9) 12 (46.2) 23 (21.9).001 Regular exercise, No. (%) 35 (25.5) 7 (26.9) 34 (32.4).50 Current smokers, No. (%) 29 (21.2) 5 (19.2) 7 (6.7).007 Diabetes mellitus, No. (%) 38 (27.7) 6 (23.1) 17 (16.2).11 Hypercholesterolemia, No. (%) 82 (59.9) 13 (50.0) 44 (41.9).02 Leg symptoms, No. (%) No exertional leg symptoms 21 (15.3) 14 (53.8) 81 (77.1) NC Atypical exertional leg symptoms 35 (25.5) 2 (7.7) 5 (4.8).001 Pain at rest 42 (30.7) 9 (34.6) 15 (14.3) NC Intermittent claudication 39 (28.5) 1 (3.8) 4 (3.8) NC *Group 1, patients with peripheral arterial disease (ankle-brachial index 0.90) identified from the blood flow laboratory; group 2, patients with peripheral arterial disease identified from a general internal medicine practice via ankle-brachial index screening; and group 3, patients without peripheral arterial disease (ankle-brachial index, 0.90-1.50) identified from general internal medicine. Differences in leg symptoms between group 1, group 2, and group 3 are significant at P.001. All patients with bilateral exertional leg symptoms had the same type of symptoms in each leg. Therefore, the P value represents differences in all leg categories between groups 1, 2, and 3, and separate P values were not calculated (NC) for each category of leg symptoms. Percentages do not add to 100% because of rounding. Table 3. Multiple Logistic Regression Analysis Relating Clinical Characteristics to Absence of Exertional Leg Symptoms and Intermittent Claudication Among 163 Patients With Peripheral Arterial Disease* Characteristic OR for Absence of Exertional Leg Symptoms OR for Intermittent Claudication Age, per year 1.09 0.96 African American race 1.13 0.90 Female sex 0.27 1.92 Ankle-brachial index, per 0.4 unit 2.78 0.03 Lower-extremity revascularization 0.52 1.18 Diabetes mellitus 2.91 0.63 Current cigarette smoking 0.95 0.79 Group 2 status 15.78 0.09 Regular exercise 0.94 1.84 Pentoxifylline use 0.002 2.93 High cholesterol level 1.04 0.82 *All variables except age and ankle-brachial index are categorical variables. For each categorical variable, the odds ratio (OR) represents the OR for patients with vs without the specified condition. P.01. P.05. Patients identified from general internal medicine with ankle-brachial index screening (group 2). Reference group is patients identified from the blood flow laboratory (group 1). women found to have PAD by ABI screening had intermittent claudication. 12 Similarly, in the Cardiovascular Health Study, 12% of men and women aged 65 years and older had an abnormally low ABI, while just 2% had intermittent claudication symptoms. 10 It is unclear from studies of community-dwelling men and women, such as those in the Study of Osteoporotic Fractures and the Cardiovascular Health Study, whether people with low ABI who do not have intermittent claudication are asymptomatic or have leg symptoms other than classic intermittent claudication. The generalizability of findings from community-based studies to patients encountered by practicing clinicians is also unclear. Among participants with PAD in the Cardiovascular Health Study and the Study of Osteoporotic Fractures, 40% and 45%, respectively, had an ABI of 0.80 or greater and less than 0.90, indicative of mild disease. 10,12 In contrast, 14% of patients with PAD in the present study had an ABI of 0.80 or greater and less than 0.90. Thus, most patients with PAD described herein had more severe disease than the previously described men and women with PAD in epidemiological studies. The most striking implication of our data is that the clinical manifestations of PAD are diverse among patients with ABI less than 0.90 at an academic medical center. Identifying the full spectrum of PAD requires attention to symptoms other than classic intermittent claudication and possibly screening for PAD even among asymptomatic patients. In the present study, 29% of patients with PAD identified from the blood flow laboratory had classic intermittent claudication, compared with 4% of patients with PAD identified with ABI screening in a general internal medicine practice and 4% of normal controls. Thus, the sensitivity of intermittent claudication for PAD appears higher in a blood flow laboratory setting than among patients with PAD identified with ABI screening from a general medicine practice. This discrepancy is in part caused by selection bias, since all patients with PAD recruited from the blood flow laboratory were referred by a physician who suspected PAD, whereas patients identified from general internal medicine had previously unrecognized disease. The PAD may have been unrecognized among general medicine patients in part because 96% did not have typical intermittent claudication symptoms. The high prevalence of pain at rest among patients with PAD in a general medicine practice was unexpected and should be distinguished from PAD-related rest pain, a symptom of severe ischemic disease that typically increases when the affected leg is elevated. In our study, pain at rest was defined as any exertional leg pain that also occurs at rest. Since none of the patients with 390

PAD from general medicine had previously diagnosed disease, it is highly unlikely that these patients were experiencing severe ischemia. Perhaps patients with PAD recruited from general medicine were experiencing mild arterial ischemia at rest. Alternatively, they may have confused other types of pain occurring at rest with their exertional leg symptoms. While comorbid diseases may contribute to the high prevalence of exertional leg symptoms other than intermittent claudication in PAD, at least 2 factors make this unlikely. Aside from diabetic neuropathy, we know of no biologically plausible explanation for a higher prevalence of comorbidities causing exertional leg symptoms, such as spinal stenosis or lower-extremity arthritis, among patients with PAD as compared with patients without PAD. Although diabetic neuropathy might increase the prevalence of atypical exertional leg symptoms or pain at rest, our logistic regression analysis showed that diabetes mellitus was associated independently with asymptomatic PAD. Therefore, our comparisons of leg symptoms between PAD and control patients should represent valid differences in leg symptoms caused by the presence or absence of PAD. Our data suggest that among patients with PAD, older age, diabetes mellitus, male sex, and ABI screening in a general medicine practice are associated with absence of exertional leg symptoms. The association between asymptomatic PAD and older age may in part be caused by lower activity levels among the elderly, since greater physical activity levels may be necessary to induce exertional leg symptoms. Similarly, increasing exercise duration may be more likely to precipitate lower-extremity arterial ischemia, accounting for the positive relationship observed between exercise duration and intermittent claudication. Our data suggest that screening for PAD with the ABI may be especially useful among older, diabetic, male patients in a general medicine setting. Physical examination is neither sensitive nor specific for diagnosing PAD in communitybased or hospital settings. 11,26 We are aware of only 1 other report describing the spectrum of leg symptoms among patients with PAD in a patient care setting. 21 Criqui et al described leg symptoms in a blood flow laboratory population including 88% men and 4% African Americans. Criqui et al reported a 26% prevalence of intermittent claudication and a 25% prevalence of pain at rest. Although their results are generally consistent with ours, the 31% prevalence of asymptomatic PAD reported by Criqui et al 21 is twice that reported herein for patients from a blood flow laboratory. Our data add to the work of Criqui et al by assessing the prevalence of leg symptoms among patients with PAD in a general medical setting and control patients without PAD. Our study had a few limitations. Our data reflect patients at 1 medical center, and results may not be generalizable to patients with PAD at other institutions. Exclusion of non English-speaking patients may limit the generalizability of our findings to patients who do not speak English. However, the consistency of our findings for patients with PAD from a blood flow laboratory with those reported by Criqui et al suggests that our data are similar to those of other academic medical centers. 21 Second, many potentially eligible subjects could not be reached, refused participation, or had transportation difficulties, limiting study enrollment. Severity of PAD was comparable between potentially eligible patients with PAD and participating patients with PAD from the blood flow laboratory. Potentially eligible patients with PAD who refused participation were slightly older than participants with PAD from the blood flow laboratory. Because increasing age was associated with a lower prevalence of exertional leg symptoms, it is conceivable that our results underestimate the prevalence of asymptomatic PAD among patients at a blood flow laboratory. Finally, the absolute number of patients with PAD identified with ABI screening in general medicine was small. Further study is necessary to confirm our findings for patients with PAD identified with ABI screening from primary care physicians practices. The 13% prevalence of previously undiagnosed PAD among men and women in our general medicine practice suggests that PAD may regularly go unrecognized in primary care settings, perhaps because the high prevalence of asymptomatic disease and exertional leg symptoms other than intermittent claudication are underappreciated by clinicians. Recognizing PAD, even when asymptomatic, is important because PAD has significant implications for cardiovascular morbidity and mortality. 27 Our data highlight the need for a better understanding of the spectrum of clinical pictures presented by a broad range of patients with PAD. Accepted for publication May 28, 1998. Reprints not available from the authors. REFERENCES 1. Fowkes FGR. The measurement of atherosclerotic peripheral arterial disease in epidemiological surveys. Int J Epidemiol. 1988;17:248-254. 2. Carter SA. Indirect systolic pressures and pulse waves in arterial occlusive disease of the lower extremities. Circulation. 1968;37:624-638. 3. Yao JS. New techniques of objective arterial evaluation. Arch Surg. 1973;106: 600-604. 4. Winsor T. Influence of arterial disease on the systolic blood pressure gradients of the extremity. Am J Med Sci. 1950;220:117-126. 5. Bernstein EF, Fronek A. Current status of noninvasive tests in the diagnosis of peripheral arterial disease. Surg Clin North Am. 1982;62:473-487. 6. Leng GC, Fowkes FGR. The Edinburgh claudication questionnaire: an improved version of the WHO/Rose questionnaire for use in epidemiological surveys. Clin Epidemiol. 1992;10:1101-1109. 7. Rose GA. The diagnosis of ischaemic heart pain and intermittent claudication in field surveys. Bull World Health Organ. 1962;27:645-658. 8. Harrison TR, Isselbacher KJ, eds. Principles of Internal Medicine. 13th ed. New York, NY: McGraw Hill Book Co; 1994:1135. 9. Tierney LM, McPhee SJ, Papadakis MA, eds. Current Medical Diagnosis and Treatment. Norwalk, Conn: Appleton & Lange; 1995:396-397. 10. Newman AB, Siscovick DS, Manolio TA, et al. Ankle-arm index as a marker of atherosclerosis in the Cardiovascular Health Study. Circulation. 1993;88:837-845. 11. Criqui MH, Fronek A, Barrett-Connor E, Klauber MR, Gabriel S, Goodman D. The sensitivity, specificity, and predictive value of traditional clinical evaluation of peripheral arterial disease: results from noninvasive testing in a defined population. Circulation. 1985;71:516-522. 12. Vogt MT, Cauley JA, Newman AB, Kuller LH, Hulley SB. Decreased ankle/arm blood pressure index and mortality in elderly women. JAMA. 1993;270: 465-469. 13. Farkouh M, Oddone EZ, Simel DL, for the US-Canadian Research Group on the Clinical Examination. The clinical examination for peripheral arterial disease [abstract]. J Gen Intern Med. 1996;11:46. 391

14. Corti MC, Williamson JD, Phillips CL, et al. Cardiovascular diseases and diabetes. In: Guralnik JM, Fried LP, Simonsick EM, Kasper JD, Lafferty ME, eds. The Women s Health and Aging Study: Health and Social Characteristics of Older Women With Disability. Bethesda, Md: National Institute on Aging; 1995:96-105. NIH publication 95-4009. 15. Bayer AJ, Chadha JS, Farag RR, Pathy MS. Changing presentation of myocardial infarction with increasing old age. J Am Geriatr Soc. 1986;34:263-266. 16. Mittelmark MB, Psaty BM, Rautaharju PM, et al, for the Cardiovascular Health Study Collaborative Research Group. Prevalence of cardiovascular diseases among older adults: the Cardiovascular Health Study. Am J Epidemiol. 1993;137: 311-317. 17. De Bruyne MC, Mosterd A, Hoes AW, et al. Prevalence, determinants, and misclassification of myocardial infarction in the elderly. Epidemiology. 1997;8: 495-500. 18. Frishman WH. Treatment of myocardial ischemia and myocardial infarction in the elderly. South Med J. 1993;86:2S-29 2S-37. 19. Guralnik JM, Fried LP, Simonsick EM, Bandeen-Roche KJ, Kasper JD. Screening the community-dwelling population for disability. In: Guralnik JM, Fried LP, Simonsick EM, Kasper JD, Lafferty ME, eds. The Women s Health and Aging Study: Health and Social Characteristics of Older Women With Disability. Bethesda, Md: National Institute on Aging; 1995:9-18. NIH publication 95-4009. 20. Fried LP, Kasper JD, Williamson JD, et al. Appendix E: disease ascertainment algorithms. In: Guralnik JM, Fried LP, Simonsick EM, Kasper JD, Lafferty ME, eds. The Women s Health and Aging Study: Health and Social Characteristics of Older Women With Disability. Bethesda, Md: National Institute on Aging; 1995: E-1 E-23. NIH publication 95-4009. 21. Criqui MH, Denenberg JO, Bird CE, Fronek A, Klauber MR, Langer RD. The correlation between symptoms and non-invasive test results in patients referred for peripheral arterial disease testing. Vascular Med. 1996;1:65-71. 22. Weitz JI, Byrne J, Clagett P, et al. Diagnosis and treatment of chronic arterial insufficiency of the lower extremities: a critical review. Circulation. 1996;11: 3026-3029. 23. GorollAH,May LA,Mulley AG.Primary Care Medicine:Office Evaluation and Management of the Adult Patient. 3rd ed. Philadelphia, Pa: JP Lippincott Co; 1995:108-109. 24. Brand FN, Abbott RD, Kannel WB. Diabetes, intermittent claudication, and risk of cardiovascular events. Diabetes. 1989;38:504-509. 25. Ernst E, Fialka V. A review of the clinical effectiveness of exercise therapy for intermittent claudication. Arch Intern Med. 1993;153:2357-2360. 26. Meade TW, Gardner MJ, Cannon P, et al. Observer variability in reading the peripheral pulse. Br Heart J. 1968;30:661-665. 27. Criqui MH, Langer RD, Fronek A, et al. Mortality over a period of 10 years in patients with peripheral arterial disease. N Engl J Med. 1992;326:381-386. 392