Peripheral Artery Disease, Diabetes, and Reduced Lower Extremity Functioning

Transcription

1 Pathophysiology/Complications O R I G I N A L A R T I C L E Peripheral Artery Disease, Diabetes, and Reduced Lower Extremity Functioning NANCY C. DOLAN, MD 1 KIANG LIU, PHD 2 MICHAEL H. CRIQUI, MD, MPH 3 PHILIP GREENLAND, MD 2 JACK M. GURALNIK, MD, PHD 4 CHEELING CHAN, MS 2 JOSEPH R. SCHNEIDER, MD, PHD 5,6 AIMEE LUNA MANDAPAT, BA 1 GARY MARTIN, MD 1 MARY M. MCDERMOTT, MD 1,2 OBJECTIVE To characterize lower extremity function and dysfunction in peripheral artery disease (PAD) patients with and without diabetes. RESEARCH DESIGN AND METHODS In this cross-sectional study, 460 men and women with PAD (147 with diabetes) were recruited from three academic medical centers. Assessments included ankle brachial index (ABI), neuropathy score, 6-min walk distance, 4-m walking velocity, Walking Impairment Questionnaire (0 100 scale, 100 best), and summary performance score (SPS) (0 12 scale, 12 best). RESULTS The mean ABI was similar in PAD patients with and without diabetes. PAD patients with diabetes were younger, had a higher BMI, had a worse neuropathy score, and had a greater number of cardiovascular comorbidities compared with those without diabetes. Participants with diabetes were less likely to report classical symptoms of intermittent claudication and more likely to report exertional leg pain, which sometimes started at rest. After adjusting for age, those with diabetes had a shorter mean 6-min walk distance (1,040 vs. 1,168 feet, P 0.001), slower fast-pace 4-m walk velocity (0.83 vs m/sec, P 0.001), and a lower SPS (7.3 vs. 8.6, P 0.001) than those without diabetes. Patients with diet-controlled diabetes performed better than those on diabetes medications. Differences in lower extremity functioning between patients with and without diabetes were largely attenuated but not abolished for SPS and fast-pace 4-m walk velocity after adjustment for type of exertional leg pain, neuropathy score, and number of cardiovascular comorbidities. CONCLUSIONS Subjects with PAD and diabetes have poorer lower extremity function than those with PAD alone. This difference in functioning appears to be largely explained by diabetes-associated neuropathy, differences in exertional leg symptoms, and greater cardiovascular disease in patients with diabetes. Diabetes is common in men and women with peripheral artery disease (PAD). In the Framingham study (1), 20% of men and women with Diabetes Care 25: , 2002 From the 1 Department of Medicine, Northwestern University Medical School, Chicago, Illinois; the 2 Department of Preventive Medicine, Northwestern University Medical School, Chicago, Illinois; the 3 Department of Family and Preventive Medicine, University of California at San Diego, San Diego, California; the 4 Epidemiology, Demography, and Biometry Program, National Institute on Aging, Bethesda, Maryland; the 5 Division of Vascular Surgery, Department of Surgery, Northwestern University Medical School, Chicago, Illinois; and the 6 Division of Vascular Surgery, Department of Surgery, Evanston/Northwestern Hospital, Evanston, Illinois. Address correspondence and reprint requests to Dr. Nancy C. Dolan, Northwestern University Medical School, 675 N. St. Clair, Suite , Chicago, IL ndo428@northwestern.edu. Received for publication 14 May 2001 and accepted in revised form 21 September Abbreviations: ABI, ankle brachial index; CHS, Cardiovascular Health Study; GDS, Geriatric Depression Scale; PAD, peripheral artery disease; PCP, primary care physician; SPS, summary performance score; WHAS, Women s Health and Aging Study; WIQ, Walking Impairment Questionnaire. A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances. intermittent claudication had diabetes compared with 6% of those without intermittent claudication. Diabetes and PAD are both frequently complicated by neuropathy and foot ulceration, and each condition is associated with an increased risk of gangrene and lower extremity amputation (2 4). Among individuals in the general population, diabetes has also been associated with increased disability, poorer physical functioning, and an increased risk of being unable to do mobility-related tasks (3,5,6). The excess risk of disability among people with diabetes may be related to the complications and comorbidities commonly associated with diabetes (5). Whether the excess risk of disability associated with diabetes is exacerbated in subjects with PAD has not been previously studied. Impaired lower extremity functioning is an important predictor of future disability, mobility loss, and nursing home placement (7,8). Maintaining function is an important public health objective because it allows men and women with chronic diseases such as diabetes and PAD to live longer. Establishing whether individuals with PAD and diabetes have greater impairment in lower extremity functioning than individuals with PAD without diabetes and, if so, why they have greater impairment could have important clinical implications for the prevention of disability in patients with PAD. The purpose of this study was to compare lower extremity functioning between PAD patients with and without diabetes. We hypothesized that PAD patients with diabetes, regardless of PAD severity, as measured by the ankle brachial index (ABI), have poorer lower extremity functioning than those without diabetes. RESEARCH DESIGN AND METHODS Participant identification We defined PAD as an ABI The study findings, therefore, are generalizable only to those PAD patients with compressible ankle vessels. Participants were identified from consecutive men and women aged 55 years with lower extremity arterial studies consistent with PAD at one of three Chicago-area medical DIABETES CARE, VOLUME 25, NUMBER 1, JANUARY

2 PAD, diabetes, and reduced lower extremity functioning centers noninvasive blood flow laboratories between 1 January 1997 and 31 October Fewer participants were identified from general medicine patients aged 55 years with ABIs 0.90 who were seen in a large practice affiliated with Northwestern University Medical School. Identified individuals were invited to return to the medical center for a study visit. The visit consisted of a comprehensive interview, lower extremity functional measurements, and a physical examination that included ABI, BMI, and monofilament testing for neuropathy. The Institutional Review Boards of Northwestern University Medical School and Catholic Health Partners approved the study protocol, and all participants gave informed consent. A total of 150 participants were excluded because their ABI was 0.90 at the study visit. Patients who resided in nursing homes, were wheelchairdependent, had leg or foot amputations, did not speak English, or had Mini- Mental Status scores of 18 were also excluded from the study. ABI measurement Based on a previous study, ABI was calculated by averaging the dorsalis pedis and posterior tibial artery pressures in each limb and dividing this pressure by the average of the four brachial artery pressures (9). The average of the brachial artery pressures in the arm with the highest pressure was used when one brachial arterial pressure was greater than the other in both measurement sets and when the right and left brachial artery pressures differed by 10 mmhg. Leg symptom groups Leg symptoms were ascertained using the San Diego claudication questionnaire (10). Participants were categorized into five mutually exclusive leg pain groups: 1) no exertional leg pain; 2) atypical exertional leg pain/carry-on, defined as exertional leg symptoms that do not begin at rest and do not cause the participant to stop walking; 3) atypical exertional leg pain/stop, defined as exertional leg symptoms that do not begin at rest, are not consistent with intermittent claudication, and cause the participant to stop walking; 4) intermittent claudication, defined as exertional calf pain that does not begin at rest, causes the participant to stop walking, and resolves within 10 min of rest; and 5) leg pain with exertion and rest, defined as exertional leg pain that sometimes begins at rest. Identifying diabetes and diabetes severity We documented the presence of diabetes using a disease-specific algorithm from the Women s Health and Aging Study (WHAS) and the Cardiovascular Health Study (CHS) (11). The algorithm combines data from patient report, the medical record, medications, and a primary care physician (PCP) questionnaire (11). Diabetes was considered present if at least two of the following criteria were satisfied: 1) the patient reported physiciandiagnosed diabetes; 2) the patient was taking diabetes medications; 3) there was a medical record report of a GHb 10%; and 4) the PCP questionnaire reported a diagnosis of diabetes. If the first three criteria were not satisfied but the PCP questionnaire indicated diabetes, the physician was recontacted for confirmation. If the diagnosis was confirmed, the patient was considered to have diabetes. The algorithm does not distinguish between type 1 and type 2 diabetes. To estimate diabetes severity, participants with diabetes were classified according to their use of diabetes medication: 1) no diabetes medication; 2) oral medications only; or 3) insulin with or without oral medication. Other medical conditions Comorbidities other than diabetes that were previously shown or expected to influence lower extremity functioning were identified using methods from the WHAS and the CHS (11). These included angina, myocardial infarction, stroke, heart failure, pulmonary disease, knee and hip arthritis, spinal stenosis, disc disease, Parkinson s disease, and hip fracture (12,13). The presence of high cholesterol was determined using medical records and PCP questionnaire, whereas hypertension was determined by self-report and PCP questionnaire. Physician-diagnosed vision problems were established by self-report. We used the Geriatric Depression Scale (GDS) short form to ascertain the presence of depressive symptoms (range 0 15, higher more depression) (14). Depression was defined as the presence of six or more depressive symptoms on the GDS (14). A monofilament was used to assess sensation on the dorsal and ventral surfaces of each foot in specific locations as previously described (15,16). The ordinal neuropathy score that was developed represented the number of items missed on the monofilament test (range 0 22, higher worse). Lower extremity functional measures Measure of walking endurance: the 6-min walk. The 6-min walk distance measures endurance and correlates significantly with time to onset of leg pain during treadmill testing, maximum distance achieved in treadmill testing, physical activity levels, and ABI (9,17,18). Using a standardized protocol, participants were instructed to walk up and down a 100- foot distance for 6 min, covering as much distance as possible (17,18). Measures of walking speed: 4-m walking velocities. The usual and fast-pace 4-m walking velocities measure the time it takes participants to walk a 4-m distance at both their usual and fastest pace, respectively. Both of these walks were performed twice, and the fastest walk in each set (usual and fastest pace) was used in analyses (8). Measures of balance and strength: tandem stand and repeated chair rises. The full tandem stand measures whether or not participants can stand with both feet together side-by-side for 10 s (7,8). The repeated chair rises test measures the time it takes to complete five chair rises and is a measure of leg strength as well as balance (7,8). Measures of physical activity Physical activity levels were measured over 7 days using the Caltrac accelerometer (19 21). Based on previous study, we programmed accelerometers using identical weight, height, age, and sex for each participant so that activity could be compared meaningfully among all study participants (19 21). Programmed in this manner, the accelerometer measures activity units. Because we had a relatively small number of accelerometers, they were distributed to participants when available. Forty-six percent of the study participants wore the accelerometer. There were no significant differences in age, sex, BMI, or diabetes prevalence between those who wore the accelerometer and those who did not. Self-reported physical activity was measured by asking all participants to estimate the number of 114 DIABETES CARE, VOLUME 25, NUMBER 1, JANUARY 2002

3 Dolan and Associates Table 1 Age-adjusted* characteristics of men and women aged >55 years with PAD according to the presence or absence of diabetes Characteristic Diabetes No diabetes P n African-American (%) Male (%) ABI Leg symptom category No exertional leg pain (%) Exertional pain/carry-on (%) Exertional pain/stop (%) Intermittent claudication (%) Leg pain with exertion and rest (%) BMI (kg/m 2 ) Neuropathy score (0 22 scale, higher worse) GDS (0 15 scale, higher worse) Lower extremity revascularization (%) Ever smoked (%) Current smoker (%) Cigarette smoking (pack years) Physician-diagnosed vision problem (%) Total number of comorbidities Any cardiovascular disease (%) Any lower extremity arthiritis (%) Pulmonary disease (%) Cancer (%) *Age was (range 55 91) for those with diabetes and (range 55 93) for those without diabetes. P was based on logistic regression for dichotomous variables and analysis of covariance for continuous variables. Leg symptom categories were mutually exclusive. The P values for leg symptoms were for comparing the rate of each leg symptom between PAD participants with and without diabetes. Total number of comorbidities other than diabetes included the following: myocardial infarction, heart failure, angina, stroke, knee arthritis, hip arthritis, hip fracture, spinal stenosis, disc disease, pulmonary disease, and cancer. Cardiovascular disease included myocardial infarction, heart failure, angina, and stroke. Lower extremity arthritis included knee arthritis, hip arthritis, hip fracture, spinal stenosis, and disc disease. stairs climbed and city blocks that they had walked in the past week. Walking Impairment Questionnaire The Walking Impairment Questionnaire (WIQ) measures self-reported walking speed and distance among PAD patients (22,23). To measure walking distance, participants ranked their ability to walk specific distances on a 0 4 Likert scale. The WIQ distance score was calculated by multiplying the Likert scale score with the corresponding distance, summing these products, and then dividing by the maximum score possible to get a percent score that ranged from 0 to 100 (100 best). Walking speed was assessed by asking subjects to rank their degree of difficulty walking a block slowly, at average speed, quickly, or running/jogging on a 0 4 Likert scale (4 best). Participants responses on the Likert scale were multiplied by the approximate miles per hour represented by each walking speed. The resultant score was divided by the maximum possible score to achieve a percent score that ranged from 0 to 100 (100 best). Composite measure of lower extremity functioning: summary performance score The summary performance score (SPS) measures leg strength, balance, and usual walking speed (8). To calculate the SPS, a 0 4 score was assigned for 4-m walking velocity, chair rises, and standing balance, respectively, based on cut points derived from normative data of representative community populations. These scores were summed to obtain the SPS, which ranged from 0 to 12 (12 best) (9,24). Statistical analyses Differences in clinical characteristics and lower extremity functional measures between PAD participants with and without diabetes were assessed using analysis of covariance while adjusting for age. Statistical significance was based on logistic regression for dichotomous variables and analysis of variance for continuous variables. Using a forward-selection method, multiple linear models were developed to test for potential predictors that accounted for differences between participants with and without diabetes in the following lower extremity functional measures: 6-min walk distance, fastest pace 4-m walking velocity, WIQ speed performance score, and SPS. These four measures were chosen based on their strong correlation with ABI (19,25). First, potential confounders, including age, sex, race (African-American or non African- American), and BMI, were simultaneously forced into the first model on the basis of previous studies. Second, other potential predictors, including four dummy variables for leg symptoms (exertional leg pain/carry-on, exertional leg pain/stop, intermittent claudication, and leg pain with exertion and rest, with no exertional leg pain used as the reference category), neuropathy score, number of cardiovascular diseases (including myocardial infarction, heart failure, angina, and stroke), number of arthritis diseases (including knee and hip arthritis, hip fracture, spinal stenosis, and disc disease), number of other diseases (pulmonary disease and cancer), depression score, and ABI were entered one at a time into the linear model. From these analyses, we selected as the next covariate for the second model the variable that explained the greatest difference (i.e., reduced most the difference between groups) in performance between participants with and without diabetes, and that still remained statistically significant (P 0.10) after adjusting for other covariates already in the model. Once a variable had been selected, it stayed in the model. This selection process continued until no variable considered for addition improved the model. We also assessed the differences in lower extremity functioning among PAD patients with different diabetes severity using analysis of covariance while adjusting for age, sex, race, and BMI. All analyses were conducted using SAS Statistical Software (SAS Institute Inc, Cary, NC). DIABETES CARE, VOLUME 25, NUMBER 1, JANUARY

4 PAD, diabetes, and reduced lower extremity functioning Table 2 Age-adjusted lower extremity functioning among men and women aged >55 years with PAD according to the presence or absence of diabetes Lower extremity functional measurements Diabetes No diabetes P* n Measures of walking endurance 6-Min walk distance (feet) 1, , Stopped during 6-min walk (%) Measures of walking speed 4-M walking velocity (m/sec) Fastest-pace 4-m walking velocity (m/sec) Measures of balance and strength Completed full tandem stand (%) Time to complete five chair rises (sec) Measures of physical activity Number of stair flights climbed in previous week Number of blocks walked in previous week Day accelerometer score (activity units) WIQ WIQ distance score (0 100 scale, 100 best) WIQ speed score (0 100 scale, 100 best) SPS (0 12 scale, 12 best) *P was based on logistic regression for dichotomous variables and analysis of covariance for continuous variables. Completed full tandem stand denotes percent that held full tandem stand for 10 sec. Measures of physical activity: number of stair flights climbed and number of blocks walked in previous week, based on self-report. Physical activity was also measured objectively over 7 days with the Caltrac accelerometer. Accelerometer data were available for 75 participants with diabetes and 150 participants without diabetes. The SPS combined performance on walking velocity, repeated chair rises, and standing balance tests (0 12 scale, 12 best). RESULTS Study participant characteristics PAD patients with diabetes were younger than those without diabetes and had a higher mean BMI, a worse neuropathy score, a higher prevalence of hypertension, and a greater number of comorbidities (predominantly cardiovascular) (Table 1). Although the mean ABI was similar between the two groups, exertional leg pain symptoms were different. PAD patients with diabetes were less likely to report intermittent claudication than those without diabetes and more likely to report leg pain on exertion and rest. Lower extremity functioning: participants with diabetes versus participants without diabetes Table 2 shows the age-adjusted measures of lower extremity function among PAD participants with and without diabetes. Those with diabetes consistently performed worse than those without diabetes. The differences between participants with and without diabetes were most marked in measures of balance, walking endurance, and walking speed. Specifically, participants with diabetes were less likely than participants without diabetes to complete the full tandem stand, took a longer time to complete five repeated chair rises, had significantly slower walking velocities, and walked significantly fewer feet during the 6-min walk. This latter difference appeared to be due to a discrepancy in walking speed because the stop rate during the 6-min walk was similar in the two groups. In contrast, there were no significant differences between participants with and without diabetes in the accelerometer score, the number of blocks walked, or the number of flights of stairs climbed in the past week. All three of these latter measures assessed physical activity. Diabetes severity Table 3 shows the adjusted association between lower extremity functioning and diabetes severity. For all four measures analyzed, after adjusting for age, sex, race, and BMI, participants without diabetes performed the best, and participants with diabetes who were on medication performed the worst. The relation between insulin versus oral diabetes medication and lower extremity functioning was inconsistent. For the WIQ speed score, patients with diabetes on insulin performed worse than those on oral medications, although this difference was not significant. For the 6-min walk distance, the SPS, and the fast-pace 4-m walking velocity, the reverse was true. Differences between diabetes severity groups remained significant for the SPS and the fastest-pace 4-m walking velocity, even after further adjusting for leg symptoms and comorbidities. Predictors of differences in the lower extremity functional measures between PAD patients with and without diabetes Table 4 shows results for the stepwise linear regression models, in which the effects of individual potential confounders are shown using stepwise methods and expressed as adjusted differences in functional measurements (6-min walk distance, SPS, fastest-pace 4-m walking velocity, and WIQ speed score) between PAD participants with and without diabetes. Differences in leg symptoms and neuropathy scores appeared to account for all of the significant differences in the 6-min walk distances between the participants with and without diabetes and much of the differences in the SPS and fastest-pace 4-m walking velocities between participants with and without diabetes. Cardiovascular disease was also a large contributor to the differences in fastestpace 4-m walk velocity. The neuropathy score largely explained the differences in WIQ speed scores between participants with and without diabetes. For both the SPS and the fastest pace 4-m walking velocity, there remained a statistically significant residual difference between participants with and without diabetes, even after adjusting for potential confounders. CONCLUSIONS In this cohort of men and women aged 55 years with PAD, participants with diabetes had poorer lower extremity functioning than those without diabetes. Differences between PAD participants with and without diabetes were greatest for the measures of balance, walking endurance, and walking speed. Poorer neuropathy scores, higher prevalence of leg pain on exertion and rest, and a greater number of cardiovascular diseases among participants with di- 116 DIABETES CARE, VOLUME 25, NUMBER 1, JANUARY 2002

5 Dolan and Associates Table 3 Adjusted association between lower leg functioning and diabetes severity among men and women aged >55 years with PAD Diabetes Model No diabetes No medications Oral medication only Insulin oral medication P n Adjusted for age, sex, race, and BMI 6-min walk distance (feet) 1, , , , SPS (0 12 scale, 12 best) Fastest-pace 4-m walking velocity (m/sec) WIQ speed score (0 100 scale, best) Adjusted for age, sex, race, BMI, and leg symptoms 6-min walk distance (feet) 1, , , , SPS (0 12 scale, 12 best) Fastest-pace 4-m walking velocity (m/sec) WIQ speed score (0 100 scale, best) Adjusted for age, sex, race, BMI, leg symptoms, and comorbidities 6-min walk distance (feet) 1, , , , SPS (0 12 scale, 12 best) Fastest-pace 4-m walking velocity (m/sec) WIQ speed score (0 100 scale, 100 best) *P derived using analysis of covariance comparing all four groups. Race was categorized as African-American or non African-American. Leg symptoms were categorized as four dummy variables (exertional leg pain/carry-on, exertional leg pain/stop, intermittent claudication, and leg pain with exertion and rest, with no exertional leg pain as reference group). Comorbidities were defined as number of cardiovascular diseases (including myocardial infarction, heart failure, angina, and stroke) and number of arthritis diseases (including knee arthritis, hip arthritis, hip fracture, spinal stenosis, and disc disease). The SPS combined performance on walking velocity, repeated chair rises, and standing balance tests (0 12 scale, 12 best). abetes accounted for much but not all of the differences observed. Although most of the differences in lower extremity functioning were associated with diabetes-related factors (neuropathy and leg pain symptoms) and cardiovascular disease, even after adjusting for these and other confounders, we were not able to account for all of the differences observed between PAD participants with and without diabetes in regard to fastest-pace 4-m walking velocity and SPS. There are several potential explanations for our results. First, our neuropathy score is only a crude measure of sensory neuropathy and does not assess propioception or autonomic dysfunction, both of which may be adversely affected by diabetes and could negatively impact on lower extremity functioning. A more thorough and comprehensive investigation of diabetic neuropathy would include measurement of sensory and motor nerve conduction velocity, vibration perception threshold, warmth and cold perception threshold, and autonomic testing. Second, it is possible that medial arterial calcinosis, which is more common in people with diabetes (26,27), may have led to falsely elevated ABIs in the participants with diabetes, thereby underestimating the severity of PAD in the group with diabetes. Such an underestimation could explain the residual differences in the fastest-pace 4-m walking velocity and SPS after adjusting for ABI. In contrast to the differences observed in measures of balance, leg strength, walking endurance, and walking speed, both self-reported and accelerometer-measured physical activities were similar between participants with and without diabetes. The reasons for this are unclear. It is possible that diabetes-related neuropathy, which accounts for much of the difference between participants with and without diabetes in the other measures, does not have a significant adverse effect on physical activity. Alternatively, it is possible that differences in physical activity levels between PAD patients with and without diabetes do exist, but that the physical activity measures used were not sensitive enough to detect these differences. Among patients with coronary artery disease, patients with diabetes are less likely than patients without diabetes to have typical angina (28). Our findings suggest that there is a similar phenomenon in PAD patients with diabetes. Compared with PAD patients without diabetes, PAD patients with diabetes were less likely to have classical intermittent claudication but more likely to have leg pain with exertion and rest. The presence of diabetes-related neuropathy may explain the majority of the differences in leg pain symptoms between patients with and without diabetes. Altered foot architecture, common in patients with diabetes but not measured in this study, may also have contributed to the higher prevalence of leg pain on exertion and rest observed in PAD patients with diabetes. DIABETES CARE, VOLUME 25, NUMBER 1, JANUARY

6 PAD, diabetes, and reduced lower extremity functioning Table 4 Stepwise multiple linear analyses for differences in lower extremity functional measures among men and women aged >55 years with PAD according to the presence or absence of diabetes* Adjusted mean measures Models Diabetes (n 147) No diabetes (n 313) Differences P 6-min walk distance (feet) Adjusted for age, sex, race, and BMI 1, , Adjusted for age, sex, race, BMI, and leg symptoms 1, , Adjusted for age, sex, race, BMI, leg symptoms, and 1, , neuropathy score Adjusted for age, sex, race, BMI, leg symptoms, neuropathy 1, , score, and number of cardiovascular diseases Adjusted for age, sex, race, BMI, leg symptoms, neuropathy 1, , score, number of cardiovascular diseases, and depression Adjusted for age, sex, race, BMI, leg symptoms, neuropathy 1, , score, number of cardiovascular diseases, depression, and ABI SPS (0 12 scale, 12 best) Adjusted for age, sex, race, and BMI Adjusted for age, sex, race, BMI, and neuropathy score Adjusted for age, sex, race, BMI, neuropathy score, and leg symptoms Adjusted for age, sex, race, BMI, neuropathy score, leg symptoms, and number of cardiovascular diseases Adjusted for age, sex, race, BMI, neuropathy score, leg symptoms, number of cardiovascular diseases, and number of arthritis diseases Adjusted for age, sex, race, BMI, neuropathy score, leg symptoms, number of cardiovascular diseases, number of arthritis diseases, and ABI Fastest-pace 4-m walking velocity (m/sec) Adjusted for age, sex, race, and BMI Adjusted for age, sex, race, BMI, and leg symptoms Adjusted for age, sex, race, BMI, leg symptoms, and neuropathy score Adjusted for age, sex, race, BMI, leg symptoms, neuropathy score, and number of cardiovascular diseases Adjusted for age, sex, race, BMI, leg symptoms, neuropathy score, number of cardiovascular diseases, and ABI WIQ speed score (0 100 scale, 100 best) Adjusted for age, sex, race, and BMI Adjusted for age, sex, race, BMI, and neuropathy score Adjusted for age, sex, race, BMI, neuropathy score, and number of cardiovascular diseases Adjusted for age, sex, race, BMI, neuropathy core, number of cardiovascular diseases, and depression Adjusted for age, sex, race, BMI, neuropathy score, number of cardiovascular diseases, depression, and number of arthritis diseases *At each step (except in the first model), the variable that explained the largest difference between participants with diabetes and without diabetes was selected. Later models contain all the variables included in earlier models. Race was categorized as African-American or non African-American. Leg symptoms were categorized as four dummy variables (exertional leg pain/carry-on, exertional leg pain/stop, intermittent claudication, and leg pain with exertion and rest, with no exertional leg pain as reference group). Neuropathy score denotes the number of items missed on the monofilament test (0 22 scale, higher worse). Cardiovascular diseases included myocardial infarction, heart failure, angina, and stroke. Arthritis diseases included knee arthritis, hip arthritis, hip fracture, spinal stenosis, and disc disease. Depression was defined as a GDS score of 6(0 15 scale, higher score greater depression). SPS combined performance on walking velocity, repeated chair rises, and standing balance tests (0 12 scale, 12 best). 118 DIABETES CARE, VOLUME 25, NUMBER 1, JANUARY 2002

7 Dolan and Associates There are several limitations to this study. First, we excluded those PAD patients from our noninvasive vascular laboratory with ABIs 0.90, a finding suggestive of calcified, noncompressible vessels that is more common in patients with diabetes than patients without diabetes (26,27). Our findings, therefore, cannot be generalized to PAD patients with diabetes who have noncompressible vessels. Second, because medial calcinosis in patients with diabetes is associated with neuropathy (29,30) to the extent that neuropathy affects lower extremity function, exclusion of PAD patients with diabetes and noncompressible vessels may have underestimated the differences in lower extremity function between PAD patients with and without diabetes. Third, despite our rigorous method of diabetes ascertainment, it is possible that our study population of PAD patients without diabetes included some subjects with undiagnosed diabetes, which might have reduced the observed differences between participants with and without diabetes. However, we believe the number of undiagnosed cases of diabetes was very small because once a patient is diagnosed with PAD, he/she is generally monitored regularly for diabetes. Fourth, we did not have information on duration of diabetes or glucose control, which may have been better measures of diabetes severity than the methods used. Although we did not have information on the type of diabetes, given the mean age of the patients with diabetes (69 years) and the small proportion of patients using insulin in the diabetic group, we estimate that the majority had type 2 diabetes. Previous studies have shown high rates of foot ulcers and lower extremity amputations in PAD patients with diabetes as well as faster rates of PAD progression in patients with diabetes compared with patients without diabetes (3,4,31). Thus, diabetes is known to be associated with more severe forms of clinically apparent PAD. This report extends the spectrum of diabetes-related complications to include lower extremity dysfunction in PAD patients. The excess disability associated with diabetes in PAD patients is consistent with what has been found in the general population (3,5,6). Our results also give further support for aggressive lifestyle modification for diabetes prevention in those with PAD at risk for developing diabetes. Patients with PAD without established diabetes should be aggressively monitored for the development of diabetes; if they do have diabetes, they should be evaluated for the presence of associated comorbidities that could contribute to disability. Based on our data, potential interventions to improve lower extremity functioning in PAD patients with diabetes might include the development of effective treatments for leg pain on exertion and rest, prevention and treatment of neuropathy, and prevention of cardiovascular disease. Given the association between poorer lower extremity function and future disability, mobility loss, and nursing home placement (7,8), these findings are likely to have important prognostic implications for PAD patients with diabetes. Acknowledgments Supported by grant no. R from the National Heart Lung and Blood Institute and by grant no. RR from the National Center for Research Resources, National Institutes of Health. M.M.M. was the recipient of an Established Investigator Award from the American Heart Association and was a Robert Wood Johnson Generalist Physician Faculty Scholar. References 1. Marubito JM, D Agostino RB, Silbershatz H, Wilson WF. Intermittent claudication: a risk profile from the Framingham Heart Study. Circulation 96:44 49, Nathan DM: Long-term complications of diabetes mellitus. N Engl J Med 328: , National Diabetes Data Group (Eds.): Diabetes in America. 2nd ed. Bethesda, MD, National Institutes of Health (publ. no. NIH ) 4. Graves EJ: National Center for Health Statistics: National Hospital Discharge Survey: Annual Summary, Hyattsville, MD, National Center for Health Statistics, 1992 (Vital and Health Statistics Ser. 13, no. 112) 5. Gregg EW, Beckles GL, Williamson DF, Leveille SG, Langlois JA, Engelgau MM, Narayan KM: Diabetes and physical disability among older U.S. adults. Diabetes Care 23: , Guccione AA, Felson DT, Andersen JJ, Anthony JM, Zhang Y, Wilson PW, Kelly- Hayes M, Wolf PA, Kreger BE, Kannel WB: The effects of specific medical conditions on the functional limitations of elders in the Framingham Study. Am J Public Health 84: , Guralnik JM, Ferrucci L, Simonsick E, Salive ME, Wallace RB: Lower extremity function in persons over 70 years as a predictor of subsequent disability. N Engl J Med 332: , Guralnik JM, Simonsick EM, Ferrucci L, Glynn RJ, Berkman LF, Blazer DG, Scherr PA, Wallace RB: A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol 49: M85 M94, McDermott MM, Criqui MH, Liu K, Guralnik JM, Greenland P, Martin GJ, Pearce W: Lower ankle/brachial index, as calculated by averaging the dorsalis pedis and posterior tibial arterial pressures, and association with leg functioning in peripheral arterial disease. J Vasc Surg 32: , Criqui MH, Denenberg JO, Bird CE, Fronek A, Klauber MR, Langer RD: The correlation between symptoms and noninvasive test results in patients referred for peripheral arterial disease testing. Vasc Med 1:65 71, 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 (NIH publ. no ) 12. Ettinger WH, Fried LP, Harris T, Shemanski L, Schulz R, Robbins J: Self-reported causes of physical disability in older people: the Cardiovascular Health Study. J Am Geriatr Soc 42: , Fried LP, Bandeen-Roche K, Kasper JD, Guralnik JM: Association of comorbidity with disability in older women: the Women s Health and Aging Study. J Clin Epidemiol 52:27 37, Lyness JM, Noel TK, Cox C, King DA, Conwell Y, Caine ED: Screening for depression in elderly primary care patients. Arch Intern Med 157: , Birke JA, Sims DS: Plantar sensory threshold in the ulcerative foot. Lepr Rev 57: , Olmos PR, Cataland S, O Dorision TM, Casey CA, Smead WL, Simon SR: The Semmes-Weinstein monofilament as a potential predictor of foot ulceration in patients with noninsulin-dependent diabetes. Am J Med Sci 309:76 82, Montgomery PS, Gardner AW: The clinical utility of a six-minute walk test in peripheral arterial occlusive disease patients. J Am Geriatr Soc 46: , Gardner AW, Womack CJ, Sieminski DJ, Montgomery PS, Killewich LA, Fonong T: Relationship between free-living daily physical activity and ambulatory measures in older claudicants. Angiology 49: , 1998 DIABETES CARE, VOLUME 25, NUMBER 1, JANUARY

8 PAD, diabetes, and reduced lower extremity functioning 19. Richardson MT, Leon AS, Jacobs DR, Ainsworth BE, Serfass R: Ability of the Caltrac accelerometer to assess daily physical activity levels. J Cardiopulm Rehabil 15: , McDermott MM, Liu K, O Brien E, Guralnik JM, Criqui MH, Martin GJ, Greenland P: Measuring physical activity in peripheral arterial disease: a comparison of two physical activity questionnaires with an accelerometer. Angiology 51:91 100, Hiatt WR, Hirsh AT, Regensteiner JG, Brass EP: Clinical trials for claudication: assessment of exercise performance, functional status, and clinical endpoints. Circulation 3: , Regensteiner JG, Steiner JF, Panzer RJ, Hiatt WR: Evaluation of walking impairment by questionnaire in patients with peripheral artery disease. J Vasc Med and Biol 2: , Guralnik JM, Ferrucci L, Pieper CF, Leveille SG, Markides KS, Ostir GV, Studenski S, Berkman LF, Wallace RB: Lower extremity function and subsequent disability: consistency across studies, predictive models, and value of gait speed alone compared with short physical performance battery. J Gerontol A Biol Sci Med Sci 55:M221 M231, McDermott MM, Liu K, Guralnik JM, Mehta S, Criqui MH, Martin GJ, Greenland P: The ankle brachial index independently predicts walking velocity and walking endurance in peripheral artery disease. Journ Am Geriatr Soc 46: , McDermott MM, Mehta S, Liu K, Guralnik JM, Martin GJ, Criqui MH, Greenland P: Leg symptoms and the ankle brachial index independently predict patient reported walking ability in peripheral arterial disease in patients without limb threatening ischemia. J Gen Intern Med 14: , Akbari CM, LoGerfo FW: Diabetes and peripheral artery disease. J of Vasc Surg 30: , Quigley FG, Faris IB, Duncan HJ: A comparison of Doppler ankle pressures and skin perfusion pressure in subjects with and without diabetes. Clin Physiol 11:21 25, Langer A, Freeman MR, Josse RG, Steiner G, Armstrong PW: Detection of silent myocardial ischemia in diabetes mellitus. Am J Cardiology 67: , Psyrogiannis A, Kyriazopoulou V, Vagenakis AG: Medial arterial calcification is frequently found in patients with microalbuminuria. Angiology 50: , Edmonds ME, Morrison N, Laws JW, Watkins PJ: Medial arterial calcification and diabetic neuropathy. Br Med J (Clin Res Ed) 284: , Bird CE, Criqui MH, Fronek A, Denenberg JO, Klauber MR, Langer RD: Quantitative and qualitative progression of peripheral arterial disease by non-invasive testing. Vasc Med 4:15 21, DIABETES CARE, VOLUME 25, NUMBER 1, JANUARY 2002