Efficacy of arm-ergometry versus treadmill exercise training to improve walking distance in patients with claudication

Transcription

1 Efficacy of arm-ergometry versus treadmill exercise training to improve walking distance in patients with claudication Diane Treat-Jacobson University of Minnesota School of Nursing; Center for Gerontological Nursing, Ulf G Bronas University of Minnesota School of Nursing and Arthur S Leon University of Minnesota School of Kinesiology, Laboratory of Physiological Hygiene and Exercise Science Abstract: The efficacy of treadmill walking training to improve pain-free (PFWD) and maximal (MWD) walking distance in patients with claudication is well documented. The effects of aerobic arm-ergometry to improve PFWD and MWD compared to treadmill walking or usual care are not known. Forty-one participants (29 male, 12 female, mean age 67.7 years, 92.7% smoking history, 36.6% with diabetes) with lifestyle-limiting claudication were randomized to 12 weeks of 3 hours/week of supervised exercise training using either arm-ergometry, treadmill walking, or a combination, versus. PFWD and MWD were assessed before and after training, and after 12 weeks of follow-up. The 12-week MWD increased significantly in the armergometry (+53%), treadmill (+69%), and combination (+68%) groups (p < versus ). The 24-week MWD was maintained in the arm-ergometry (p = 0.009) and treadmill (p = 0.019) groups, whereas the combination group declined (p = 0.751) versus. The 12-week PFWD increased significantly in the arm-ergometry group (+82%; p = versus ). PFWD in treadmill (+54%; p = versus ) and combination (+60%; p = versus ) groups did not reach statistical significance. PFWD improvement was maintained in the arm-ergometry group after a 12-week follow-up (+123%; p = versus ). In conclusion, these pilot data demonstrate for the first time that dynamic arm exercise training can improve walking capability in people with peripheral arterial disease (PAD)-induced claudication compared to participants receiving usual care and that improvement was not different from that seen with treadmill walking exercise training. Dynamic arm exercise may be a therapeutic exercise option for patients with PAD. Keywords: claudication, exercise therapy, intermittent claudication, peripheral arterial disease Introduction Peripheral arterial disease (PAD) is a progressive atherosclerotic occlusive disease affecting approximately nine million Americans. 1,2 It is associated with increased mortality and significant morbidity, including major limitations in mobility and physical functioning, and decreased quality of life. 3 6 Insufficient arterial blood flow to the lower extremities results in skeletal muscle ischemia-induced debilitating pain associated with walking (claudication). Approximately four million Americans with PAD have moderate to severe impairment in walking ability due to claudication. This contributes to a significant reduction in aerobic capacity, resulting in Corresponding author: Diane Treat-Jacobson, University of Minnesota School of Nursing, Weaver-Densford Hall, 308 Harvard Street SE, Minneapolis, MN 55455, USA. treat001@umn.edu an average peak oxygen consumption (VO 2 peak) of 15 ml/kg per minute in PAD patients, similar to that of individuals with stage III heart failure. 7,8 The efficacy of supervised exercise training for improving walking capacity in individuals with claudication is well documented Exercise training, consisting of repeated bouts of treadmill or track walking until moderate claudication pain is induced, generally results in an increase in the distance that patients can walk prior to onset of claudication (i.e. the pain-free walking distance, PFWD), as well as the distance that patients can walk before pain forces them to stop (i.e. the maximal walking distance, MWD). Walking exercise is currently considered the most effective method to improve walking capacity in this population. Improvement is hypothesized to result from local lower extremity skeletal muscle metabolic adaptations. 7,8,11,12,15 However, systemic adaptations resulting from exercise training also appear The Author(s), Reprints and permissions: / X

2 204 D Treat-Jacobson et al. to contribute to the clinical improvements seen in those with claudication. 16,17 Although walking exercise and associated induced muscle ischemia may stimulate local changes in involved skeletal muscles, claudication reduces tolerance of the duration and intensity of this mode of training and thus limits conditioning effects on the cardiovascular system. Walking exercise also induces undesirable, ischemia-induced, recurrent muscle and vascular inflammatory responses. In contrast, upper extremity aerobic exercise training, such as via arm-ergometry, does not cause lower limb ischemia and associated pain. Therefore, patients should be able to achieve a higher exercise intensity with arm-ergometry than with walking exercise training, thereby resulting in a greater potential for cardiovascular conditioning effects. Walker et al. 16 and Zwierska et al. 17 investigated the effects of arm versus leg cycle ergometry training on walking capacity in individuals with claudication, and demonstrated a comparable improvement with both forms of exercise. However, these investigations did not compare ergometry training with the accepted gold standard of treadmill exercise training (walking), nor did they measure training-induced changes in peak oxygen consumption using a standard graded treadmill walking protocol. Therefore, the purpose of this pilot study was to determine the relative improvement in walking distance in patients with claudication in response to supervised armergometry exercise training as compared to standard treadmill training, a combination of training modalities (treadmill walking and arm-ergometer), and usual medical care. We also determined whether training-induced effects were maintained or improved 12 weeks following completion of supervised exercise training. The effectiveness of training in improving aerobic power and associated hemodynamic parameters was also evaluated in each study group. We hypothesized that all groups performing supervised exercise would significantly improve PFWD and MWD, relative to the usual care group, and that the group performing a combination of both treadmill walking and arm-ergometry training would achieve the greatest benefit. Methods Design The Exercise Training for Claudication (ETC) study was a single-center, randomized, led pilot study comparing the relative efficacy of armergometry, treadmill exercise training, combination training (arm/treadmill), and usual care in improving walking distance and VO 2 peak in patients with lifestyle-limiting claudication. The trial was approved by the University of Minnesota s Institutional Review Board; written informed consent was obtained from each participant. Sample Study participants were recruited from the Twin Cities metropolitan area through clinician referral and media advertisements between 2003 and Following initial telephone interview screening, potentially qualifying individuals were scheduled for an initial screening visit. Volunteers were included if they were aged 18 years and had lifestyle-limiting claudication; an ankle brachial index (ABI) of 0.90 and/or a decrement in their ABI of 10% following a symptom-limited treadmill exercise test; were able to walk at a rate of 2.0 mph on a treadmill; and able and willing to participate in a 12-week supervised exercise program. Patients with diabetes mellitus were eligible for participation provided that their fasting glucose levels were within the acceptable range for exercise training according to standard guidelines. 18 Exclusion criteria included: unled hypertension (> 200 SBP (systolic blood pressure) and/ or DBP (diastolic blood pressure) > 100) 18 ; ischemic rest leg pain and/or leg/foot ulceration, or impending gangrene; exercise capacity limited by health problems other than claudication such as angina pectoris, severe arthritis, marked dyspnea on exertion; a recent myocardial infarction or unstable coronary heart disease; or a coronary or lower extremity revascularization procedure within the past 3 months. Participants were instructed to maintain their current dietary habits and prescribed medications throughout the study, and compliance was monitored. Those currently taking pentoxyphylline or cilostazol for treatment of claudication were included if medication use was initiated at least 3 months prior to enrollment into the study; however, they were instructed not to initiate or discontinue these medications during study participation unless recommended by their primary clinician. Sample size was calculated based on expected changes in the primary endpoint (MWD) based on data from previously published studies. 8 With 40 individuals randomized to one of the four arms, the study was powered at 90% using a two-tailed alpha of 0.05, assuming a 30% increase over the usual care group. Our goal was to enroll 45 individuals to adjust for an anticipated 10% drop-out rate. Procedures All measurements and training were performed at the Laboratory of Physiological Hygiene and Exercise

3 Exercise training for claudication 205 Science (Exercise Laboratory) at the University of Minnesota. During the initial screening visit, prior to initiation of any testing procedures, study procedures were explained, written informed consent was obtained, and a standard medical history questionnaire was completed. Qualified participants performed a symptom-limited, graded, cardiopulmonary treadmill exercise test (GXT) to determine their PFWD and MWD. Those who met study inclusion criteria following this first GXT were asked to return to repeat the test. Individuals whose second MWD was within 25% of the first MWD qualified for the study. If the second MWD varied > 25% from the first MWD, participants were asked to return for a third test. If the third MWD was within 25% of the second MWD, the individual qualified for the study. Those whose third MWD varied > 25% from the second MWD were excluded from further study participation. Data from the last screening GXT were used as baseline measurements. Following completion of baseline assessments, eligible participants were randomized by simple randomization tables to one of four study groups: Arm-ergometry, treadmill, combination, or usual medical care. The GXT was repeated within 1 week after completion of the 12-week exercise program and at 24 weeks (12 weeks after completion of training or usual care). Measurements Ankle brachial index (ABI) Resting ABI measurements were performed according to standard procedures to confirm the presence of PAD. 19 Participants rested supine for 10 minutes before brachial and ankle systolic blood pressures were assessed using a 5-MHz broad beam Doppler ultrasound. The higher systolic pressure in either arm was used as the denominator and the higher systolic ankle pressure (posterior tibial or dorsalis pedis artery) of each leg was used as the numerator for calculation of the ABI. Blood pressure and heart rate measurements SBP and DBP were measured in the seated and standing positions by auscultation and a sphygmomanometer in the right arm. Heart rate was measured by ECG. All measurements were performed following 5 minutes of quiet seated rest prior to exercise testing. Graded cardiopulmonary treadmill test (GXT) Walking distance and VO 2 uptake during exercise were measured during a symptom-limited GXT using a Quinton 4000 Exercise Monitoring System (Bothell, WA, USA). The GXT was performed in the post-absorptive state; no smoking or caffeinated beverages were permitted 3 hours prior to testing. Participants were instructed to take their usual medications with the exception of hypoglycemic agents, which were withheld until after the exercise test. All GXTs were supervised by a physician. Heart rate and rhythm during exercise were continuously monitored by ECG. Blood pressure was measured immediately prior to testing, at 3-minute intervals during the test, and for 6 minutes during recovery. The GXT protocol consisted of walking on the treadmill at a speed of 2 mph starting at a 0% grade (flat). The treadmill grade was increased 3.5% every 3 minutes until a 10.5% grade was obtained, at which time the speed was increased by 0.5 mph every 3 minutes, while maintaining the grade at 10.5%. This approximates an increase in exercise intensity of 1 metabolic equivalent (MET) per test stage. Participants were only permitted to hold the handrail lightly in order to maintain balance. Respiratory gas measurement was performed with a calibrated Sensormedic 2900 Metabolic Cart (Yorba Linda, CA, USA). During the GXT, participants rated their claudication pain severity every 30 seconds, using the following claudication pain scale: 0 = no pain, 1 = initial onset of mild claudication pain, 2 and 3 = moderate pain, 4 = submaximal, and 5 = maximal claudication pain (point at which the participant requested to stop). The distance walked prior to initial onset of pain was defined as PFWD; the distance at which the participant stopped due to claudication pain was defined as the MWD. The VO 2 peak was assessed during the last minute of exercise. Arm-ergometry exercise test Prior to randomization, within 1 week after the final screening GXT, participants underwent an assessment of upper-limb peak exercise capacity using a Monarch Arm Cycle Ergometer (Stockholm, Sweden) to establish an initial exercise prescription for participants randomized to one of the two groups performing arm exercise training. Following a 2- minute warm-up against no resistance, participants maintained a constant rate of 60 revolutions/minute, starting at a workload of 10 watts. The intensity was manually increased by 10 watts at 3-minute intervals, until the participant was unable to continue. The maximal power achieved was recorded as the power at which the patient stopped exercising. Monitoring of unsupervised exercise All participants (including those in the group) were asked to maintain a record of any exercise performed beyond the supervised exercise training. Participants were instructed to record the mode, intensity, and time spent in unsupervised exercise. Recorded unsupervised exercise was analyzed as a

4 206 D Treat-Jacobson et al. dichotomous variable (yes or no). Participants were considered to have performed unsupervised exercise if their exercise log listed moderate-intensity exercise (usually walking), for a minimum of 30 minutes a session, at least three times per week for the group participants and at least twice per week for exercise group participants. If such data were not recorded, it was assumed that unsupervised exercise was not performed. Intervention protocol Control group Control group participants were instructed to continue the prescribed usual medical care for their PAD. However, they were provided with specific, standardized, written walking instructions for claudication patients and a daily exercise record, and were asked to record the type and amount of any daily exercise. They also were requested to come to the Exercise Laboratory once a week for review of their exercise records, and measurement of resting blood pressure and heart rate. These weekly visits, specific exercise instructions, and exercise logs provided them professional attention as an incentive to maintain retention. Training programs Participants in all three exercise groups performed exercise in the Exercise Laboratory three times per week for 12 weeks, for a total of 36 sessions. Each session was 70 minutes in length, including 5 minutes of warm-up, 60 minutes of intermittent exercise, and 5 minutes of cool down. The training protocol is summarized in Figure 1. Heart rate, blood pressure, rating of perceived exertion (RPE), and claudication pain were recorded every 3 5 minutes during exercise training. Exercise intensity was modified to maintain a RPE of out of 20 using the Borg RPE scale The armergometry exercise prescription was designed to provide a similar relative intensity and volume of exercise (calculated by MET minutes of exercise) as the treadmill exercise (based on intermittent walking to moderate claudication pain). To achieve an exercise prescription similar in MET-minutes of exercise between groups, the arm-ergometry exercise group was prescribed progressively longer exercise bouts and shorter rest periods, which allowed the arm-ergometry group to gain similar METminutes of exercise as the treadmill group. Exercise adherence was monitored and recorded and participants were strongly encouraged to make up any missed sessions. Participants making up missed sessions were not permitted to exercise in the laboratory more than four times per week. All participants were strongly encouraged to complete 36 exercise training sessions. Arm-ergometry The arm-ergometry training protocol was modified from previous studies. 16,17 Each participant in the arm-ergometry group began exercising at one work level (10 watts) below the maximal level achieved during their baseline arm-ergometry test at a rate of 50 cycles per minute. Participants worked against this load intermittently for periods of 2 minutes of exercise, followed by 2 minutes of rest, for a total of up to 60 minutes. After 3 weeks of training, exercise intensity was increased to the work level in watts achieved during the baseline arm-ergometry test. Exercise time was progressively increased in each cycle by 1 minute every 2 3 weeks during the training period and rest periods were decreased to 1 minute, for a maximal volume of 5 minutes of exercise and 1 minute of rest for 60 minutes (50 minutes of exercise). Treadmill walking Participants randomized to the treadmill walking group initially began exercising at 2 mph (3.2 km/h) at a 0% grade. They walked until their claudication pain became moderately severe (4 of 5 on the claudication scale), and they then stopped, sat down, and rested until the pain subsided. This exercise/rest cycle was repeated throughout each 60-minute exercise session. When a participant was able to walk for 8 minutes at the initial workload without having to stop because of moderately severe claudication, the treadmill grade was increased by increments of 0.5% until an 8 10% grade was achieved. Subsequently, exercise intensity was increased during training sessions by increasing the treadmill speed by increments of mph ( km/h) as tolerated. Arm-ergometry and treadmill walking combination Combination group participants completed both upper and lower body training at each session. After a 5-minute warm-up period, sessions consisted of 20 minutes of intermittent arm-ergometry exercise utilizing a similar protocol as the armergometry group, followed by 40 minutes of intermittent treadmill walking, utilizing a similar protocol as the treadmill group, for a total of 60 minutes of training per session. Statistical analysis Data were entered by study personnel and verified for accuracy. Variables were tested for normality and descriptive statistics were completed. Group differences in baseline demographic and medical variables were assessed using chi-squared tests and t-tests. Values for MWD and PFWD were log transformed to normalize the distribution due to large,

5 Exercise training for claudication 207 Figure 1 (A) Treadmill exercise training program. (B) Arm-ergometry exercise program. positively skewed standard deviations. Differences between groups in the primary endpoint (MWD) and secondary endpoints (PFWD, VO 2 peak, resting blood pressure and heart rate) at 12 and 24 weeks were assessed using analysis of covariance (ANCOVA) with unsupervised exercise and baseline values for each of these variables entered as covariates. Planned comparisons between individual groups were performed using Bonferroni adjustment for multiple comparisons. Relationships between changes in primary and secondary end points were assessed via Pearson s productmoment correlations. The authors had full access to the data and take responsibility for its integrity. All three authors have read and approve the manuscript as written. Results Figure 2 presents the flow of participants through the study. To summarize, following telephone

6 208 D Treat-Jacobson et al. interview, 102 potentially qualified participants were screened in the Exercise Laboratory by medical history review and ABI assessment. Approximately 40% of these individuals were excluded from participation because of failure to confirm the presence of PAD and/or claudication. An additional 17 patients did not meet entry criteria following the GXT. Forty-five eligible participants were subsequently randomized; however, four of them withdrew from the study before completing exercise training. Forty-one participants completed 12 weeks of intervention and follow-up assessment and 31 (76%) participants who completed the 12- week testing also completed the 12-week posttraining (24 week) follow-up assessments. Participant characteristics The study sample comprised predominantly male Caucasians with moderately severe PAD. The majority had medical histories that included hypertension, dyslipidemia, cigarette smoking, and/or coronary heart disease (Table 1). There were no statistical differences between groups in baseline demographic or medical characteristics or in currently prescribed medications. Although not statistically significant, there were fewer participants with diabetes mellitus in the treadmill group (9.1%) than in the other three study groups ( %). Participant adherence to training Twenty (61%) participants from the exercise training group completed all 36 exercise sessions within 14 weeks; 12% completed all 36 sessions but took longer than 14 weeks; and 97% of participants completed at least 75% of the prescribed training sessions. Reasons for non-attendance included illness, vacations, scheduling conflicts or transportation difficulties. MET-minutes of exercise training performed per session was [mean (SD)] (99.7), (96.9), and (75.8) for the arm-ergometry, treadmill and combination groups, respectively (p = 0.983). In addition to the supervised training sessions, all exercise and participants were encouraged to exercise outside of the laboratory setting. Based on the exercise records submitted by the exercise participants, 45% of participants in the treadmill walking group reported participating in outside exercise (at least 2 additional days per week in addition to the supervised sessions) versus 25% in the combination group, and Figure 2 Diagram of study flow.

7 Exercise training for claudication 209 Table 1 Sample characteristics Demographic and medical variables a Arm ergometry or n (%) Treadmill walking or n (%) Combination or n (%) Control or n (%) Total or n (%) Age, years 64 (8.6) 64 (11.7) 71.9 (11.3) 70 (7.8) 67.7 (10.5) Male 8 (80%) 7 (64%) 7 (58%) 7 (88%) 29 (70.7%) Caucasian 9 (90%) 11 (100%) 9 (75%) 6 (75%) 35 (85.4%) Diabetes 6 (60%) 1 (9%) 5 (42%) 3 (37.5%) 15 (36.6%) Current/past smoking 10 (100%) 11 (100%) 10 (83%) 7 (88%) 38 (92.7%) Pack years 45.7 (24.8) 46.6 (35.3) 35.4 (19.5) 36.4 (21.6) 41.5 (26.1) Dyslipidemia 10 (100%) 9 (81.8%) 10 (100%) 6 (75%) 37 (90.2%) Hypertension 7 (70%) 9 (82%) 10 (83%) 7 (88%) 33 (80.5%) Lowest resting ABI 0.66 (0.15) 0.68 (0.11) 0.65 (0.10) 0.69 (0.15) 0.67 (0.1) Prior leg revascularization 2 (20%) 4 (36.4%) 3 (25%) 3 (37.5%) 12 (29.3%) Coronary heart disease 7 (70.0%) 7 (63.6%) 7 (58.3%) 4 (50%) 27 (65.9%) Body mass index, mm/kg (8.8) 26.4 (3.1) 28.3 (3.8) 29.1 (5.4) 28.3 (5.6) Current medications a Cilostazol 2 (20.0) 3 (27.3) 2 (16.7) 0 (0.0) 7 (17.1) Antiplatelet agent 10 (100.0) 8 (72.7) 9 (75.0) 6 (75.0) 33 (80.5) Warfarin 0 (0.0) 3 (27.3) 1 (8.3) 0 (0.0) 4 (9.8) Lipid-lowering agent 9 (90.0) 8 (72.7) 11 (91.7) 4 (50.0) 32 (78.0) Beta-blocking agent 5 (50.0) 4 (36.4) 6 (50.0) 4 (50.0) 19 (46.3) ACE inhibitor 8 (80.0) 4 (36.4) 7 (58.3) 2 (25.0) 21 (51.2) a No significant differences between groups by chi-squared tests or t-tests. 20% in the arm-ergometry group. Conversely, 75% of group participants reported participating in outside exercise at least 3 days per week. Maximal walking distance (MWD) There were no baseline differences in MWD between groups. There were significant effects of baseline MWD and outside exercise at both 12 and 24 weeks on improvement in MWD. After ling for these covariates, MWD increased significantly from baseline to 12 weeks across all three exercise groups, compared to the group, with no significant differences observed between exercise groups (p = 1.0). Percent improvement was 53%, 69%, and 68% in the arm-ergometer, treadmill, and combination groups, respectively, compared to an 11% increase in the group (Table 2). Increased MWD in the exercise groups was equivalent to two to three and a half blocks, compared to less than one block in the group. After 12 weeks of follow-up (24 weeks), improvement in MWD was maintained in the treadmill and arm-ergometry groups, whereas it declined in the combination group to a point where it was no longer statistically different from the group (Table 2; Figure 3A). Pain-free walking distance (PFWD) There was no difference in mean PFWD between treatment groups at baseline. The arm-ergometry, treadmill, and combination groups increased PFWD from baseline to 12 weeks of intervention by 82%, 54%, and 60%, respectively (approximately equivalent to one additional city block) and were not statistically different from one another (p = 1.0). The group increased PFWD by only 1%. The percent change in PFWD from baseline to 24 weeks (12 weeks post intervention) was 123%, 80%, 38%, and 18% in the arm-ergometry, treadmill, combination, and groups, respectively. There were significant effects of baseline PFWD at 12 and 24 weeks (p > 0.001) and for outside exercise at 24 weeks only (p = 0.02) on improvement in PFWD. After ling for these covariates, there remained a significant difference in PFWD in the overall ANCOVA from baseline to 12 weeks and from baseline to 24 weeks (Table 3; Figure 3B). Post-ANCOVA planned comparisons indicated that the arm-ergometry group significantly increased PFWD both from baseline to 12 weeks and baseline to 24 weeks as compared to the group; however, the statistically adjusted improvement in PFWD from baseline to 12 weeks and 24 weeks in the treadmill and combination groups failed to reach statistical significance. Resting hemodynamic variables There were no significant differences between the four study groups in baseline resting ABI, heart rate or seated blood pressure levels. There also were no significant changes in any group in resting ABI, heart rate, or DBP values following 12 weeks of intervention or at the 24-week follow-up measurement point. However, the mean resting SBP

8 210 D Treat-Jacobson et al. Table 2 maximal walking distance (MWD) from baseline to 12 and 24 weeks Group Baseline MWD (m) MWD (m) Baseline to 12 weeks MWD log transformation 12 weeks Adj. mean (SE) ANCOVA F = 10.3 p < MWD (m) 24 weeks MWD log transformation 24 weeks Adj. mean (SE) ANCOVA F =5.7 p =0.004 Arm-ergometry (188.7) (126.7) 0.18 (0.03) p = vs (164.1) 0.23 (0.04) p =0.01vs Treadmill (290.9) (163.5) 0.23 (0.03) p < vs (162.2) 0.20 (0.03) p =0.02vs Combination (184.1) (72.7) 0.22 (0.03) p < vs (159.6) 0.12 (0.03) p =0.73vs Control (185.2) 45.3 (92.7) 0.02 (0.04) 73.3 (65.6) 0.03 (0.04) Planned comparisons: Bonferroni adjustment for multiple comparisons. level was significantly lower after 12 weeks of exercise training only in the arm-ergometer group, as compared to the group. This decrease was maintained at 24 weeks (Table 4). Functional capacity The mean VO 2 peak in ml/kg per minute improved with exercise training from 14.9 to 16.3 (12%) in the treadmill group, from 14.0 to 14.7 (4%) in the combination group, and from 14.1 to 15.6 (11%) in the arm-ergometry group, while the group showed a slight decrease ( 4%) from 13.6 to 13.0 ml/kg per minute. However, improvements in the exercise groups did not reach statistical significance when compared to the group, possibly due to the large variability observed within groups in response to exercise training. There were no significant changes in VO 2 peak in any group from 12 to 24 weeks. Across groups, the correlation between change in PFWD and change in MWD was significant (r = 0.43; p = 0.006). A significant association also was observed between change in VO 2 peak and change in MWD (r = 0.45, p = 0.005). Discussion Figure 3 (A) Maximal walking distance at baseline, at 12- week intervention and at 24 weeks (logarithmic transformation). (B) Pain-free walking distance at baseline, at 12-week intervention and at 24 weeks (logarithmic transformation). To our knowledge, this is the first study to directly compare supervised exercise training by dynamic arm exercise to treadmill walking in individuals with claudication. Participants performing 12 weeks of supervised arm-ergometry demonstrated substantial improvements in both PFWD and MWD, similar to those performing treadmill training. This finding provides evidence that exerciseinduced improvement in walking ability in patients with claudication can result from changes in systemic as well as local skeletal muscle adaptations. Surprisingly, after statistical adjustment for multiple variables, only the arm-ergometry group showed a significant improvement in PFWD, compared with the group. There was a positive trend toward improvements in PFWD observed in both the treadmill and combination groups, but this did not achieve statistical significance, possibly due to the large variability in response to exercise training observed in these two groups. Previous studies also

9 Exercise training for claudication 211 Table 3 pain-free walking distance (PFWD) from baseline to 12 and 24 weeks Group Baseline PFWD (m) PFWD (m) 12 weeks PFWD log transformation 12 weeks Adj. mean (SE) ANCOVA F =3.3 p = PFWD (m) 24 weeks PFWD log transformation 24 weeks Adj. mean (SE) ANCOVA F = 5.29 p =0.006 Arm-ergometry (64.1) 89.6 (74.0) 0.24 (0.07) p = 0.03 vs 39.7 (97.2) 0.34 (0.08) p =0.01vs Treadmill (151.4) 91.6 (148.4) 0.14 (0.06) p = 0.20 vs (180.7) 0.22 (0.07) p =0.11vs Combination (100.4) (109.94) 0.17 (0.06) p = 0.11 vs 21.6 (81.3) 0.04 (0.07) p =1.00vs Control (62.2) 4.0 (45.4) 0.10 (0.08) 10.9 (27.4) 0.07(0.09) Planned comparisons: Bonferroni adjustment for multiple comparisons. demonstrated a great degree of variability in responsiveness of PFWD to treadmill exercise training. 16,17,20,21 PFWD was similar by sex and race, between smokers and non-smokers, and in those with or without diabetes. Significant improvements in MWD occurred in all three exercise training groups, as compared to the group, with no statistical differences between exercise groups. The increase in MWD in the present study is within the range of improvement in walking capacity reported in previous reviews. 14,20,21 The finding of improvement in MWD in response to both arm-ergometry and treadmill training indicates that the mechanism of improvement is not solely an adaptation to exercise-induced ischemia in leg skeletal muscle. Although local skeletal muscle adaptations may have contributed significantly to improvement in MWD in the treadmill group, such adaptations are not plausible in the arm-ergometry group. Further, the improvement observed in MWD after both modes of exercise training was associated with increased peak VO 2, supporting the contribution of improved central cardiorespiratory function to increased MWD. This also is compatible with a transfer-effect, whereby exercise training in one muscle group leads to improvement in the exercise capacity of untrained muscle groups through a generalized systemic training effect. Previous studies involving sedentary healthy individuals with low initial fitness levels have demonstrated significant transfer effects following arm-ergometry, during either maximal cycle ergometer or treadmill testing Since individuals with claudication generally have very low fitness levels, they are more likely to experience such a transfer effect than people with higher baseline fitness levels. In this pilot study, the treadmill group had a trend toward a greater increase in MWD as compared to the arm-ergometry group (294 meters, 69% improvement versus 182 meters, 53% improvement; p = ns). This difference may be clinically important and could become statistically significant if the trends seen in this study were repeated with a larger population. Conversely, the arm-ergometry group trended toward a greater percent improvement in PFWD compared to the treadmill group. This lends support to the theory that improvement in PFWD and MWD with exercise training in PAD patients could arise from different mechanisms as Table 4 resting SBP (mmhg) from baseline to 12 and 24 weeks Group Resting SBP (mmhg) Baseline Resting SBP (mmhg) 12 weeks resting SBP (mmhg) 12 weeks Adj. mean (SE) ANCOVA F =3.58 p =0.023 Resting SBP (mmhg) 24 weeks resting SBP (mmhg) 24 weeks Adj. mean (SE) a Arm-ergometry (15.22) (12.61) (4.77) p = 0.033vs (15.82) (6.75) Treadmill (18.18) (20.02) 2.44 (4.49) p = 1.00 vs (21.07) 5.51 (5.89) Combination (20.58) (16.18) 0.53 (4.40) p = 1.00 vs (18.65) 9.23 (6.68) Control (15.24) (14.85) 5.12 (5.53) (23.20) 4.00 (7.36) Planned comparisons: Bonferroni adjustment for multiple comparisons. SBP, systolic blood pressure. a ANCOVA F = 0.68; p = NS.

10 212 D Treat-Jacobson et al. previously postulated. 8 These findings suggest that systemic adaptations may play a greater role in improvement in PFWD, whereas improvement in MWD, particularly with treadmill training, may result from a combination of systemic and local skeletal muscle adaptations. In this study, the armergometry group had a greater opportunity for a systemic training effect because their exercise training intensity and duration was only limited by exercise-induced fatigue as compared to lower extremity claudication in the treadmill group. The improvements in PFWD and MWD in the arm-ergometry group compare favorably with the results of two previous studies using armergometry exercise training in patients with claudication. 16,17 However, it should be noted that these two studies did not use a standard GXT to assess MWD. Instead, they used a so-called shuttle test, which reduces the comparability of their results to those of the present study. These studies also did not have a treadmill exercise training group for comparison purposes, but instead compared the response of those performing arm or leg cycle ergometry only to participants. The improvement seen in PFWD in the arm-ergometry group is also comparable to that found following a 12-week pain-free treadmill training program. 25 Additionally, the relatively small response seen in the group (1% increase in PFWD and 18% increase in MWD) is similar to that reported in other randomized led exercise studies on patients with claudication. 8,12,26,27 An unexpected finding in the present study was that the combination group had similar improvements to the other exercise groups, rather than an expected greater improvement due to the additive effects of arm-ergometry and treadmill exercise training. It is not known whether the outcome would be improved if the two types of exercise were preformed on alternate days, rather than during the same session. It may be that the total volume of exercise (which was similar in all exercise groups) may be a more important contributor to change in walking distance than the mode of exercise. The observation of a significant decrease in resting SBP from baseline to 12 weeks only in the armergometry group also was an unexpected finding since this study was not powered to detect exercise-induced changes in blood pressure levels. This observation warrants further investigation with a larger sample with sufficient power to detect such differences should they exist. Limitations The relatively small sample size in this pilot study is a limitation, and thus these results should be interpreted with caution until they are replicated in a larger randomized clinical trial. Another limitation of this study was that although the physician supervising the treadmill tests was blinded as to treatment group assignment, other staff assisting with testing were not blinded. To mitigate for this limitation, care was taken to ensure standardization of exercise testing procedures. Each participant was given the same pretest instructions and tests were performed according to a strict protocol across groups. While there were mean improvements in PFWD and MWD in each treatment group, several participants (at least one in each group) did not demonstrate improvement despite having the same intervention. The lack of responsiveness in this small subset of participants might be related to a number of factors including the intensity of exercise training performed in the arm-ergometry group, or individual pain tolerance and perceived pain intensity during treadmill training. The broad inclusion criteria and wide range of baseline walking capacity of study participants increases the variability of the results, but this is offset by the increase in the generalizability of our findings to other populations of PAD patients. The low drop-out rate and excellent exercise training compliance rate was likely related to well-motivated individuals and flexible training schedules. Regularly scheduled weekly meetings with the group also may have contributed to the low drop-out rate in this group. Conclusion The data from this study demonstrate for the first time that dynamic arm exercise training can improve walking capacity in people with PADinduced claudication compared to participants receiving usual care, and improvement was not different from that seen in response to treadmill walking exercise training. Confirmation of these findings in a larger study sample is needed to provide support for the novel approach of dynamic arm exercise as a therapeutic exercise option for PAD patients whose disease severity and claudication limit their ability to perform lower extremity exercise training, or who have an aversion to exercise-induced ischemic pain. Acknowledgements We wish to thank Professors Ruth Lindquist and Jean Wyman for careful review of the manuscript, Kay Savik for invaluable statistical advice, the research assistants who helped with the implementation of this study, and especially the participants who generously donated their time and energy.

11 Exercise training for claudication 213 Funding sources This study was supported with funding from the American Heart Association Northland Affiliate: Scientist Development Grant # Z (Treat- Jacobson, PI), from the Fesler Lampert Chair of Aging Studies Award, and from the University of Minnesota Academic Health Center Clinical Research Scholar Award. Dr Leon is partially supported by the Henry L Taylor Professorship in Exercise Science and Health Enhancement. Disclosures The authors have no conflicts to report. References 1 Ostchega Y, Paulose-Ram R, Dillon CF, Gu Q, Hughes JP. Prevalence of peripheral arterial disease and risk factors in persons aged 60 and older: data from the National Health and Nutrition Examination Survey J Am Geriatr Soc 2007; 55: Allison MA, Ho E, Denenberg JO, et al. Ethnic-specific prevalence of peripheral arterial disease in the United States. Am J Prev Med 2007; 32: Criqui MH, Fronek A, Barrett-Connor E, Klauber MR, Gabriel S, Goodman D. The prevalence of peripheral arterial disease in a defined population. Circulation 1985; 71: 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: Khaira HS, Hanger R, Shearman CP. Quality of life in patients with intermittent claudication. Eur J Vasc Endovasc Surg 1996; 11: Pell JP. Impact of intermittent claudication on quality of life. The Scottish Vascular Audit Group. Eur J Vasc Endovasc Surg 1995; 9: Hiatt WR. Exercise physiology in cardiovascular diseases. Curr Opin Cardiol 1991; 6: Hiatt WR, Regensteiner JG, Hargarten ME, Wolfel EE, Brass EP. Benefit of exercise conditioning for patients with peripheral arterial disease. Circulation 1990; 81: Regensteiner JG, Steiner JF, Hiatt WR. Exercise training improves functional status in patients with peripheral arterial disease. J Vasc Surg 1996; 23: Regensteiner JG. Exercise in the treatment of claudication: assessment and treatment of functional impairment. Vasc Med 1997; 2: Stewart KJ, Hiatt WR, Regensteiner JG, Hirsch AT. Exercise training for claudication. N Engl J Med 2002; 347: Hiatt WR, Wolfel EE, Meier RH, Regensteiner JG. Superiority of treadmill walking exercise versus strength training for patients with peripheral arterial disease. Implications for the mechanism of the training response. Circulation 1994; 90: Gardner AW, Katzel LI, Sorkin JD, Goldberg AP. Effects of long-term exercise rehabilitation on claudication distances in patients with peripheral arterial disease: a randomized led trial. J Cardiopulm Rehabil 2002; 22: Leng GC, Fowler B, Ernst E. Exercise for intermittent claudication. Cochrane Database Syst Rev 2000; 2: CD Hiatt WR, Nawaz D, Brass EP. Carnitine metabolism during exercise in patients with peripheral vascular disease. J Appl Physiol 1987; 62: Walker RD, Nawaz S, Wilkinson CH, Saxton JM, Pockley AG, Wood RF. Influence of upper- and lower-limb exercise training on cardiovascular function and walking distances in patients with intermittent claudication. J Vasc Surg 2000; 31: Zwierska I, Walker RD, Choksy SA, Male JS, Pockley AG, Saxton JM. Upper- vs lower-limb aerobic exercise rehabilitation in patients with symptomatic peripheral arterial disease: a randomized led trial. J Vasc Surg 2005; 42: American College of Sports Medicine. Guidelines for exercise testing and prescription. Philadelphia: Lea and Febiger, 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. Vasc Med 1996; 1: Gardner AW, Poehlman ET. Exercise rehabilitation programs for the treatment of claudication pain. A metaanalysis. JAMA 1995; 274: Bulmer AC, Coombes JS. Optimising exercise training in peripheral arterial disease. Sports Med 2004; 34: Loftin M, Boileau RA, Massey BH, Lohman TG. Effect of arm training on central and peripheral circulatory function. Med Sci Sports Exerc 1988; 20: Lewis S, Thompson P, Areskog NH, et al. Transfer effects of endurance training to exercise with untrained limbs. Eur J Appl Physiol 1980; 44: Rosler K, Hoppeler H, Conley KE, Claassen H, Gehr P, Howald H. Transfer effects in endurance exercise. Adaptations in trained and untrained muscles. Eur J Appl Physiol 1985; 54: Mika P, Spodaryk K, Cencora A, Unnithan VB, Mika A. Experimental model of pain-free treadmill training in patients with claudication. Am J Phys Med Rehabil 2005; 84: Tsai JC, Chan P, Wang CH, et al. The effects of exercise training on walking function and perception of health status in elderly patients with peripheral arterial occlusive disease. J Intern Med 2002; 252: Holm J, Dahllöf AG, Björntorp P, Scherstén T. Enzyme studies in muscles of patients with intermittent claudication. Effect of training. Scand J Clin Lab Invest Suppl 1973; 128: