Author + information
- Received September 7, 2016
- Revision received January 10, 2017
- Accepted January 27, 2017
- Published online April 3, 2017.
- Ambarish Pandey, MDa,
- Subhash Banerjee, MDa,
- Christian Ngo, MDa,
- Purav Mody, MDa,
- Steven P. Marso, MDa,
- Emmanouil S. Brilakis, MD, PhDa,
- Ehrin J. Armstrong, MD, MSb,
- Jay Giri, MD, MPHc,
- Marc P. Bonaca, MD, MPHd,
- Aruna Pradhan, MD, MPHd,
- Anthony A. Bavry, MD, MPHe and
- Dharam J. Kumbhani, MD, SMa,∗ ()
- aDivision of Cardiology, Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, Texas
- bSection of Cardiology, Denver VA Medical Center and University of Colorado School of Medicine, Denver, Colorado
- cCardiovascular Medicine Division, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania
- dDivsion of Cardiovasular Medicine, Brigham and Women’s Hospital, Boston, Massachusetts
- eDepartment of Medicine, University of Florida, Gainesville, Florida
- ↵∗Address for correspondence:
Dr. Dharam J. Kumbhani, UT Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, Texas 75390-9047.
Objectives The authors performed a meta-analysis of randomized controlled trials to compare the efficacy of initial endovascular treatment with or without supervised exercise training (SET) versus SET alone in patients with intermittent claudication.
Background Current guidelines recommend SET as the initial treatment modality for patients with intermittent claudication, in addition to optimal medical therapy. The role of endovascular therapy as primary treatment for claudication has been controversial.
Methods The primary outcome was treadmill-measured maximal walk distance at the end of follow-up. Secondary outcomes included resting ankle brachial index (ABI) and treadmill-measured ischemic claudication distance on follow-up. Risk of revascularization or amputations was also compared. Pooled estimates of the difference in outcomes between endovascular therapy with or without SET and SET-only groups were calculated using fixed and random effects models.
Results A total of 987 patients from 7 trials were included. In pooled analysis, compared with SET only (reference group), patients that underwent combined endovascular therapy and SET had significantly higher maximum walk distance (standardized mean difference 0.79 [95% confidence interval (CI): 0.18 to 1.39]; weighted mean difference 98.9 [95% CI: 31.4 to 166.4 feet], and lower risk of revascularization or amputation (odds ratio 0.19 [95% CI: (0.09 to 0.40]; p < 0.0001, number needed to treat = 8) over a median follow-up of 12.4 months. By contrast, revascularization was not associated with significant improvement in exercise capacity or risk of future revascularization or amputation, compared with SET alone. Follow-up ABI was significantly higher among patients that underwent endovascular therapy with or without SET as compared with SET alone.
Conclusions Compared with initial SET only, endovascular therapy in combination with SET is associated with significant improvement in total walking distance, ABI, and risk of future revascularization or amputation. By contrast, endovascular therapy-only was not associated with any improvement in functional capacity or clinical outcomes over an intermediate duration of follow-up.
- exercise capacity
- intermittent claudication
- peripheral arterial disease
- supervised exercise training
Peripheral arterial disease (PAD) is a manifestation of systemic atherosclerosis that affects more than 8.5 million adults >40 years of age and confers an increased risk for cardiovascular morbidity and mortality (1–5). Intermittent claudication is the classical symptomatic clinical presentation of PAD, and is associated with reduced exercise capacity and poor quality of life among these patients (1,6,7). Current American College of Cardiology/American Heart Association guidelines recommend a program of supervised exercise training (SET) as an initial treatment modality for patients with intermittent claudication (Class I) (8). Endovascular therapies are recommended for patients with intermittent claudication when they do not respond to initial exercise and medical therapies (Class IIa) and have favorable anatomy for intervention. However, the role of endovascular therapies as an initial treatment option alone or as an upstream adjunct to SET in patients with intermittent claudication remains controversial. Several recent trials have compared endovascular therapies with or without SET versus SET alone as an initial treatment strategy for PAD patients with intermittent claudication, and reported inconsistent results (9–12). In this study, we performed a meta-analysis of randomized controlled trials to compare the efficacy of initial endovascular treatment with or without SET versus SET alone in patients with intermittent claudication.
A search of MEDLINE and Embase was performed using these key terms: intermittent claudication, claudication, claudicants, peripheral arterial disease, exercise training, exercise therapy, supervised exercise training, angioplasty, stents, endovascular procedure, surgical revascularization, and percutaneous revascularization. Search was limited to human studies published after 1990, with no restriction for language. From these lists, published clinical trials comparing the efficacy of SET versus endovascular therapy (balloon angioplasty or stenting) in patients with PAD and ischemic claudication were identified. The references cited in the included papers were also examined for additional studies. Trials presented at recent national and international cardiovascular conferences were also assessed. The search strategy, study selection and analysis were carried out in accordance with the PRISMA statement for systematic reviews (Figure 1) (13).
All selected abstracts and titles were scanned independently by 2 reviewers (A.P. and D.J.K.) to identify papers for potential inclusion. Studies included in this analysis were required to have: 1) a randomized controlled trial design with an endovascular therapy arm and a SET arm; 2) data on baseline symptom status of study participants; 3) clearly defined intervention (endovascular therapy alone or in combination with SET) and control groups (SET only); and 4) objective measures of exercise capacity at the end of the study. Studies involving unsupervised or home-based exercise training or exclusively surgical revascularization of PAD lesions were excluded from the study; however, hybrid revascularization (endovascular therapy + surgical revascularization) studies were included.
Data extraction and risk of bias assessment
Data extraction was performed independently by 3 investigators (A.P., P.M., and D.J.K.). Clinical characteristics, target lesion location, type of endovascular intervention performed, outcomes measured at baseline and end of study, follow-up period, and interval clinical events were extracted from individual studies and entered into a data extraction form. For multiple studies published from a single dataset, the largest study with primary findings was included in the analysis. Discrepancies between reviewers were resolved through discussion and consensus. Risk of bias was assessed using the Cochrane risk of bias assessment tool by 2 independent investigators (A.P. and D.J.K.) and studies were classified as low-, moderate-, or high-quality study (14).
Primary outcome of interest for this study was maximal walking distance at follow-up. Secondary outcomes included resting ankle brachial pressure index (ABI) and ischemic claudication distance at follow-up. The follow-up data on outcomes were consistently reported in most included studies for 12 to 18 months post-randomization. Pooled incidence of need for revascularization procedures and amputations during follow-up was also compared between the different study arms.
The included studies were analyzed as per the intention-to-treat principle. Two separate analyses were conducted: endovascular therapy with SET versus SET alone, and endovascular therapy alone versus SET alone, as the initial treatment option. For continuous variables, the standardized and weighted mean differences (SMD, WMD) and corresponding 95% confidence intervals (CIs) were computed, using fixed and random effects modeling. The SMD values are preferentially reported to account for differences in treadmill protocols, but where necessary, the WMD values are provided for clinical relevance. If treadmill measures were reported in time units, they were converted to distance units, based on the treadmill protocol. For the Gardner protocol, distance in meters was calculated as: distance (m) = time (in min) × (53.33 m/min [3.2 km/h]) (15). Summary odds ratios (OR) with 95% CI were computed for each dichotomous outcome. In situations where no events were noted in one arm of a trial, we used an automatic “zero cell” correction so that studies with no events in a given arm would still be included for analysis. To assess the effect of the location of dominant lesion (aortoiliac vs. femoropopliteal) and the type of endovascular therapy (stent vs. no stent), random-effect meta-regression models were constructed for the primary outcome.
Heterogeneity between studies was assessed using I2 statistic and publication bias with Egger’s method (16). For outcomes with significant heterogeneity, the random effects model is reported in the text and figures; for all others, the fixed effects models are reported. All p values were 2-tailed with statistical significance specified at 0.05 and CI computed at the 95% level. All statistical analyses were conducted using STATA version 12.0 statistical software (Stata Corporation, College Station, Texas).
A total of 987 patients from 7 randomized control trials (constituting 9 total comparison arms) (11,12,17–21) with a weighted median follow-up duration of 12.4 months (range 10 to 18 months) were enrolled. Of these, 530 were randomized to endovascular therapy versus SET and 457 to endovascular therapy + SET versus SET. All trials emphasized optimal medical therapy in both arms. The baseline, demographic, and clinical characteristics of the study participants are summarized in Table 1, and angiographic features in Table 2. An assessment of anatomic location of stenosis was performed in all the included studies. Overall, aortoiliac lesions were observed in 45% of the patients.
The SET protocol, endovascular therapy used, and outcomes assessed in the included studies are detailed in Tables 3 and 4. Treadmill protocols differed among the included studies. Percutaneous endovascular therapy with balloon angioplasty and selective stent placement was used in 5 studies, whereas 2 studies used only balloon angioplasty (stent-use range 0% to 100%, mean 38%). One study included both surgical and endovascular revascularization intervention. The MIMIC (Mild to Moderate Intermittent Claudication) trial reported separate outcome measures among subgroups of participants with femoropopliteal and aortoiliac disease (18). Thus, data from these subgroups are reported separately in the pooled analysis.
Quality assessment was performed through use of the Cochrane risk of bias assessment tool. One study (Perkins et al. ) was identified as low quality due to lack of reporting on study design including randomization technique, lack of blinding in outcome assessment, and missing information on dropout among study participants. Other studies (n = 6) were moderate quality with random sequence generation, allocation concealment, and blinded assessment of outcome. None of the studies were double blind due to the nature of intervention. Incomplete outcome or selective reporting was not observed in most (n = 6) included studies.
Effect of endovascular therapy only versus SET only (n = 5 comparison arms)
Pooling across the studies using a random effects model showed no significant difference in the maximum walk distance (SMD: −0.11 [95% CI: −0.59 to 0.36]; p = 0.64) (Figure 2) on follow-up between the 2 groups. There was evidence of significant heterogeneity for this endpoint (I2 = 87.5%). There was no evidence of publication bias (p = 0.43). On meta-regression, there were no significant differences in the pooled maximum walk distance based on stent use (p = 0.31) or presence of aortoiliac disease (p = 0.54) for this outcome.
Four studies reported ischemic claudication distance on follow-up. On pooled analysis, ischemic claudication distance was not different between the endovascular and SET arms (WMD: −39.18 [95% CI: −85.9 to 7.54]; p = 0.10). ABI measurement on follow-up was reported by all 5 studies. On pooled analysis, ABI was significantly higher among participants that underwent endovascular therapy as compared with SET only (SMD: 0.64 [95% CI: 0.38 to 0.90]; p < 0.0001; WMD: 0.15 [95% CI: 0.10 to 0.19]; p < 0.0001) (Figure 3). All 5 studies reported on the need for subsequent revascularization or amputation on follow-up. On pooled analysis, there was no difference in the risk of revascularization or amputation on follow-up (8.7% vs. 12.3%, OR: 0.68 [95% CI: 0.33 to 1.43]; p = 0.31) (Figure 4).
Effect of combined endovascular therapy and SET versus SET (n = 4 comparison arms)
On pooled analysis using random effects models, endovascular therapy with SET was associated with significantly higher maximal walking distance on follow-up compared with SET alone (SMD: 0.79 [95% CI: 0.18 to 1.39]; p = 0.01) (Figure 2). Pooled WMD for maximal walking distance between the 2 study groups was 98.9 feet (95% CI: 31.4 to 166.4). There was evidence of significant heterogeneity for this endpoint (I2 = 88.2%). There was no evidence of publication bias (p = 0.46). On meta-regression, there were no significant differences in the pooled maximal walk distance based on stent use (p = 0.11) or presence of aortoiliac disease (p = 0.25).
Ischemic claudication distance was reported in 2 studies with no difference between the 2 groups in random effects pooled analysis (SMD: 0.31 [95% CI: −0.26 to 0.87]; p = 0.29). All 4 studies reported ABI on follow-up. On pooled analysis, the combination of endovascular therapy and SET was associated with significantly higher ABI on follow-up compared with SET only (SMD: 0.62 [95% CI: 0.33 to 0.91]; WMD: 0.14 [95% CI: 0.10 to 0.17]; p < 0.0001) (Figure 3). All 4 studies reported on the need for subsequent revascularization or amputation on follow-up. On pooled analysis, the combination was associated with a significantly lower risk of revascularization or amputation on follow-up (3.5% vs. 17.3%, OR: 0.19 [95% CI: 0.09 to 0.40]; p < 0.0001) (Figure 4). The corresponding number needed to treat was 8 patients (95% CI: 6 to 12).
In this meta-analysis of 987 patients with stable intermittent claudication on medical therapy, endovascular revascularization only as an initial treatment strategy did not improve exercise capacity or lower risk of revascularization or amputation on intermediate-term follow-up compared with SET only. By contrast, the combination of endovascular therapy with SET significantly improved maximal walking distance and lowered the risk of downstream revascularization or amputation compared with SET only. Finally, endovascular revascularization with or without SET was associated with a significant 20% improvement in ABI.
Our study has important clinical and policy implications. SET has been shown to improve exercise capacity and quality of life in patients with coronary artery disease, heart failure, and other cardiopulmonary conditions. Similar benefits have also been reported in PAD patients, and SET currently carries a Class I recommendation for initial management of intermittent claudication (8). These beneficial effects of SET among PAD patients are attributed to improvement in endothelial function, improved vascular obstruction, reduced inflammation, and favorable changes in skeletal muscle phenotype with improved mitochondrial function (22). Current guidelines also state that revascularization should be considered only after failure of SET and guideline directed medical therapy (Class IIa). These recommendations are validated in our study, given the lack of difference in functional and limb outcomes observed between SET alone and revascularization alone. However, our study also suggests that the combination of the 2 therapies may have an additive effect over SET alone and deserves consideration in the next iteration of the guidelines.
Despite being a Class I recommendation, a major practical limitation of SET is that it is currently not reimbursed by Centers for Medicare & Medicaid Services (23). In addition, and to some extent as a result of this, easy availability of supervised exercise training programs for PAD patients is poor (24). Further, compliance throughout the prescribed duration of exercise therapy is poor, as also noted in the included studies. Our study suggests that renewed attempts are necessary to improve access and address reimbursement policies for SET. Besides SET, recent studies have also evaluated the role of home-based exercise training for management of PAD patients. In a recent clinical trial, McDermott et al. (25) demonstrated significant improvements in exercise capacity with use of home-based walking exercise among PAD participants. Other studies comparing home-based exercise versus revascularization in PAD patients have demonstrated comparable changes in maximal walking distance (26). Thus, home-based exercise may be a useful alternative for PAD patients when SET is not available as the primary treatment option, and is endorsed as a Class IIa recommendation by the guidelines as well (8). Future trials are needed to directly compare revascularization to home-based walking programs. To date, the only trial that we encountered on this topic is the IRONIC (Invasive Revascularization or Not in Intermittent Claudication) trial (26). One hundred fifty-eight patients with claudication were randomized to either revascularization with a home-based walking program or a home-based program alone. At 1 year, the combination was superior to the home-based program alone in improving claudication distance. A number of quality-of-life indices were also higher in the combination group (26).
From a mechanistic standpoint, we observed an interesting dichotomy between ABI values and functional improvements. Although resting ABI values were higher by approximately 20% during follow-up among participants receiving endovascular therapy with or without concurrent SET, the clinical benefits of endovascular therapy were limited to the combined endovascular therapy and SET group. Prior studies have reported that ABI is associated with exercise capacity and leg function in PAD patients (27,28). However, the association between therapy-related improvement in arterial blood flow and improvement in functional capacity is less well established, suggesting that additional mechanisms beyond anatomic disease may play a major role in mediating the functional benefits among PAD patients (20,29,30). Taken together, these findings suggest that whereas ABI may be an effective screening tool to identify PAD patients, its usefulness in guiding therapeutic decisions may be limited.
In a previous meta-analysis, Ahimastos et al. (31) demonstrated improvement in ischemic claudication distance among PAD patients with intermittent claudication that underwent a combination of revascularization and SET as compared with SET alone. However, this meta-analysis was limited by the small number of included studies, inclusion of observational studies, and predominant use of balloon angioplasty during revascularization. Moreover, several randomized trials have been published in the last 3 years that have demonstrated variable effects of revascularization on functional outcomes (11,12,26). To our knowledge, the current meta-analysis with 987 participants with PAD and intermittent claudication is one of the largest and most contemporary analyses on this topic to date.
An important limitation of our study is our inability to adequately assess a standardized major adverse limb outcome because information on acute and critical limb ischemia events during follow-up was not separately reported. For this reason, we could not differentiate between subsequent revascularization for stable symptoms (which is expected and not necessarily a “failure” of exercise) from revascularization for limb-threatening events. We observed a small number of amputations in the current analysis, and the indications for the same were not reported. Procedural safety for endovascular revascularization also could not be commented upon because it did not appear to be systematically reported by all studies. Other limitations include significant heterogeneity in the pooled analysis for exercise capacity. This could be related to differences in patient characteristics such as location of the lesions, differences in adherence to exercise training protocols, and differences in intervention strategies used. Further, because it is a meta-analysis, the validity of our results is dependent on the validity of the studies included. We did not have access to patient-level data. Furthermore, data on medical management of study participants in most trials were not reported completely, limiting our ability to assess the extent to which optimal medical therapy was used. Next, only intermediate-duration (12 to 18 months) follow-up results were consistently available in most studies; longer-term data are lacking. Moreover, most data comparing endovascular therapy with SET are from Europe, which may limit the generalizability of study findings to the U.S. population. Finally, our study protocol was not published a priori in a central database.
Among patients with stable PAD and intermittent claudication, compared with SET only, endovascular revascularization in combination with SET is associated with an improvement in total walking distance, ABI, and lower risk of revascularization or amputation over an intermediate duration of follow-up. By contrast, revascularization only is not associated with significant improvement in any clinical or functional outcomes despite improvement in ABI. These findings highlight the need to consider initial revascularization as an adjunctive therapy to SET but not as a primary treatment option in the initial management approaches for intermittent claudication. Future studies that have longer-term follow-up, compare contemporary revascularization techniques and exercise training protocols, and are adequately powered to assess relevant clinical outcomes, are needed to determine the most effective management strategy for these patients.
WHAT IS KNOWN? Supervised exercise training is recommended as an initial treatment modality for patients with intermittent claudication.
WHAT IS NEW? Endovascular revascularization only as an initial management strategy for intermittent claudication does not improve functional outcomes as compared with supervised exercise training alone over an intermediate duration of follow-up. In contrast, when used as an adjunct to exercise training, endovascular revascularization is associated with greater improvements in functional and anatomic outcomes as compared with exercise training alone.
WHAT IS NEXT? Future studies, with long follow-up and relevant clinical outcomes, that compare contemporary revascularization techniques and exercise training protocols are needed to determine the most effective management strategy for these patients.
Dr. Banerjee has received speakers honoraria from Medtronic, CSI, and Gore; and research and educational grants from Boston Scientific and Merck. Dr. Marso has received grants to his institution from The Medicines Company, Novo Nordisk; and personal fees from Novo Nordisk, Abbott Vascular, and AstraZeneca. Dr. Brilakis has received consulting/speaker honoraria from Abbott Vascular, Asahi, Cardinal Health, Elsevier, GE Healthcare, and St. Jude Medical; and research support from InfraRedx and Boston Scientific. Dr. Armstrong has been a consultant for Abbott Vascular, Boston Scientific, Cardiovascular Systems, Medtronic, Merck, and Spectranetics. Dr. Giri has received research funds to his institution from St. Jude Medical. Dr. Bavry has received honoraria from the American College of Cardiology. Dr. Bonanca has received research grant support from AstraZeneca and Merck; and is a consultant for AstraZeneca, Merck, Bayer, and Azalez. Dr. Kumbhani has received honoraria and research support from the American College of Cardiology. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- ankle brachial index
- confidence interval
- odds ratio
- peripheral arterial disease
- supervised exercise training
- standardized mean difference
- weighted mean difference
- Received September 7, 2016.
- Revision received January 10, 2017.
- Accepted January 27, 2017.
- 2017 American College of Cardiology Foundation
- de Lemos J.A.,
- Kumbhani D.J.
- Mozaffarian D.,
- Benjamin E.J.,
- Go A.S.,
- et al.
- McDermott M.M.
- Gerhard-Herman M.D.,
- Gornik H.L.,
- Barrett C.,
- et al.
- Anderson J.L.,
- Halperin J.L.,
- Albert N.M.,
- et al.
- Murphy T.P.,
- Cutlip D.E.,
- Regensteiner J.G.,
- et al.
- Murphy T.P.,
- Cutlip D.E.,
- Regensteiner J.G.,
- et al.
- Liberati A.,
- Altman D.G.,
- Tetzlaff J.,
- et al.
- Higgins J.P.,
- Altman D.G.,
- Gotzsche P.C.,
- et al.
- Higgins J.P.,
- Thompson S.G.,
- Deeks J.J.,
- Altman D.G.
- Gelin J.,
- Jivegard L.,
- Taft C.,
- et al.
- Greenhalgh R.M.,
- Belch J.J.,
- Brown L.C.,
- et al.
- Hamburg N.M.,
- Balady G.J.
- McDermott M.M.,
- Guralnik J.M.,
- Criqui M.H.,
- et al.
- Nordanstig J.,
- Taft C.,
- Hensater M.,
- Perlander A.,
- Osterberg K.,
- Jivegard L.
- Watson L.,
- Ellis B.,
- Leng G.C.