Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print November 5, 2001, 10.1148/radiol.2221001743
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2221001743v1
222/1/25    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by de Vries, S. O.
Right arrow Articles by Hunink, M. G. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by de Vries, S. O.
Right arrow Articles by Hunink, M. G. M.
(Radiology 2002;222:25-36.)
© RSNA, 2001

Evidence-based Practice

Intermittent Claudication: Cost-effectiveness of Revascularization versus Exercise Therapy1

Sybolt O. de Vries, MD, PhD, Karen Visser, MSc, Jelle A. de Vries, BSc, John B. Wong, MD, Magruder C. Donaldson, MD and M. G. Myriam Hunink, MD, PhD

1 From the Department of Health Sciences, University of Groningen, the Netherlands (S.O.d.V., J.A.d.V., M.G.M.H.); Department of Medicine, New England Medical Center, Tufts University School of Medicine, Boston, Mass (J.B.W.); Division of Vascular Surgery, Brigham and Women’s Hospital, Harvard Medical School, Boston, Mass (M.C.D.); Departments of Epidemiology and Biostatistics and Radiology, Erasmus Medical Center Rotterdam, Rm EE2140, Dr Molewaterplein 50, 3015 GE Rotterdam, the Netherlands (K.V., M.G.M.H.); and Department of Health Policy and Management, Harvard School of Public Health, Boston, Mass (M.G. M.H.). Received October 31, 2000; revision requested January 8, 2001; final revision received July 6; accepted August 3. Supported by a PIONIER award from the Netherlands Organization for Scientific Research. Address correspondence to M.G.M.H. (e-mail: hunink@epib.fgg.eur.nl).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 
PURPOSE: To compare the costs, effectiveness, and cost-effectiveness of alternative treatment strategies for intermittent claudication.

MATERIALS AND METHODS: By combining data from the literature and original patient data, a Markov decision model was developed to evaluate the societal cost-effectiveness. Patients presented with previously untreated intermittent claudication, and treatment options were exercise, percutaneous transluminal angioplasty (with stent placement, if necessary), and/or bypass surgery. Treatment strategies were defined as the initial therapy in combination with secondary treatment options should the initial therapy fail. The main outcome measures were quality-adjusted life days, expected lifetime costs (in 1995 U.S. dollars), and incremental cost-effectiveness ratios.

RESULTS: Compared with an exercise program, revascularization (either angioplasty or bypass surgery) improved effectiveness by 33–61 quality-adjusted life days among patients with no history of coronary artery disease. The incremental cost-effectiveness ratio was $38,000 per quality-adjusted life year gained when angioplasty was performed whenever feasible, as compared with exercise alone, and $311,000 with additional bypass surgery. The incremental cost-effectiveness ratios were sensitive to age, history of coronary artery disease, estimated health values for no or mild claudication versus severe claudication, and revascularization costs.

CONCLUSION: The results suggest that, on average, the expected gain in effectiveness achieved with bypass surgery for intermittent claudication is small compared with the costs. Angioplasty performed whenever feasible was more effective than was exercise alone, and the cost-effectiveness ratio was within the generally accepted range.

Index terms: Arteries, interventional procedures, 92.1268, 92.1269, 92.128 • Arteries, transluminal angioplasty, 92.1268, 92.128 • Cost-effectiveness • Extremities, abnormalities, 92.721, 92.76


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 
Treatment strategies for patients with intermittent claudication traditionally have been conservative (1,2). Few patients eventually develop limb-threatening symptoms, and the risks of perioperative mortality and morbidity—at least with surgical procedures—are considerable (3–5). Revascularization procedures are therefore generally postponed until initial conservative management fails. Many physicians consider exercise an inexpensive and effective method of improving the symptoms of claudication and recommend it as the initial treatment (6–9). However, individual responses to exercise vary considerably, and long-term compliance varies from 65% to 87% (10,11). Furthermore, because patients must invest time before reaping any health reward, exercise may not be as inexpensive as it seems when the societal perspective is considered: The cost of the time spent exercising must be taken into account.

With the development of percutaneous revascularization techniques, the interventional armamentarium for peripheral arterial disease has expanded considerably. Because the risks of periprocedural mortality and morbidity with percutaneous revascularization are low, some physicians advocate the use of such procedures at an early stage of treatment for intermittent claudication (12). Many patients with intermittent claudication eventually undergo revascularization, and earlier use of the currently available low-risk alternatives might prevent unnecessary suffering. On the other hand, both percutaneous and surgical procedures are expensive, and with a limited health care budget, one needs to consider whether the gain in quality-adjusted life expectancy justifies the costs.

The medical literature does not provide a direct answer as to what the optimal comprehensive treatment strategy is for patients with intermittent claudication. Research efforts have been directed mainly at comparing various initial revascularization procedures—for example, different types of material for aortic bifurcation grafts (13), primary versus secondary stent placement in iliac angioplasty (14), and bypass surgery versus angioplasty for femoropopliteal arterial disease (15,16). In two published clinical trials (17–19), revascularization was compared with exercise. Results of one of these studies (17) showed that the maximum walking distance was improved more with either bypass surgery or a combination of surgery and exercise than with exercise alone. In the other study (18,19), the investigators found that exercise improved the walking distance more than angioplasty. The numbers of patients in these two studies were small. Furthermore, treatment strategies in both studies were defined on the basis of the initial treatment only, and the economic consequences of the strategies were not taken into account.

The purpose of the present study was to evaluate the costs, effectiveness, and relative cost-effectiveness of various comprehensive treatment strategies for the treatment of intermittent claudication by assessing a combination of exercise, percutaneous transluminal angioplasty (PTA), and bypass surgery.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 
A decision model was developed to estimate, from the societal perspective (20), the quality-adjusted life expectancy, lifetime costs, and cost-effectiveness associated with five alternative treatment strategies for unilateral intermittent claudication. Therapeutic options included an exercise program, PTA, and/or bypass surgery. A treatment strategy was defined as the initial therapy combined with secondary treatment options should the initial treatment fail (Table 1). Treatment failure was defined as discontinuation of the exercise program in combination with severe claudication, graft failure or restenosis in combination with severe claudication, or progression to critical limb ischemia.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Evaluated Treatment Strategies

 
We assumed that all patients would undergo general examination and treatment for other atherosclerotic disease, and risk factor modification was not explicitly modeled. The treatment of critical limb ischemia—defined as pain at rest, ulcer, or gangrene—was the same for all strategies: In patients with suitable lesions, angioplasty was performed; in all other patients, bypass surgery was performed.

For analysis, data from the medical literature were combined with original patient data from three sources (3,14,21–52) (Appendix, Tables A1 and A2). Original patient data included those from a 5-year (1990–1995) consecutive series from (a) the vascular registry database at Brigham and Women’s Hospital, Boston, Mass, which included 722 patients (ie, Boston database); (b) the database of the exercise program at the University Hospital Groningen, the Netherlands, which included 329 patients (ie, Groningen database); and (c) a Dutch trial on oral anticoagulants at the Dijkzigt Hospital, Rotterdam, the Netherlands, which included 547 patients (ie, Rotterdam database) (21). The protocols for these studies were approved by the appropriate institutional review boards, and informed consent was obtained from all patients. Baseline-case values and ranges for the key parameters of the model, technical details about the model structure, and data sources are presented in the Appendix. Pertinent assumptions are summarized in the following text.

Model Structure
Figure 1 is a schematic representation of the model. It is a Markov model (54) that simulates individual disease histories, from presentation to death, for all alternative treatment strategies and keeps track of the time spent in various health states and the accumulated costs. Health states were defined as all possible combinations of symptom severity in each limb: (a) asymptomatic or mild claudication; (b) severe claudication; (c) critical limb ischemia; (d) below-knee amputation, including transmetatarsal amputations; and (e) above-knee amputation. We did not define a separate health state for asymptomatic patients, because the available data did not allow us to distinguish them from patients with mild claudication.



View larger version (32K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Schematic representation of the decision model structure. The model is a state transition Markov model that defines a number of health states and the possible transitions between those states. The structure and parameter values define how patients may move from one health state to another during one time cycle. Each box in this figure represents different possible events that may lead to such state transitions. For example, the Symptom changes box represents possible changes in symptom severity, with the actual probabilities of these changes depending on whether a patient participates in an exercise program and on graft patency. Where events are modeled separately for each limb, two boxes are shown. Patients participating in an exercise program move from the Start cycle box through the Symptom changes box to the End cycle box. These patients can only move to the Angiography box if invasive treatment (ie, PTA or bypass surgery) is performed because exercise failed to cause alleviation of claudication or if critical limb ischemia develops.

 
Implicit in the definition of treatment failure is the assumption that no or mild claudication does not require treatment, other than general measures for atherosclerotic disease. A threshold maximum walking distance of 250 m—the upper tercile value at presentation in the patient cohort in the Groningen database—was used to distinguish severe claudication from no or mild claudication. In a sensitivity analysis, we used the lower tercile of this distribution, 175 m, as a threshold. Apart from symptom severity, additional details of disease history were modeled. These details included age, sex, duration of symptoms, history of angina pectoris or myocardial infarction, resting ankle-brachial index (lowest value of the two limbs), patient compliance with exercise, number of revascularization procedures performed, time since the last procedure, patency and predictors of failure, and effects on quality of life and costs.

At baseline-case analysis, we calculated the results for a 60-year-old man presenting with a 1-year history of severe unilateral claudication. We assumed that this patient had no history of coronary artery disease (ie, angina pectoris or myocardial infarction) and that the initial ankle-brachial index (lowest value of two limbs) was 0.70. By using first-order Monte Carlo analysis, we simulated the disease history of this patient multiple times (n = 100,000) for each of the five strategies (54). In Monte Carlo analysis, a computer-generated random factor is used at each point in the model where chance plays a role. As a result, outcomes may differ from simulation to simulation, as in real life, where the outcomes of patients with the same characteristics may differ due to chance factors.

The average outcome per strategy was used for further calculations. Strategies were ordered according to increasing effectiveness (in quality-adjusted life years [QALYs]), and a dominated strategy was defined as a strategy with a lower effectiveness and higher cost than another strategy. Next, dominated strategies were eliminated and incremental cost-effectiveness ratios were calculated as the difference in mean lifetime costs divided by the difference in mean QALYs for each strategy compared with that for the next best strategy (55).

Exercise Program
We considered a 6-month exercise program developed at the Department of Internal Medicine of the University Hospital Groningen (41,51). In this program, patients are asked to walk a certain fixed distance each day. The exact distance varies from patient to patient (2–6 km) and depends on his or her performance at baseline. Patients are instructed to pause when symptoms of claudication occur. There are four control visits at the hospital during the first 6 months. After these 6 months, patients continue the exercises at home. We assumed that patients who discontinue the exercises do not re-enter the program.

Revascularization
Revascularization was assumed to be preceded by angiography. Findings at angiography, such as the level of disease and the feasibility of PTA, were incorporated as possible events in the decision model. We assumed that in a minority of patients (5%), the angiographic findings are such that no revascularization will be considered unless the patient develops critical limb ischemia. Of the remaining 95% of patients, those with suitable lesions will undergo PTA.

In the current study, a focal stenosis of 50%–99% above the knee joint was considered suitable for PTA. To make the problem tractable, we incorporated only the most commonly used procedures—specifically, PTA with selective stent placement and aortic bifurcation surgery for suprainguinal disease (14) and PTA and bypass surgery for infrainguinal disease. Thrombolysis was not considered a treatment option. The maximum number of possible revascularization procedures per limb was set at three. More than three interventions per limb were rarely encountered in the Boston database, and in a sensitivity analysis, we varied the number of interventions from two to four.

Quality of Life
QALYs were calculated as the sum of the time spent in each health state multiplied by a correction factor representing the quality of life in each of these health states. Correction factors generally are based on health value measures, which are instruments geared to measure quality of life on a 0–1 scale. Examples of such instruments include the standard gamble and the time trade-off (56). In the present study, we used time trade-off estimates. For the health states of no or mild claudication and severe claudication, responses on the EuroQol questionnaire were modified to estimate time trade-off values (43,49). We used the EuroQol questionnaire because it was the best for discriminating among patients with different symptom severities. Values obtained with other instruments were used in the sensitivity analyses.

For patients with critical limb ischemia or with amputations, we used time trade-off estimates from the medical literature and explored a wide range of alternative values at sensitivity analyses (42). With bilateral symptoms, we assumed that the most severe symptoms would determine the quality of life, ignoring the possible additional effects of milder contralateral symptoms on the quality of life. The average time trade-off among survivors of myocardial infarction was used as an approximation for the quality of life with a systemic complication (44), which was incorporated with the assumption of a multiplicative relationship.

Costs
Both medical and nonmedical costs were included. Medical costs included the costs of all diagnostic and therapeutic procedures, professional services, short- and long-term care after complications, follow-up visits, and rehabilitation and long-term care for patients with amputations. Estimates of the hospital costs for each revascularization procedure were obtained from the Boston database (46). Future medical costs for unrelated diseases were not considered, because none of the proposed treatment strategies offers any substantial survival advantage. If these medical costs had been included, all strategies would have been equally affected and the incremental cost-effectiveness ratio would not have changed.

Nonmedical costs included transportation costs and the opportunity costs of patient time invested in, for example, undergoing a procedure or exercising. At sensitivity analysis, the cost of patient time spent on exercise was set at zero with the assumption that patients enjoy this activity. Costs associated with productivity changes were considered to be negligible and not to differ across strategies. All costs were converted to 1995 U.S. dollars by using the medical care specific consumer price index of the Bureau of Labor Statistics (53). Both costs and QALYs were discounted at an annual rate of 3% (20). The discount rate was varied from 0% to 10% in the sensitivity analysis.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 
Baseline-Case Analysis
At baseline-case analysis, we calculated the outcomes for a 60-year-old man with no history of coronary artery disease. This model predicts a lower probability of severe claudication with increasingly intensive treatment efforts (Fig 2) and, correspondingly, a higher probability of having no or mild claudication (not shown). The difference in the predicted probability of severe claudication between the treatment strategies in which revascularization was considered a primary option and the corresponding strategies in which the same revascularization procedure(s) was considered a secondary option after a trial of exercise (ie, PTA/EX vs EX ± PTA and PTA/BP/EX vs EX ± PTA/BP) initially was large but quickly disappeared over time (Fig 2).



View larger version (34K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Graph illustrates proportions of patients with severe claudication during the first 5 years of follow-up, conditional on survival, at baseline-case analysis. The model predicts a lower probability of severe intermittent claudication for strategies that consist of more invasive treatment options. Initially, the difference between EX ± PTA/BP ({circ}) and PTA/BP/EX ({bullet}) was large because invasive treatment was directly successful in many patients and walking exercises may take some time before leading to a health reward, but over time, the difference disappeared. {square} = EX, {triangleup} = EX ± PTA, {blacktriangleup} = PTA/EX. For an explanation of the strategies, see Table 1.

 
Invasive treatment of patients can be directly successful, whereas patients starting with walking exercises have to invest time before reaping any health reward. With the two strategies that include bypass surgery, the probability of severe claudication decreased sharply during the 1st year and increased only slightly during subsequent years. The two bypass surgery strategies were, however, associated with a substantially higher risk of periprocedural mortality (Fig 3) and morbidity (not shown) than were the other strategies. For the baseline case, the higher periprocedural mortality and morbidity did not substantially affect the unadjusted life expectancy: Without quality adjustment or discounting, estimates ranged from 10.37 years for PTA/BP/EX to 10.40 years for EX ± PTA. Estimates of QALYs varied more substantially, ranging from 6.05 for EX to 6.22 for EX ± PTA/BP (including discounting).



View larger version (34K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Graph illustrates cumulative procedural mortality rates (ie, mortality from angiography, amputation, bypass surgery, or percutaneous intervention) during the first 5 years of follow-up at baseline-case analysis, including mortality from initial invasive treatment (start of follow-up) for PTA/EX ({blacktriangleup}) and PTA/BP/EX ({bullet}). The strategies that included bypass surgery had a higher risk of procedural mortality than did the other strategies. {square} = EX, {triangleup} = EX ± PTA, {circ} = EX ± PTA/BP. For an explanation of the strategies, see Table 1.

 
Generally, the expected lifetime costs also increased with increasing QALYs (Fig 4). EX ± PTA was inferior to PTA/EX by dominance, and PTA/BP/EX was inferior to EX ± PTA/BP by dominance. The differences between the dominant and dominated strategies, however, were minimal. The incremental cost-effectiveness ratios of the two remaining revascularization strategies were $38,000 per QALY gained for PTA/EX (compared with EX) and $311,000 per QALY gained for EX ± PTA/BP (compared with PTA/EX).



View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Graph illustrates expected lifetime costs versus QALYs at baseline-case analysis. With increasing QALYs, the lifetime costs also increased. EX ± PTA ({triangleup}) and PTA/BP/EX ({bullet}) were inferior by dominance. {square} = EX, {circ} = EX ± PTA/BP, {blacktriangleup} = PTA/EX.

 
Sensitivity Analysis
For most parameters, we found that alternative assumptions either did not substantially affect the outcomes or affected all strategies similarly, without substantial effects on the incremental cost-effectiveness ratios. For example, varying the discount rate from 0% to 10% affected both the costs and the effects of all five strategies, but the incremental cost-effectiveness ratios hardly changed. Similarly, a higher or lower risk of mortality associated with peripheral arterial disease did not substantially affect the incremental results.

The results were, however, particularly sensitive to varying health values for severe claudication versus no or mild claudication (Table 2). The smaller the difference in health values between severe claudication and no or mild claudication was, the higher the incremental cost-effectiveness ratios were, whereas a larger difference between the health values resulted in lower incremental cost-effectiveness ratios. Also, the results of strategies that included bypass surgery were sensitive to the quality of life of a patient with systemic complications. For example, if the quality of life with systemic complications was lower, then the incremental cost-effectiveness ratio for EX ± PTA/BP increased. The effects of varying the health values for angina pectoris, above- or below-knee amputation, and critical ischemia were minimal.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Results of Four-Way Sensitivity Analysis Based on Age, History of Coronary Artery Disease, Revascularization Costs, and Health Value Difference between No or Mild Claudication and Severe Claudication

 
For all strategies, the expected lifetime costs were sensitive to the costs of revascularization procedures. Varying these costs between 50% and 150% of the original estimates resulted in incremental cost-effectiveness ratios ranging from $25,000 to $46,000 per QALY gained for PTA/EX and from $266,000 to $453,000 for EX ± PTA/BP. Assuming that the patient enjoyed walking—for which time costs for exercise were set to zero—led to a reduction in the total costs for strategies that started with exercise of approximately $3,500 and thus resulted in incremental cost-effectiveness ratios of $63,000 per QALY gained for PTA/EX and $230,000 per QALY gained for EX ± PTA/BP.

Assuming a small survival benefit to the patients who participated in the exercise program did not affect the results substantially. Using a threshold maximum walking distance of 175 m instead of 250 m to distinguish severe claudication from no or mild claudication increased the incremental cost-effectiveness ratios for PTA/EX ($53,000/QALY gained) and EX ± PTA/BP ($359,000/QALY gained). Assuming that the proportion of lesions suitable for PTA initially would be 50% higher than that during subsequent years decreased the incremental cost-effectiveness ratio for PTA/EX ($31,000/QALY gained) and increased the incremental cost-effectiveness ratio for EX ± PTA/BP ($1,504,000/QALY gained). Alternative assumptions concerning the effect of exercise or bifurcation grafts on the development of contralateral symptoms did not change the results substantially. For all five strategies, the outcomes with alternative assumptions regarding the maximum number of procedures per limb, which varied from two to four, were essentially the same.

For all patient characteristics except age and history of coronary artery disease, the effects on the incremental results were modest. Generally, the incremental cost-effectiveness ratios for the interventional strategies increased with increasing age or with a positive history of coronary artery disease. The latter characteristic was associated with a markedly shorter life expectancy with interventional strategies—especially those that included bypass surgery—owing to the increased procedural risk in patients with cardiac ischemia.

Table 2 summarizes the combined effects of those variables that had substantial effects on the incremental cost-effectiveness ratios for PTA/EX and EX ± PTA/BP at one-way sensitivity analyses: age, history of coronary artery disease, costs of revascularization, and health value difference between no or mild claudication and severe claudication. Each cell in Table 2 is based on simulations of 100,000 disease histories for each treatment strategy under the given parameter values. For example, with the assumption of a patient aged 40 years with a positive history of coronary artery disease, revascularization costs that are 50% lower than the baseline-case estimates, and a health value difference of 0.08 between no or mild claudication versus severe claudication, our analysis results indicate that the incremental cost-effectiveness ratio for PTA/EX was approximately $36,000 per QALY gained, whereas EX ± PTA/BP was dominated by the other two strategies.

The data in Table 2 suggest that EX ± PTA/BP is unattractive under most circumstances, especially for patients aged 80 years and those with a history of coronary artery disease, because it is dominated by EX or PTA/EX or because the incremental cost-effectiveness ratios are exceptionally high.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
APPENDIX
 REFERENCES
 
The development of percutaneous procedures for treatment of peripheral arterial disease during the past 2 decades has stirred up the debate about the optimal treatment strategy for patients with intermittent claudication (1,2,12,57–59). In this study, we addressed the question of whether the traditional conservative approach to treating claudication can still be justified in the present era of low-risk revascularization procedures. The results suggest that, from a societal perspective and with consideration of cost-effectiveness, a fairly conservative approach is warranted. The net effect on the quality-adjusted life expectancy with revascularization was relatively small, not only because of the modest difference in health values between patients with no or mild claudication and those with severe claudication, but also because the benefits of revascularization in terms of symptom severity were partly offset by the risks of procedural mortality and morbidity, especially for bypass surgery.

The expected lifetime costs of revascularization procedures, especially those strategies including bypass surgery, were substantially higher than those of exercise therapy alone. Furthermore, the results of strategies with a revascularization procedure as a possible initial step (ie, PTA/EX and PTA/BP/EX) were very similar to those of the corresponding strategies that considered the same procedure(s) as a secondary option after a trial of exercise (ie, EX ± PTA and EX ± PTA/BP, respectively).

The results of the sensitivity analysis emphasize that the expected gain in quality of life should be a crucial factor in the choice of treatment for intermittent claudication. We found that the cost-effectiveness ratios of the revascularization strategies depended to a large extent on the health values for no or mild claudication and severe claudication. The quality of life values for no or mild claudication and severe claudication in the current analysis were based on the results of a previous study (43), which were in close agreement with the values obtained in other studies (47,60).

The difference in the average observed EuroQol values (which were used in our analysis) for patients with no or mild claudication versus those with severe claudication was only modest (0.08) and led to high cost-effectiveness ratios for the revascularization strategies. With other quality-of-life instruments, differences between groups were even smaller and would have resulted in even higher cost-effectiveness ratios. The estimated health values were, however, average values, assuming that patients could be divided into fairly homogeneous groups on the basis of their walking distance. The effect of a change in walking distance from, for example, 300 to 150 m may for some patients have a far greater effect on their quality of life than the average effect on quality of life that we estimated. Thus, although our study results suggest that bypass surgery is not cost-effective, given the average observed difference in quality of life values between no or mild claudication and severe claudication, the sensitivity analysis results support the use of bypass surgery in exceptional cases where this difference is very large.

With constant improvements in percutaneous treatment techniques, more lesions will be suitable for PTA. Under these circumstances, our model predicted that the incremental cost-effectiveness ratio of the strategy with angioplasty as the only and the initial invasive treatment would decrease slightly and the incremental cost-effectiveness ratio of the strategy with bypass surgery would increase tremendously. We also assumed that invasive treatment was always preceded by angiography. Currently, angiography is often replaced by noninvasive imaging modalities such as magnetic resonance angiography and duplex ultrasonography. These modalities involve lower costs and risks than does angiography, but they can lead to false test results.

The cost-effectiveness ratio of more than $200,000 per QALY gained is outside the generally reported range of cost-effectiveness ratios (61); this observation implies that bypass surgery for intermittent claudication is an inefficient use of the limited resources in health care. Although practice patterns vary considerably, in many centers, bypass surgery is performed not only for limb-threatening symptoms but also for severe claudication. Our analysis results suggest that bypass surgery, even as only a secondary option, is very expensive relative to the achieved gain in effectiveness.

The incremental cost-effectiveness ratios for angioplasty were within the range of those reported for currently accepted technologies. For example, Wong et al (62) observed the cost of coronary artery bypass surgery for three-vessel disease in patients with severe angina to be $105,000 per QALY gained (adjusted to 1995 U.S. dollars). Compared with the cost-effectiveness ratios for coronary angioplasty for severe angina, $10,000–$18,000 per QALY gained (62), however, the cost-effectiveness ratios for peripheral angioplasty were somewhat unfavorable.

Our study was a synthesis of the existing knowledge of the risks, benefits, and costs of different therapeutic options for patients with peripheral arterial disease. The limitations of our study reflect shortcomings in the current knowledge and the necessity to make several simplifying modeling assumptions to keep the problem tractable. To test the robustness of our conclusions, we examined, where possible, the effect of choosing alternative assumptions about the outcomes of our study. For many assumptions, we observed that the alternatives either did not change the results substantially or changed the results for all strategies similarly; this finding indicates that the conclusions remained the same. However, a number of important points of discussion remain.

The primary limitation of our study was that differences among the various sources of evidence that we used may have biased the results. For example, we used two different data sets to model changes in the severity of claudication: one for patients participating in an exercise program and one for patients who underwent a revascularization. We adjusted the transition probabilities from both data sets for important potential confounders, including age, ankle-brachial index, presence of angina, and duration of the symptoms. Similarly, we corrected other parameter estimates for potential confounders, where possible. Thus, formal decision analysis offers opportunities to truly integrate information from different sources as an alternative to just comparing the outcomes of different studies. Nevertheless, in terms of preventing confounders, our analysis cannot replace a properly conducted randomized controlled trial. In our view, the current analysis is a prelude, rather than true alternative, to such a future trial, and the presented results may help focus the question-and-trial design.

Another limitation related to cost-effectiveness analysis involving decision models is the use of fixed values for the parameters in the model with the assumption that all of these values are precisely known. For example, in our model, we incorporated that 8.3% of patients undergoing aortic bifurcation bypass surgery would have systemic complications. This value was uncertain, however, and could have been 7% or 10%. Thus, there was uncertainty not only as to whether an event would occur, which was modeled with probabilities, but also surrounding the value of all the model parameters. We explored the effect of the uncertainty in the model parameters on the decision by performing sensitivity analysis, at which all estimates were varied across a plausible range of values. A sensitivity analysis varying up to four parameters simultaneously provided insight into the uncertainty surrounding the results.

Furthermore, we could have assessed the uncertainty in the model by performing a probabilistic sensitivity study with second-order Monte Carlo analysis (63,64). In a second-order Monte Carlo analysis, simulations are performed with a set of parameter estimates drawn from prespecified distributions, and this process is repeated multiple times, with a new set of parameter estimates drawn from the distributions each time. The results provide a distribution of the effectiveness and cost outcomes. Second-order Monte Carlo analysis, although useful at times, increases the complexity of the methods and results and does not provide more insight into which parameters are driving the decision. Moreover, the decision of which treatment strategy to use in clinical practice still would need to be made, and according to welfare economics, would still be best made on the basis of the incremental cost-effectiveness ratios calculated by using the means of the parameter estimates (65).

We recognize that our definitions of health states, especially those of no or mild claudication and severe claudication, are simplifications of more complex clinical realities. Most recommendations regarding revascularization procedures for intermittent claudication distinguish between disabling and nondisabling claudications, categories that take into account a patient’s lifestyle and occupation. Such subjective definitions, however, would make it difficult, if not impossible, to obtain the relevant data and to interpret the results. Thus, we used maximum walking distance to define no or mild claudication versus severe claudication. The threshold value that we used, 250 m, was necessarily an arbitrary one: This distance is in the upper tercile of the distribution of maximum walking distance at baseline for patients with intermittent claudication in the Groningen database. Some vascular surgeons would find 250 m too lenient a criterion and use lower thresholds to define severe claudication.

In a sensitivity analysis, we recalculated the transition probabilities in the model and the health values for no or mild claudication and severe claudication on the basis of a lower cutoff value (175 m, the lower tercile). The incremental cost-effectiveness ratios of the revascularization strategies increased with the lower threshold value, but the conclusions remained the same.

Another limitation of our analysis may be that we considered only one conservative treatment option. Exercise is generally regarded as the best conservative option. The exercise program that we considered in the current analysis is intermediate between more intensely supervised programs and a pure home-based exercise program. There are conflicting reports regarding the additional effect of supervision in exercise programs, and further research on a larger scale is required (66,67). There is little doubt, however, that the more intensely supervised programs are more expensive than is our program.

In addition to recommending exercise, many physicians would advise their patients to change their smoking habits. Cigarette smoking has been shown to increase the incidence of intermittent claudication (68), and smoking cessation may reduce the incidence of adverse events once intermittent claudication has developed (7,69). In many exercise programs, including ours, patients are motivated to stop smoking; thus, the effects of exercise and smoking cessation may overlap. We, therefore, chose not to include smoking cessation as a separate option. Of the drugs that have been marketed for intermittent claudication, pentoxifylline has probably been studied most widely; however, the results of a meta-analysis of 12 randomized trials could not provide conclusive evidence of this drug’s effectiveness (7). Finally, we did not include platelet aggregation inhibitor therapy as a separate treatment option, but rather we considered this to be adjuvant therapy, the main benefit being the reduction of coronary and cerebrovascular events, which was not the focus of the current analysis.

Finally, we found that in the setting of coronary artery disease, revascularization for claudication is not cost-effective. We did not, however, model the option of coronary revascularization prior to peripheral revascularization, because this issue is different from that which we set out to address and was beyond the scope of this article. However, we did consider the presence of coronary artery disease and found that due to the increased procedural risk in patients with cardiac ischemia, the benefits of the interventional strategies—especially those that included bypass surgery—compared with those of the conservative strategies, were substantially reduced.

In conclusion, the results suggest that, on average, the small gain in effectiveness achieved with bypass surgery for intermittent claudication does not justify the additional costs. Angioplasty as an alternative to exercise, when feasible, was more effective than was exercise alone, and the cost-effectiveness ratio was within the generally accepted range.


   APPENDIX
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 APPENDIX
REFERENCES
 
Model Structure
Transitions were modeled back and forth between no or mild claudication and severe claudication and between claudication and critical ischemia. We assumed that exercise does not reduce the risk of critical ischemia (50), that critical limb ischemia would not improve without revascularization, and that critical limb ischemia in a revascularized limb would develop only after loss of patency. Patients who developed critical limb ischemia were assumed to have undergone a revascularization procedure, unless the maximum number of procedures had been reached, in which case, amputation would follow. Amputation in patients with critical limb ischemia and progression from below-knee to above-knee amputation also were modeled.

Patency, symptom severity, and other details of the disease history were tracked for each limb separately. In comparing the results of limb- and patient-based reports of patency for bifurcation grafts, we found no evidence suggesting that patency in one limb depends on patency in the contralateral limb (26). Therefore, we assumed independent patency and failure in the two limbs. Patient characteristics, however, affected symptom severity in both limbs, and, thus, the events in the two limbs were related (39,41). Implantation of a bifurcation graft for unilateral symptoms was assumed not to affect the development of contralateral symptoms, and this assumption was examined in a sensitivity analysis. Exercise was assumed not to affect the development of contralateral symptoms, and at sensitivity analysis, we tested whether assuming a preventive effect of exercise (ranging from 0% to 100%) would substantially change the results.

The degree of clinical detail incorporated in our decision model exceeded the capacity of standard decision analysis software. Therefore, a C++ based programming language (Fast Decision Language; J. A. de Vries, Groningen, the Netherlands) was developed to construct the model. At the initial stage, for debugging purposes, a simpler version of the model was constructed in both Fast Decision Language and DATA (Decision Analysis by TreeAge; TreeAge Software, Williamstown, Mass) and the results were compared. Extensive sensitivity analyses were performed to check for inconsistencies in the model.

Exercise Program
Estimates of the transition probabilities between the health states associated with exercise were obtained from the Groningen database and adjusted for age, duration of symptoms, ankle-brachial index, presence of angina, and compliance (51). On the basis of 3-year follow-up at the University Hospital Groningen, we assumed that the rate of discontinuing the exercises was constant over time. Although no survival benefit following exercise programs for intermittent claudication has been demonstrated (52), in a sensitivity analysis, we tested whether a small survival benefit (ie, 20% reduction in mortality) among participants of the exercise program would substantially affect the results.

Revascularization
Patency estimates were obtained from three published meta-analyses (26,27,34) and adjusted for presenting symptoms (claudication vs critical limb ischemia) and, where applicable, lesion type (stenosis vs occlusion), level of distal anastomosis (above vs below the knee), and graft material (ie, autologous vein vs polytetrafluoroethylene). For iliac PTA, we used data from studies in which PTA was combined with stent placement, assuming that a stent would be placed if angioplasty alone yielded suboptimal results (14). We assumed that patients would not undergo multiple revascularization procedures at different arterial levels simultaneously, but the model does allow for new procedures in the same limb at a different (usually infrainguinal) arterial level during follow-up. In following up multiple procedures in one limb, we assumed that the procedure that was performed last would determine the overall probability of graft or PTA failure.

Transition probabilities between health states following revascularization were obtained from the Rotterdam database and adjusted for covariates and the effects of anticoagulant medication. In addition, we incorporated the effect of failure on changes in maximum walking distance by using a decrease in ankle-brachial index of more than 0.15 relative to the postrevascularization value as a patency criterion (39,40). Estimates of the rate of development of contralateral symptoms were obtained from the Boston database (36).

Costs
The main cost of the exercise program was the opportunity cost of the time invested by the patient (20). Questionnaire answers indicated that the participants at the University Hospital Groningen spent, on average, 6.4 hours per week exercising. Theoretically, the value of patient time depends on how much the patient enjoys or dislikes the activity (20,50). At baseline-case analysis, we used the 1995 U.S. average hourly wage rate of $11.35 (from Bureau of Labor Statistics [53]) as the value of patient time for all disease-related activities. For in-hospital vascular procedures, the value of patient time was estimated as the average length of stay (from the Boston database), in days, multiplied by 7.5 times the hourly wage rate. We also included transportation costs based on an average distance from the home address to the hospital of 49 km.

Estimates of the hospital costs for each of the revascularization procedures were obtained from the Boston database (46) and adjusted for age, sex, presenting symptoms, and history of coronary artery disease. More detailed information on the cost estimation can be found in the article by Jansen et al (46). The extra costs incurred from systemic complications or procedural mortality also were obtained from these data. We found that local complications had a minimal effect on the total hospital costs. We assumed that the extra long-term medical costs for patients with systemic complications would equal the yearly medical costs for survivors of myocardial infarction (48). Myocardial infarction is the most important major complication associated with invasive treatment of peripheral arterial disease (70). The long-term costs from systemic complications were varied over a wide range to explore the effects of major complications other than myocardial infarction.

All of the data used in the decision model are presented in Tables A1 and A2. The rates and probabilities are presented in Table A1, and the health values and costs are presented in Table A2.


View this table:
[in this window]
[in a new window]

  		TABLE A1. Rates and Probabilities

 

View this table:
[in this window]
[in a new window]

  		TABLE A2. Health Values and Costs

 


   FOOTNOTES
 
Abbreviations: PTA = percutaneous transluminal angioplasty, QALY = quality-adjusted life year

Author contributions: Guarantor of integrity of entire study, M.G.M.H.; study concepts, S.O.d.V., J.B.W., M.C.D., M.G.M.H.; study design, S.O.d.V., J.B.W., M.G.M.H.; literature research, S.O.d.V.; clinical studies, S.O.d.V., M.C.D.; data acquisition, S.O.d.V.; data analysis/interpretation, S.O.d.V., J.A.d.V., K.V., M.G.M.H.; statistical analysis, S.O.d.V., J.A.d.V., K.V.; manuscript preparation, S.O.d.V.; manuscript definition of intellectual content, S.O.d.V., J.B.W., M.C.D., K.V., M.G.M.H.; manuscript editing, S.O.d.V., M.G.M.H.; manuscript revision/review, S.O.d.V., J.B.W., M.C.D., K.V., M.G.M.H.; manuscript final version approval, all authors.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 

  1. Coffman JD. Intermittent claudication: be conservative. N Engl J Med 1991; 325:577-578.
  2. Hertzer NR. The natural history of peripheral vascular disease: implications for its management. Circulation 1991; 83(2 suppl):I12-I19.
  3. Bloor K. Natural history of arteriosclerosis of the lower extremities. Ann R Coll Surg Engl 1961; 28:36-52.
  4. Cronenwett JL, Warner KG, Zelenock GB, et al. Intermittent claudication: current results of nonoperative management. Arch Surg 1984; 119:430-436.
  5. Rosenbloom MS, Flanigan DP, Schuler JJ, et al. Risk factors affecting the natural history of intermittent claudication. Arch Surg 1988; 123:867-870.
  6. Taylor LM, Porter JM. Natural history and nonoperative treatment of chronic lower extremity ischemia. In: Rutherford RB, eds. Vascular surgery. Philadelphia, Pa: Saunders, 1995; 751-766.
  7. Radack K, Wyderski RJ. Conservative management of intermittent claudication. Ann Int Med 1990; 113:135-146.
  8. Hiatt WR, Regensteiner JG, Hargarten ME, Wolfel EE, Brass EP. Benefit of exercise conditioning for patients with peripheral arterial disease. Circulation 1990; 81:602-669.
  9. Gardner AW, Poehlman ET. Exercise rehabilitation programs for the treatment of claudication pain: a meta-analysis. JAMA 1995; 274:975-980.
  10. Carter SA, Hamel ER, Paterson JM, Snow CJ, Mymin D. Walking ability and ankle systolic pressures: observations in patients with intermittent claudication in a short-term walking exercise program. J Vasc Surg 1989; 10:642-649.
  11. Ekroth R, Dahllof AG, Gundevall B, Holm J, Schersten T. Physical training of patients with intermittent claudication: indications, methods, and results. Surgery 1978; 84:640-663.
  12. Isner JM, Rosenfield K. Redefining the treatment of peripheral artery disease: role of percutaneous revascularization. Circulation 1993; 88:1534-1557.
  13. Friedman SG, Lazzaro RS, Spier LN, Moccio C, Tortolani AJ. A prospective randomized comparison of Dacron and polytetrafluoroethylene aortic bifurcation grafts. Surgery 1995; 117:7-10.
  14. Tetteroo E, van der Graaf Y, Bosch JL, et al. Randomised comparison of primary stent placement versus primary angioplasty followed by selective stent placement in patients with iliac-artery occlusive disease: Dutch Iliac Stent Trial Study Group. Lancet 1998; 351:1153-1159.
  15. Wilson SE, Wolf GL, Cross AP. Percutaneous transluminal angioplasty versus operation for peripheral arteriosclerosis: report of a prospective randomized trial in a selected group of patients. J Vasc Surg 1989; 9:1-9.
  16. Wolf GL, Wilson SE, Cross AP, Deupree RH, Stason WB. Surgery or balloon angioplasty for peripheral vascular disease: a randomized clinical trial—principal investigators and their Associates of Veterans Administration Cooperative Study Number 199. J Vasc Interv Radiol 1993; 4:639-648.
  17. Lundgren F, Dahllof AG, Lundholm K, Schersten T, Volkmann R. Intermittent claudication: surgical reconstruction or physical training?. Ann Surg 1989; 209:346-355.
  18. Perkins JM, Collin J, Creasy TS, Fletcher EW, Morris PJ. Exercise training versus angioplasty for stable claudication: long and medium term results of a prospective, randomised trial. Eur J Vasc Endovasc Surg 1996; 11:409-413.
  19. Creasy TS, McMillan PJ, Fletcher EWL, Collin J, Morris PJ. Is percutaneous transluminal angioplasty better than exercise for claudication? Preliminary results from a prospective randomised trial. Eur J Vasc Surg 1990; 4:135-140.
  20. Gold MR, Siegel JE, Russell LB, Weinstein MC. Cost-effectiveness in health and medicine New York, NY: Oxford University Press, 1996.
  21. de Smit P. Orale anticoagulantia en perifeer vaatlijden: dissertation Rotterdam, the Netherlands: Erasmus University, 1994.
  22. Jelnes R, Gaardsting O, Hougaard Jensen K, Baekgaard N, Tonnesen KH, Schroeder T. Fate in intermittent claudication: outcome and risk factors. Br Med J (Clin Res Ed) 1986; 293:1137-1140.
  23. Juergens JL, Barker NW, Hines EA. Arteriosclerosis obliterans: review of 520 cases with special reference to pathogenic and prognostic factors. Circulation 1960; 21:188-195.
  24. Kallero KS. Mortality and morbidity in patients with intermittent claudication as defined by venous occlusion plethysmography: a ten-year follow-up study. J Chron Dis 1981; 34:455-462.
  25. Begg TB, Richards RL. The prognosis of intermittent claudication. Scot Med J 1962; 7:341-352.
  26. de Vries SO, Hunink MG. Results of aortic bifurcation grafts for aortoiliac occlusive disease: a meta-analysis. J Vasc Surg 1997; 26:558-569.
  27. Bosch JL, Hunink MG. Meta-analysis of the results of percutaneous transluminal angioplasty and stent placement for aortoiliac occlusive disease. Radiology 1997; 204:87-96.
  28. Hunink MGM, Wong JB, Donaldson MC, Meyerovitz MF, de Vries JA, Harrington DP. Revascularization for femoropopliteal disease: a decision and cost-effectiveness analysis. JAMA 1995; 274:165-171.
  29. Hessel SJ, Adams DF, Abrams HL. Complications of angiography. Radiology 1981; 138:273-281.
  30. Waugh JR, Sacharias N. Arteriographic complications in the DSA era. Radiology 1992; 182:243-246.
  31. Plecha FR, Bertin VJ, Plech EJ, et al. The early results of vascular surgery in patients 75 years of age and older: an analysis of 3,259 cases. J Vasc Surg 1985; 2:769-774.
  32. Kandarpa K, Gardiner GAJ. Angiography. In: Kandarpa K, Aruny JE, eds. Handbook of interventional radiological procedures. Little, Brown: Boston, Mass, 1995; 21.
  33. Weiss GN, Gorton A, Read RC, Neal LA. Outcomes of lower extremity amputations. J Am Geriatr Soc 1990; 38:877-883.
  34. Hunink MGM, Wong JB, Donaldson MC, Meyerovitz MF, Harrington DP. Patency results of percutaneous and surgical revascularization for femoropopliteal arterial disease. Med Decis Making 1994; 14:71-81.
  35. Jonason T, Ringqvist I. Factors of prognostic importance for subsequent rest pain in patients with intermittent claudication. Acta Med Scand 1985; 218:27-33.
  36. de Vries SO, Donaldson MC, Hunink MG. Contralateral symptoms after unilateral intervention for peripheral occlusive disease. J Vasc Surg 1998; 27:414-421.
  37. Hunink MGM, Donaldson MC, Meyerovitz MF, et al. Risks and benefits of femoropopliteal percutaneous balloon angioplasty. J Vasc Surg 1993; 17:183-194.
  38. Mannick JA, Whittemore AD, Donaldson MC. Clinical and anatomic considerations for surgery in tibial disease and the results of surgery. Circulation 1991; 83(2 suppl):I81-I85.
  39. de Vries SO, de Smit P, Bossuyt PMM. Peripheral arterial occlusive disease: from a decrease in ankle-brachial index to changes in symptom severity. In: de Vries SO, eds. Management strategies for intermittent claudication: dissertation. Groningen, the Netherlands: University of Groningen, 1998; 52-60.
  40. Rutherford RB. Standards for evaluating results of interventional therapy for peripheral vascular disease. Circulation 1991; 83(2 suppl):I6-I11.
  41. de Vries SO, Fidler V, Kuipers WD, Hunink MG. Fitting multistate transition models with autoregressive logistic regression: supervised exercise in intermittent claudication. Med Decis Making 1998; 18:52-60.
  42. Sculpher M, Michaels J, McKenna M, Minor J. A cost-utility analysis of laser-assisted angioplasty for peripheral arterial occlusions. Int J Technol Assess Health Care 1996; 12:104-125.
  43. de Vries SO, Kuipers WD, Hunink MG. Intermittent claudication: symptom severity versus health values. J Vasc Surg 1998; 27:422-430.
  44. Tsevat J, Goldman L, Lamas GA, et al. Functional status versus utilities in survivors of myocardial infarction. Med Care 1991; 29:1153-1159.
  45. Gupta SK, Veith FJ, Ascer E, White Flores SA, Gliedman ML. Cost factors in limb-threatening ischaemia due to infrainguinal arteriosclerosis. Eur J Vasc Surg 1988; 2:151-154.
  46. Jansen RM, de Vries SO, Cullen KA, Donaldson MC, Hunink MG. Cost-identification analysis of revascularization procedures on patients with peripheral arterial occlusive disease. J Vasc Surg 1998; 28:617-623.
  47. Bosch JL, Tetteroo E, Mali WP, Hunink MG. Iliac arterial occlusive disease: cost-effectiveness analysis of stent placement versus percutaneous transluminal angioplasty—Dutch Iliac Stent Trial Study Group. Radiology 1998; 208:641-648.
  48. Wittels EH, Hay JW, Gotto AM. Medical costs of coronary artery disease in the United States. Am J Cardiol 1990; 65:432-440.
  49. Busschbach JJ, McDonnell J, Essink-Bot ML, van Hout BA. Estimating parametric relationships between health description and health valuation with an application to the EuroQol EQ-5D. J Health Econ 1999; 18:551-571.
  50. Hatziandreu EI, Koplan JP, Weinstein MC, Caspersen CJ, Warner KE. A cost-effectiveness analysis of exercise as a health promotion activity. Am J Public Health 1988; 78:1417-1421.
  51. Andriessen MP, Barendsen GJ, Wouda AA, de Pater L. Changes of walking distance in patients with intermittent claudication during six months intensive physical training. VASA 1989; 18:63-68.
  52. Jonason T, Ringqvist I. Mortality and morbidity in patients with intermittent claudication in relation to the location of the occlusive atherosclerosis in the leg. Angiology 1985; 36:310-314.
  53. Luce BR, Manning WG, Siegel JE, Lipscomb J. Estimating costs in cost-effectiveness analysis. In: Gold MR, Siegel JE, Russell LB, et al., eds. Cost-effectiveness in health and medicine. New York, NY: Oxford University Press, 1996; 176-213.
  54. Sonnenberg FA, Beck JR. Markov models in medical decision making: a practical guide. Med Decis Making 1993; 13:322-338.
  55. Weinstein MC, Stason WB. Foundations of cost-effectiveness analysis for health and medical practices. N Engl J Med 1977; 296:716-721.
  56. Torrance GW. Measurement of health state utilities for economic appraisal: a review. J Health Econ 1986; 5:1-30.
  57. Doubilet PM, Abrams HL. The cost of underutilization: percutaneous transluminal angioplasty for peripheral vascular disease. N Engl J Med 1984; 310:95-102.
  58. Hunink MGM, Meyerovitz MF. The use of angioplasty, bypass surgery, and amputation in the management of peripheral vascular disease. N Engl J Med 1992; 326:414.
  59. Tunis SR, Bass EB, Steinberg EP. The use of angioplasty, bypass surgery, and amputation in the management of peripheral vascular disease. N Engl J Med 1991; 325:556-562.
  60. Bosch JL, Hunink MG. The relationship between descriptive and valuational quality-of-life measures in patients with intermittent claudication. Med Decis Making 1996; 16:217-225.
  61. Graham JD, Corso PS, Morris JM, Segui-Gomez M, Weinstein MC. Evaluating the cost-effectiveness of clinical and public health measures. Annu Rev Public Health 1998; 19:125-152.
  62. Wong JB, Sonnenberg FA, Salem DN, Pauker SG. Myocardial revascularization for chronic stable angina: analysis of the role of percutaneous transluminal coronary angioplasty based on data available in 1989. Ann Intern Med 1990; 113:852-871.
  63. Halpern EF, Weinstein MC, Hunink MGM, Gazelle GS. Representing both first- and second-order uncertainties by Monte Carlo simulation for groups of patients. Med Decis Making 2000; 20:314-322.
  64. Hunink MGM, Bult JR, de Vries J, Weinstein MC. Uncertainty in decision models analyzing cost-effectiveness: the joint distribution of incremental costs and effectiveness evaluated with a nonparametric bootstrap method. Med Decis Making 1998; 18:337-346.
  65. Garber AM, Weinstein MC, Torrance GW, Kamlet MS. Theoretical foundations of cost-effectiveness analysis. In: Gold MR, Siegel JE, Russell LB, et al., eds. Cost-effectiveness in health and medicine. New York, NY: Oxford University Press, 1996; 25-53.
  66. Jonason T, Ringqvist I, Oman-Rydberg A. Home-training of patients with intermittent claudication. Scand J Rehabil Med 1981; 13:137-141.
  67. Regensteiner JG, Meyer TJ, Krupski WC, Cranford LS, Hiatt WR. Hospital vs home-based exercise rehabilitation for patients with peripheral arterial occlusive disease. Angiology 1997; 48:291-300.
  68. Kannel WB, McGee DL. Update on some epidemiologic features of intermittent claudication: the Framingham Study. J Am Geriatr Soc 1985; 33:13-18.
  69. Hughson WG, Mann JI, Tibbs DJ, Woods HF, Walton I. Intermittent claudication: factors determining outcome. Br Med J 1978; 1:1377-1379.
  70. Szilagyi DE, Elliott JP, Smith RF, Reddy DJ, McPharlin M. A thirty-year survey of the reconstructive surgical treatment of aortoiliac occlusive disease. J Vasc Surg 1986; 3:421-436.



This article has been cited by other articles:


Home page
 CirculationHome page
C. J. White and W. A. Gray
Endovascular Therapies for Peripheral Arterial Disease: An Evidence-Based Review
Circulation, November 6, 2007; 116(19): 2203 - 2215.
[Abstract] [Full Text] [PDF]

Home page
 Chronic IllnessHome page
L. Gordon, N. Graves, A. Hawkes, and E. Eakin
A review of the cost-effectiveness of face-to-face behavioural interventions for smoking, physical activity, diet and alcohol
Chronic Illness, June 1, 2007; 3(2): 101 - 129.
[Abstract] [PDF]

Home page
 NEJMHome page
C. White
Intermittent Claudication
N. Engl. J. Med., March 22, 2007; 356(12): 1241 - 1250.
[Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
H. A. Deutschmann, H. Schoellnast, W. Temmel, M. Deutschmann, G. Schwantzer, G. A. Fritz, M. Brodmann, and K. A. Hausegger
Endoluminal Therapy in Patients with Peripheral Arterial Disease: Prospective Assessment of Quality of Life in 190 Patients
Am. J. Roentgenol., January 1, 2007; 188(1): 169 - 175.
[Abstract] [Full Text] [PDF]

Home page
 Chronic IllnessHome page
N. Graves, L. Mckinnon, M. Reeves, P. Scuffham, L. Gordon, and E. Eakin
Cost-effectiveness analyses and modelling the lifetime costs and benefits of health-behaviour interventions
Chronic Illness, June 1, 2006; 2(2): 97 - 107.
[Abstract] [PDF]

Home page
 RadiologyHome page
S. Spronk, J. L. Bosch, H. F. Veen, P. T. den Hoed, and M. G. M. Hunink
Intermittent Claudication: Functional Capacity and Quality of Life after Exercise Training or Percutaneous Transluminal Angioplasty--Systematic Review
Radiology, June 1, 2005; 235(3): 833 - 842.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
M. A. Mauro
The Battle of Intimal Hyperplasia in the War against Femoropopliteal Disease
Radiology, May 1, 2004; 231(2): 299 - 301.
[Full Text] [PDF]

Home page
 RadiologyHome page
K. Visser, M. C. J. M. Kock, K. M. Kuntz, M. C. Donaldson, G. S. Gazelle, and M. G. Myriam Hunink
Cost-effectiveness Targets for Multi-Detector Row CT Angiography in the Work-up of Patients with Intermittent Claudication
Radiology,