Jump to
Abstract
Background Long-term treatment with aspirin is recommended in patients with large-vessel peripheral arterial disease since these patients have a high risk of death from cardiovascular causes. Recent studies have demonstrated the prophylactic effect of low-dose aspirin in reducing the risk of cardiovascular events. Since aspirin is also recommended for prevention of late recurrence after peripheral angioplasty, the present study was undertaken to compare the effects of high-dose (1000 mg/d) and low-dose (100 mg/d) aspirin on long-term patency after femoropopliteal angioplasty.
Methods and Results Two hundred sixteen patients treated successfully by percutaneous transluminal angioplasty for femoropopliteal lesions were randomly allocated to therapy with either 1000 or 100 mg aspirin daily. The follow-up was 24 months. The long-term results were analyzed using the Kaplan-Meier method, and differences between curves of cumulative patency were determined with the Wilcoxon and log-rank statistics. Complete follow-up information (patency after 24 months, restenosis, and death) was obtained in 207 patients. During the 2-year follow-up period, 72 patients—36 in the high-dose and 36 in the low-dose aspirin group, respectively—developed angiographically verified reobstruction within the recanalized segment. By intention-to-treat analysis, the cumulative patency rates at 24 months were 62.5% in the high-dose and 62.6% in the low-dose aspirin group (Wilcoxon, P=.97; log-rank, P=.97). The cumulative survival at 24 months of follow-up was 86.6% in the high-dose and 87.7% in the low-dose aspirin group. The number of patients discontinuing therapy was 30 in the high-dose and 11 in the low-dose aspirin group (P<.01). Fewer patients receiving 100 mg of aspirin discontinued therapy because of gastrointestinal symptoms (4 versus 20).
Conclusions The data indicate that 100 mg aspirin is no less effective in the prevention of restenosis after femoropopliteal PTA than a 1000-mg dose and has fewer side effects.
Patients with large-vessel peripheral arterial disease have a high risk of death from cardiovascular causes.1 2 They are particularly likely to have concomitant coronary and/or cerebrovascular disease. Antiplatelet therapy with acetylsalicylic acid (aspirin) is well established as secondary prophylaxis in patients with arterial thrombotic disorders.3 4 Recent studies have also demonstrated the prophylactic effect of low-dose aspirin in reducing the risk of future cardiovascular events in a variety of clinical settings.5 6 7 8 9 10 11 Several studies compared the effects of different doses of aspirin on thrombotic events, and there is no evidence that low doses are either more or less effective than high doses.7 9 12 There is evidence, however, that low doses produce fewer gastric side effects.7 9 12
Percutaneous transluminal angioplasty (PTA) is a well-established method of recanalizing chronic arterial obstructions, and it has gained increased acceptance as a safe and effective therapy for peripheral arterial occlusive disease. Late restenosis constitutes the most important problem after successful PTA. Many studies using a large variety of pharmacological interventions have been done in patients after coronary angioplasty,13 whereas such studies concerning prevention of late recurrence after peripheral angioplasty are lacking. Only recently, Heiss et al14 —in a comparison of two different aspirin dosages (990 versus 300 mg/d) in combination with 225 mg dipyridamole with placebo during a follow-up period of 6 months after femoropopliteal PTA—reported a significant difference between the placebo and the high-dose aspirin group (recurrence rate of 63% versus 39%).
Because patients with peripheral arterial diseases often need long-term treatment with aspirin for concomitant cardiovascular diseases, the present study was undertaken to compare the effects of high-dose (1000 mg/d) and low-dose (100 mg/d) aspirin on long-term patency (24 months) after femoropopliteal PTA.
Methods
Patients
Between January 1, 1987, and June 30, 1989, a total of 320 consecutive patients who were treated in our center with PTA for femoropopliteal lesions were screened for entry into the study. The patients had to have a history of claudication for more than 3 months or pain at rest with or without tissue damage. All of the patients included in the study had adequate inflow in the aortoiliac vessels documented by angiography. Exclusion criteria included failed PTA (n=28; if the obstructive lesion could not be traversed or if the angiographically documented final stenosis remained >50%), early recurrence within 3 days (n=16), complications requiring surgical intervention (n=6), recent gastroduodenal ulcer (n=25; gastroduodenoscopy had been done in all patients before PTA; patients with gastroduodenal ulcer in the history but with no documented ulcer during gastroduodenoscopy were included in the study), treatment for recurrent stenosis after previous PTA (n=30), serious associated disease with life expectancy of <2 years (n=5), severe renal insufficiency (n=3), need for regular therapy with nonsteroidal antiphlogistic drugs (n=10), need for long-term treatment with anticoagulant drugs (n=14), and full cooperation considered unlikely or patient refusal to participate in the study (n=14). In some patients, there was more than one reason to exclude them from entry into the study.
According to these criteria, 216 patients (121 men and 95 women; median age, 66 years; age range, 34 to 80 years) were included in the study. The patients were randomly allocated to therapy with either 1000 mg/d acetylsalicylic acid (aspirin, Colfarit twice daily; Bayer) or 100 mg/d aspirin (Kinderaspirin; Bayer). The patients were assigned to either group through adaptive randomization.15 Prognostic factors that were accounted for included sex, age (<65 years or >65 years), history of diabetes, smoking habits, cholesterol (>250 or <250 mg/dL), clinical staging (claudication versus rest pain or tissue damage), runoff vessel disease (runoff status was defined according to the number of patent infrapopliteal arteries—poor indicates no or only one patent and good indicates two or three patent arteries), and femoropopliteal lesion (stenosis versus occlusion ≤10 cm in length versus occlusion >10 cm; stenoses were not stratified according to length because when the study was planned, there were no data concerning any influence of length of stenosis on recurrence rate—however, in retrospective, we have also measured the length of stenosis in the two study groups, and the data are given in Table 1⇓).
Baseline Characteristics of the 216 Patients According to Assigned Treatment
The baseline characteristics of the 216 study patients with presenting symptoms, associated diseases and risk factors, and concomitant drug therapy are listed in Table 1⇑ according to the assigned treatment.
Angioplasty Procedure and Follow-up
All procedures were performed with a 6F introducer sheath that was removed immediately after the end of the procedure, and an ipsilateral antegrade puncture was used in all patients. Uniplanar intra-arterial digital subtraction angiography—using a Bicor Medical System (Fa. Siemens)—was done in all patients immediately before and after the procedure. Contrast medium with 300 mg iodine/mL was injected in a diluted form (100 mL of contrast medium was diluted with 50 mL 0.9% saline solution) manually at 2 to 5 mL/s for a total volume of 10 to 15 mL per injection. Video acquisition was obtained from a 23-cm image intensifier at two frames per second into a 512×512 matrix. Postprocessing was used to achieve optimal mask/image combinations, and a hard copy was obtained with a standard multiformat camera. The degree of residual stenosis immediately after PTA—or recurrent stenosis in case of later control arteriography—was determined by comparison of the width of the contrast column (the measurements were made with a ruler) at the point of maximal diameter reduction within the dilated segment to that of an unaffected arterial segment immediately proximal. The measurements were taken blindly with no knowledge of clinical data and treatment assignment. Procedural success was evaluated according to technical success, which required angiographic patency with post-PTA residual stenosis of <50% diameter reduction.
Treatment with intravenous application of 500 mg aspirin (Aspisol; Bayer) was initiated on the morning of PTA at least 1 hour before the planned procedure, and the same dosage was applied for 2 additional days. During the intervention and after successful passage of the lesion with the guide wire, 5000 IU heparin was administered through the sheath before inflation of the balloon. After successful PTA, the patients also received heparin intravenously for 3 days. Administration of heparin was started immediately after PTA in a dosage of 1000 IU/h and was adjusted twice daily according to the thrombin time (prolongation to at least three times the normal value). The patients without early recurrence within 3 days after PTA were randomly allocated to receive therapy with high-dose or low-dose aspirin as described above. The patients and the main investigators were not blinded to dose assignment. However, noninvasive laboratory testing, duplex ultrasound investigations, and angiographic follow-up investigations were performed and analyzed by investigators without knowledge of group randomization.
Each patient gave his or her informed consent to participate in the study. The patients were also asked to avoid other aspirin-containing drugs and nonsteroidal anti-inflammatory drugs. The referring physicians were informed about the aim of the present study and asked not to stop or change study medication without contacting our center.
Follow-up examinations were performed the day after PTA, at the time of hospital discharge (4 to 5 days after PTA), and 1, 3, 6, 12, and 24 months after PTA. Follow-up examinations included assessment of symptoms, assessment of compliance with study medication and of adverse effects (especially gastric discomfort) clinical examination (pulse palpation and auscultation of the treated vessel), and noninvasive laboratory testing (pulse volume recordings and ankle-brachial arterial pressure measurement with Doppler ultrasound). The higher value of the ankle pressure measured at either the posterior tibial or dorsal pedal artery was used to calculate the Doppler index (ratio of ankle pressure to brachial artery pressure). In diabetic patients with incompressible arteries because of mediasclerosis, which precludes accurate pressure measurements, pulse waveform analysis and measurement of toe blood pressure using digit plethysmography were used to judge deterioration.
In all patients for whom there was no suspicion of recurrent stenosis on the basis of history, clinical examination, or laboratory findings, color flow duplex sonography of the femoropopliteal segment also was performed at the end of follow-up (after 24 months) to exclude otherwise undetected recurrence.
All duplex examinations were performed with an Acuson-128 duplex scanner (Acuson Inc) with a 5-MHz transducer by one observer with extensive experience. Determination of ≥50% stenosis was based on previously published duplex velocity criteria for lower extremity arterial stenosis.16 A focal increase in peak-systolic velocity of at least 100% in the dilated segment compared with that in the immediate preceding arterial segment (velocity ratio ≥2) was considered indicative of a ≥50% stenosis at that site. However, there was no patient with recurrence detected newly only by duplex sonography.
If recurrent stenosis was suspected on the basis of clinical or laboratory findings (the Doppler ankle-brachial index deteriorated by more than 0.15 from the maximum early postprocedural level), angiography was performed. This was done by intravenous digital subtraction angiography, and in patients requiring repeat PTA, an intra-arterial angiogram (by ipsilateral antegrade puncture) was obtained. Recurrence was defined as an angiographically documented stenosis of >50% narrowing of the current luminal diameter or a reocclusion within the recanalized segment. As stated, the angiograms and duplex investigations were performed and analyzed by an experienced angiographer (sonographer) without knowledge of group randomization.
The time of recurrence was judged by recurrence of symptoms or, for patients with asymptomatic recurrence, the failure date was taken as halfway between the last two examining points.
The follow-up was 24 months in all patients (or until recurrence or death). It was not the aim of the study to evaluate the efficacy of different aspirin doses on patient survival. However, comparison of survival is interesting because of the increased risk of cardiovascular death of patients with peripheral arterial occlusive disease and a potential influence of aspirin. Because the numbers of patients were too small and the follow-up period was too short for survival analysis, the survival data reported in the results section must be interpreted only descriptively.
Statistical Analysis
For data storage and numerical analysis, SAS software was used.17 The Kaplan-Meier method was used to calculate the survival function, ie, the curve of the cumulative percentage patency rate versus time of follow-up. To test if there was a statistically significant difference between survival curves (P<.05), we used the generalized Wilcoxon test and the log-rank test.18
For multivariate analysis of the factors associated with late success of PTA, a rank test using a stepwise procedure for the association of covariates was carried out.18 Comparison of the treatment groups was made on an intention-to-treat basis; regardless of whether medication was taken, the patients were analyzed according to the originally assigned treatment until the last follow-up visit. In addition, we performed treatment analysis, in which we included only recurrences while study treatment was being taken. In this analysis. patients were excluded from further consideration when study medication was stopped.
Differences between treatment groups in the distribution of baseline characteristics were tested by contingency table analysis or two-sample test.
Results
The two treatment groups that resulted from the randomization process were well balanced. No differences were seen in any of the demographic, clinical, or angiographic patient characteristics (Table 1⇑).
Complete follow-up information (patency after 24 months, restenosis or reocclusion, and death) was obtained in 207 patients, and 9 patients (4.2%) were lost to follow-up after hospital discharge (5 in the treatment group with 1000 mg aspirin and 4 in the treatment group with 100 mg aspirin). In 4 of these 9 patients, information could be obtained about their death without knowledge concerning patency of the recanalized segment.
During the 2-year follow-up period, 72 patients (36 in the high-dose and 36 in the low-dose aspirin group) developed angiographically verified restenosis of >50% within the recanalized segment. Twenty-six of these 72 patients had a complete reocclusion at the time of angiography (12 in the high-dose and 14 in the low-dose group). By intention-to-treat analysis, the cumulative patency rates at 24 months of follow-up were 62.5% in the high-dose and 62.6% in the low-dose aspirin group (Wilcoxon, P=.97; log-rank, P=.97) (Fig 1⇓). The patency rates calculated according to the treatment analysis differed little from those of the intention-to-treat analysis (high-dose aspirin group, 63.1%; low-dose aspirin group, 60.7%; Wilcoxon, P=.70; log-rank, P=.72).
Plot of cumulative patency rate (Kaplan-Meier) after femoropopliteal percutaneous transluminal angioplasty, according to assigned treatment. Numbers below the figure indicate the number of patients at risk.
We have further analyzed (by univariate analysis) different variables concerning their influence on long-term patency after 24 months. However, this analysis was not the main concern of the present study, and therefore the results should mainly be interpreted descriptively.
The following variables were analyzed concerning long-term patency after 24 months: (1) indication for PTA (the cumulative 2-year patency rate was 64% in patients with PTA because of claudication and 50% in patients with critical leg ischemia; Wilcoxon, P=.06; log-rank, P=.11); (2) runoff vessel disease (patients with good runoff had a better 2-year patency [67.5%] than patients with no or only one patent calf artery [49%]; Wilcoxon, P=.02; log-rank, P=.02); (3) type of treated lesion (single stenosis versus multiple stenoses versus occlusion ≤10 cm versus occlusion >10 cm—the cumulative 2-year patency rates were 73.8%, 62.8%, 58.7%, and 46.6%, respectively); (4) diabetes (nondiabetics had better long-term results than diabetics; 2-year cumulative patency, 68% and 54%, respectively; Wilcoxon, P=.06; log-rank, P=.05); (5) sex (the long-term results were better for men than for women; 68.2% versus 55.5%; Wilcoxon, P=.06; log-rank, P=.06—however, the aspirin dosage had no influence on the cumulative patency in either sex [men: Wilcoxon, P=.93; log-rank, P=.88; women: Wilcoxon, P=.80; log-rank, P=.76]); and (6) hypertension (for definition, see Table 1⇑), smoking habits (defined as current smoking of any number of cigarettes), and age (>65 years)—these variables were not related to long-term success of PTA. Sample size was not large enough to exclude any influence of these variables on long-term patency definitely. However, the differences were so small that any influence should be without clinical importance.
With multivariate analysis by a rank test using a stepwise procedure, two variables proved to be significant predictors of long-term success: the runoff (Wilcoxon, P=.03; log-rank, P=.02) and the type of treated lesion (Wilcoxon, P=.02; log-rank, P=.03).
Angiography was performed in 88 patients. In 75, indication for angiography was suspicion of recurrent stenosis, which could be confirmed in 72 patients, whereas 3 patients had clinical or hemodynamic deterioration because of progression of atherosclerosis at other sites. In 13 patients, angiography was primarily performed because of symptoms of the contralateral leg. Color-flow duplex sonography, which was performed at the end of follow-up in all patients without indication for angiography (no suspicion of recurrent stenosis on the basis of history or clinical or laboratory findings), did not detect any further recurrence.
Redilation by means of a second PTA was performed in 26 (36%) of the patients with restenosis (15 in the high-dose and 11 in the low-dose aspirin group), whereas 5 patients (2 in the high-dose and 3 in the low-dose group) underwent bypass surgery. PTA of the contralateral leg was performed in 36 patients (21 in the high-dose and 15 in the low-dose group) during the 2-year follow-up period.
No patient required amputation. During the follow-up, 27 patients died (12.8% of the 211 patients with follow-up information concerning survival)—14 in the high-dose and 13 in the low-dose aspirin group. Fourteen patients died from coronary heart disease, 5 from stroke, and 8 had other (including unknown) causes. The cumulative survival at 24 months of follow-up was 86.6% in the high-dose and 87.7% in the low-dose aspirin group.
The clinical status of the patients after 2 years is listed in Table 2⇓ according to the clinical stage at study entry. Symptomatic patients were listed according to the worse leg. Before PTA, the ankle-brachial blood pressure ratio averaged 0.55±0.17, and it reached its maximum 1 month after PTA (0.92±0.17). Fig 2⇓ demonstrates the course of this index on the primarily treated leg and the contralateral leg during follow-up (values for patients with secondary interventions because of recurrence are included).
Plot of ankle-brachial index (mean±SD) before and after percutaneous transluminal angioplasty (PTA) in the treated leg compared with the contralateral leg.
Clinical Status of the Patients After 24 Months According to Clinical Stage at Study Entry
The number of patients discontinuing therapy before the end of the 24 months of follow-up was 30 in the high-dose aspirin group and 11 in the low-dose group (P<.01). The median time from randomization until discontinuation of trial medication was 2 months in the high-dose and 3 months in the low-dose aspirin group. Twenty patients in the high-dose aspirin group discontinued therapy because of gastric discomfort (eg, epigastric pain, nausea or vomiting) compared with 4 in the low-dose group (P<.01). Aspirin medication was stopped by the physician in 3 patients receiving 1000 mg/d because of melena (in 1 patient, this complication required hospitalization). We did not search for occult blood loss in this study. There were no fatal bleeding complications.
Discussion
The follow-up period of 24 months should be sufficient to test the effect of pharmacological intervention on the development of restenosis; most recurrent lesions develop within this time.19 20 21 22
Our findings concerning factors that determine the long-term success of PTA are in accordance with reported results. Critical leg ischemia,19 21 22 23 24 poor runoff,20 21 23 24 25 angiographically severe lesions,20 21 22 and presence of diabetes20 21 22 24 are factors negatively influencing long-term patency. A higher recurrence rate in women—as observed in the present study—was also reported by others.22 26
The controversy regarding the optimal dosage of aspirin is not solved definitely. The early clinical trials of antiplatelet therapy used aspirin dosages of >900 mg/d.3 4 Subsequent studies in humans suggested that lower doses might be more effective because they could dissociate the opposing effects of aspirin on platelet and vascular reactivity and could favor prostacyclin synthesis.27 28 However, recent data failed to confirm the concept of a differential effect of low doses of conventionally formulated aspirin on platelet and vascular prostaglandin synthesis in humans.29 30 The clinical importance of a controlled-release preparation of low-dose aspirin for preserving systemic prostacyclin biosynthesis is unknown.30 As a result of the biochemical studies on the mechanism of action of aspirin and stimulated by the clinical necessity to minimize any side effects, several clinical studies using low doses of aspirin have been done during the past years in various categories of patients with vascular diseases.5 6 7 8 9 10 11 The results of these trials and of the present study suggest no difference in clinical effectiveness concerning cardiovascular risk reduction among high, medium, and low doses of aspirin. Otherwise, as demonstrated in the present study, the gastrointestinal side effects of aspirin are dose dependent,7 9 12 and the main advantage of a low dose is the lower frequency of serious gastrointestinal adverse effects.
It was not the aim of the present study to evaluate the efficacy of PTA, and therefore only patients with successful PTA were included. We started randomization after successful PTA to avoid a dropout rate of approximately 10% to 20% because of procedural failure or complications. The heparin regimen followed in the study was part of our standard procedural protocol at the beginning of the trial. It proved useful to avoid early thrombosis.31 The low early recurrence rate (only 16 of 320 patients, or 5%, treated during the study period by PTA for femoropopliteal lesions had to be excluded from entry into the study because of reocclusion within the first 3 days after recanalization) confirms the effectiveness of this regimen. Otherwise, only 6 patients (1.8%) needed surgery because of bleeding at the puncture site (2) or false aneurysm (4).
Initial experience indicated beneficial effects of both anticoagulant and antiplatelet therapy when used after catheter recanalization either with the Dotter or balloon catheter technique in patients with peripheral arterial occlusive disease.32 33 34 Consequently, when the present study was started, randomization of patients to placebo therapy would not have been considered clinically feasible because of ethical objections.35 Moreover, the increased risk of death from cardiovascular causes in these patients,1 2 the well-established prophylactic effect of aspirin in reducing the risk of future cardiovascular events,3 4 and the demonstrated prophylactic effect of aspirin concerning the natural history of peripheral arterial occlusive disease36 37 were reasons to favor long-term use of aspirin.
Studies in animals and humans have shown that platelet deposition occurs after angioplasty in the dilated vascular segment,38 39 40 41 and restenosis after angioplasty was significantly attenuated by antiplatelet agents in experimental models.42 However, there are different problems involved in applying the results of animal models to humans.43 Although antiplatelet agents are useful to reduce the acute thrombotic complications in patients undergoing peripheral and coronary angioplasty,33 44 they have not been shown to influence the chronic restenosis rate after percutaneous transluminal coronary angioplasty.44 Recently, Taylor et al45 reported a small beneficial effect of low-dose aspirin (100 mg/d) on restenosis after coronary angioplasty. However, to our knowledge, no placebo-controlled trials of any antithrombotic interventions for prevention of recurrence after PTA have been performed for a long time. Heiss et al14 reported a study in which patients were randomly assigned to one of three therapeutic groups: a placebo or a fixed combination of dipyridamole (75 mg) and aspirin (either 100 or 330 mg) three times a day for 6 months after femoropopliteal PTA. These authors reported a significantly lower recurrence rate with high-dose aspirin plus dipyridamole (39%) compared with placebo (63%). In the low-dose aspirin group, 51% of the patients deteriorated during the first 6 months after PTA. Unfortunately, results were not published according to the suggested standards for evaluating results of interventional therapy for peripheral vascular disease,46 and the unexpectedly high recurrence rate cannot be explained because of lack of data about the underlying lesions.
We cannot definitely exclude the possibility that the lack of a difference in long-term patency between high-dose and low-dose aspirin in the present study was caused by the identical periprocedural treatment during the first 3 days after PTA. However, this seems unlikely because in a former study41 we could not observe any difference in the amount of platelet accumulation early after PTA in patients treated with the same aspirin dosages as used in the present study.
The present study may be criticized because control arteriography was not done in every patient. However, we are sure that according to our protocol we did not miss any recurrence. Follow-up angiography was performed in every case of clinical suspicion of recurrence (by history or clinical examination) and/or of strictly defined laboratory findings. We suspected restenosis if the Doppler ankle-brachial index deteriorated by >0.15 from the maximum early postprocedural level. It is generally agreed that the ankle-brachial pressure index must change at least 0.15 before it can be considered significant.47 A change of <0.15 is considered an acceptable level of variability. Rutherford and Becker46 also stated in their article concerning standards for evaluating and reporting the results of surgical and percutaneous therapy for peripheral arterial disease that the maintenance of achieved improvement in the pressure index without deterioration by >0.15 from the maximum early postprocedural level is an objective method of determining patency. If, however, we had missed any recurrence by clinical examination or by Doppler ankle-brachial index measurement, we should have detected such a recurrence by duplex sonography.
A peak systolic velocity ratio of ≥2 showed excellent sensitivity (93%) to detect reduction in luminal diameter of ≥50% in the femoropopliteal region.48 Seifert and Jäger49 reported a sensitivity of 98% to detect a >50% stenosis. In a recent study, Ranke et al50 reported an optimal cutoff value of ≥2.4 for the peak systolic velocity ratio to detect ≥50% diameter reduction. Therefore, by using a value of ≥2 as in the present study, we should have detected even lower-grade stenoses.
The most effective medical maintenance therapy remains to be established in reducing the incidence of restenosis after angioplasty. The current evidence indicates that restenosis involves a fibroproliferative response to vascular injury by complex processes involving several cell types and different mediating mitogens, and this implies that no single intervention is likely to be fully effective.43 51 However, independent of any beneficial effect of aspirin on restenosis, many patients undergoing PTA because of peripheral vascular disease are prescribed aspirin because of its well-established effect in reducing cardiovascular morbidity and mortality. Based on results from the present study, we see no reason to recommend doses of >100 mg/d, because this dosage was no less effective in the prevention of restenosis after femoropopliteal PTA than was 1000 mg/d.
Acknowledgments
We thank Dr Barbara Schneider (Institute of Medical Statistics, Vienna, Austria) for help in statistical analyses.
- Received June 8, 1994.
- Accepted November 25, 1994.
- Copyright © 1995 by American Heart Association
References
This Issue
Jump to
Article Tools
Share this Article
- Comparison of Effects of High-Dose and Low-Dose Aspirin on Restenosis After Femoropopliteal Percutaneous Transluminal Angioplasty
