Greater life expectancy and the increasing prevalence of diabetes mellitus in developed countries lead to a progressively rising number of patients with critical limb ischemia (CLI) and intermittent claudication (IC) due to infrapopliteal artery (IPA) disease (1).

Over the past years prospective nonrandomized and randomized trials revealed that sirolimus-eluting stents (SES) are significantly superior to percutaneous transluminal balloon angioplasty (PTA) and bare-metal stents (BMS) concerning patency rates, and in parts target lesion revascularization (TVR) rates and improvement in Rutherford-Becker class (2- 5). However, none of these studies confirm an advantage of SES regarding amputation rate, limb salvage, and mortality. Moreover, due to the fact of higher costs and inconclusive evidence from nonrandomized, single-center studies advocating a clinical impact of SES in a long-term follow-up, PTA and BMS potentially remain the gold standard in treatment of focal IPA lesions (6- 7).

To gain more evidence regarding the clinical impact of SES placement in treatment of IPA lesions, the follow-up period of a prospective, double-blind, randomized, multicenter trial comparing polymer-free SES with BMS was extended. This trial confirms significantly higher patency rates and greater improvement in Rutherford-Becker class in patients treated with SES in comparison with BMS. No significant difference could be observed concerning limb salvage rate, target lesion revascularization, and event-free survival 1 year after index procedure (5).

Between April 2006 and April 2008 CLI and IC patients were enrolled in this prospective, randomized, double-blind trial. Patients were eligible for the study if they were at least 21 years old, were not pregnant, and suffered from peripheral artery disease Rutherford-Becker class of 3 to 5. Angiographic eligibility criteria were the presence of a single de novo lesion in an IPA that did not exceed 45 mm in length. We allocated patients to the 2 treatment groups (polymer-free SES and BMS) using a computer-generated random sequence. All patients received aspirin (100 mg daily) and clopidogrel (loading dose of 600 mg before the procedure followed by 75 mg daily for 6 months). Study design, study procedures, and 1-year results of this trial were published (5).

After completion of the 1-year results the extension of the follow-up period was approved by the ethics committee on August 12, 2011. All patients gave written informed consent.

Clinical follow-up was obtained through outpatient visits, direct phone call assessments, and correspondence with the primary physician with attention directed to a questionnaire regarding the clinical course of each patient including target limb major and minor amputation, TVR, myocardial infarction, and causes of death. In addition, in patients who were willing and capable a determination of the Rutherford-Becker class was performed.

The main study endpoint was the event-free survival rate defined as freedom from TVR, major and minor target limb amputation, myocardial infarction, and death. Secondary endpoints included amputation rate, TVR, and change in Rutherford-Becker class. All events were determined cumulatively for the 1,100 days after stent placement.

Data for all endpoints were evaluated in the intention-to-treat analysis. Continuous data are expressed as mean ± SD. Categorical variables were compared with the use of the 2-sided chi-square test and continuous variables were compared with the use of the 2-sided Student t test. Changes in Rutherford-Becker class were expressed as median with interquartile range, and group comparisons were performed using the Mann-Whitney U test. Event-free survival was compared by Kaplan-Meier analysis with the use of the Mantel-Cox log-rank. A 2-sided p value <0.05 was considered to indicate statistical significance. To summarize differences in the primary endpoint between the study groups, we derived hazard ratios (HR) with associated 95% confidence intervals (CI) from the Cox proportional hazards model. To assess the interaction of stent type with stage of disease (CLI or IC) and to adjust for remaining imbalance, we performed Cox regression analyses. In addition to the interaction term, the multivariable models included demographic, clinical, and interventional variables (Table 1) with a difference between the 2 study groups at a value of p ≤ 0.1. All statistical analyses were performed with the SPSS software, version 18.0 (SPSS Inc., Chicago, Illinois).

At baseline 161 patients were included in this trial. Eighty-two patients were randomly assigned to receive the polymer-free SES, and 79 were assigned to receive the BMS. With exception of a significantly higher body mass index in the SES group, the 2 groups were similar with respect to all baseline variables. A total of 155 patients (96.3%) had 0 or just 1 patent IPA to the ankle joint (SES group, 95.1%; BMS group, 96.2%; p = 0.9) at baseline (Table 1).

The mean follow-up time was 1,005 ± 139 days in the SES group and 1,027 ± 123 days in the BMS group (p = 0.38). Concerning the primary endpoint, event-free survival rate compared by Kaplan-Meier analysis was 65.8% in the SES group and 44.6% in the BMS group (p = 0.02) (Figure 1). Thus, in comparison with SES, BMS placement was associated with a HR for adverse events of 1.8 (95% CI: 1.1 to 2.9; p = 0.02). The higher risk of adverse event occurrence associated with BMS prevailed after adjustment for renal insufficiency, CLI, and body mass index. The corresponding adjusted HR was 1.7 (95% CI: 1.1 to 2.8; p = 0.03). No significant interaction (p = 0.29) could be observed between stent type (SES or BMS) and stage of disease (CLI and IC) regarding the primary endpoint when added to the Cox regression model. The impact of the SES on the incidence of adverse events in various subgroups of patients is presented in (Figure 2).

Seventeen patients (22.6%) in the SES group and 18 patients (24%) in the BMS group (p = 0.84) died during follow-up. Thirteen patients (8.6%) died because of major cardiac events (myocardial infarction, heart failure). Seven patients (4.6%) died in consequence of gastrointestinal and pulmonary infections or cancer. In 15 patients (10%) the cause of death remained uncertain. Ten (6.2%) patients (6 in the SES group and 4 in the BMS group; p = 0.75) were lost to follow-up.

Freedom from any target limb amputation was documented in 97.4% in the SES group and 87.8% in the BMS group (p = 0.03). Limb salvage rates were 98.7% and 94.6% (p = 0.17), respectively.

Clinically driven (recurrent symptoms of IC and/or worsening skin lesion) TVR was performed in 7 patients (9.2%) in the SES group and in 15 patients (20%) in the BMS group (p = 0.06). The median (interquartile range) improvement in Rutherford-Becker class was –2 (–3 to –1) and –1 (–2 to 0; p = 0.006), respectively (Table 2).

Concerning patients with CLI event-free survival was 57.9% in the SES group and 32.3% (p = 0.07) in the BMS group (Figure 3). Amputation was needed in 2 patients treated with SES (1 major and 1 minor, 5.3%) and in 7 patients treated with BMS (4 major and 3 minor, 22.6%; p = 0.04) (Table 3).

In patients with IC event-free survival compared by Kaplan-Meier analysis was 71.1% in the SES group and 50% (p = 0.07) in the BMS group, TVR rates were 7.9% and 25% (p = 0.04), and changes in Rutherford-Becker class were –2 (–2 to –1) and –1 (–1 to 0; p = 0.03) (Table 3), respectively.

To our knowledge, this is the first prospective, randomized, multicenter study showing favorable clinical outcomes for SES compared with BMS in IPA application during long-term follow-up. While PTA and BMS placement provide promising acute procedural results in treatment of IPA the technical durability is hampered by restenosis rates of more than 50% after 1 year (2,8- 9). There is evidence from nonrandomized and randomized trials that SES is superior to PTA and BMS concerning patency rates, TVR, and improvement in Rutherford-Becker class (2,5,10). However, no data from randomized studies are available concerning the clinical impact of this finding demonstrating fewer amputations and higher limb salvage rates after SES placement in comparison with PTA and BMS.

The present trial reveals that the SES has significant advantages regarding event-free survival and amputation rates over BMS. Moreover, significantly greater improvement in Rutherford-Becker class and a clearly lower TVR rate compared with BMS was noticeable in the long-term follow-up. No significant interaction was found between stent type and stage of disease (CLI or IC) concerning the primary endpoint.

To date, there are only few data released from nonrandomized trials regarding long-term results after treatment of IPA lesions. Moreover, most of these trials exclusively address CLI patients. Therefore an accurate comparison with the present trial investigating patients with CLI and IC is restricted. Although the number of patients included in the present trial limits the validity of a subgroup analysis limb salvage and mortality rates of CLI patients treated with BMS are comparable with the results of CLI patients treated with PTA in former studies (8,11).

A nonrandomized, single-center study including 106 patients illustrates a limb salvage rate of 94 ± 2% in patients treated with SES in a 27 ± 19 months follow-up period. Target limb reintervention was performed in 15% and the mortality rate was 29% (6). In a prospective registry investigating the performance of SES versus BMS for CLI at 3 years a significantly better primary patency (HR: 4.81; 95% CI: 2.91 to 7.94; p < 0.001), and a better intervention-free survival (HR: 2.56; 95% CI: 1.30 to 5.00; p = 0.006) in patients treated with SES could be observed (7). The present study reveals comparable results in particular concerning CLI patients treated with SES regarding amputation rate, TVR rate, and mortality. Moreover, in comparison with the BMS group significantly fewer amputations occur, and a considerably higher limb salvage rate could be documented in the SES group.

Although it has become an established treatment for CLI (1,12) endovascular therapy of IPA in patients with IC is not yet accepted as a therapy option, despite published data underlining the beneficial clinical effect (13- 15). In this context, results of the present study support findings that the degree of IC is positively affected by successful and long-lasting interventions of IPA lesions (4- 5,16). In addition, compared with BMS, SES placement in this patient cohort was associated with a significantly reduced TVR rate and significantly greater improvement in Rutherford-Becker class.

Changes in quality of life with standardized questionnaires and possible cost savings by using drug-eluting stents in IPA lesions needs to be clarified in upcoming studies.

It must be highlighted that this trial only addresses focal lesions. Particularly in patients with diabetes and renal failure the vast majority of IPA lesions appear as diffuse, long lesions. However, due to the absence of long devices and restricted applicability in juxta-articular regions (rigid scaffold of metallic stents) the use of stents (SES) in long IPA lesions is limited. Moreover, the polymer-free SES used in the present trial is not available in the United States. Whether comparable results can be achieved with next-generation stent devices (with thinner struts, rapamycin analogs) and drug-eluting balloons requires further investigation.

In a long-term follow-up, this prospective, randomized, multicenter study demonstrates significantly higher event-free survival rates and reduced amputation rates after treatment of focal IPA lesions with SES as compared with treatment with BMS. Both patients with CLI and IC profit from treatment with SES.