Author + information
- Received July 20, 2016
- Revision received October 3, 2016
- Accepted October 20, 2016
- Published online January 2, 2017.
- Tatsuya Nakama, MDa,∗ (, )
- Nozomi Watanabe, MDa,
- Takuya Haraguchi, MDb,
- Hiroshi Sakamoto, MDc,
- Daisuke Kamoi, MDd,
- Yoshinori Tsubakimoto, MDe,
- Kenji Ogata, MDa,
- Katsuhiko Satoh, MDb,
- Kazushi Urasawa, MDb,
- Hiroshi Andoh, MDc,
- Hiroshi Fujita, MDe and
- Yoshisato Shibata, MDa
- aMiyazaki Medical Association Hospital, Cardiovascular Center, Miyazaki, Japan
- bCardiovascular Center, Tokeidai Memorial Hospital, Sapporo, Japan
- cDepartment of Cardiology, Kasukabe Chuo General Hospital, Saitama, Japan
- dDepartment of Cardiology, Nagoya Kyoritsu Hospital, Nagoya, Japan
- eDepartment of Cardiology, Japanese Red Cross Kyoto Daini Hospital, Kyoto, Japan
- ↵∗Reprint requests and correspondence:
Dr. Tatsuya Nakama, Miyazaki Medical Association Hospital, Cardiovascular Center, 738-1 Funato Shinbeppu-tyo, Miyazaki 880-0834, Japan.
Objectives The aim of this study was to investigate the clinical outcomes of pedal artery angioplasty (PAA) for patients with critical limb ischemia.
Background Pedal artery disease is considered a predictor of delayed wound healing (DH) after endovascular therapy. Adjunctive PAA might improve the speed and extent of wound healing.
Methods Consecutive patients with critical limb ischemia (n = 257) presenting with de novo infrapopliteal and pedal artery disease were retrospectively reviewed from a multicenter registry. Patients were divided into 2 groups according to whether PAA was performed (n = 140) or not (n = 117). The rate of wound healing and time to wound healing were compared between these groups. DH score was calculated using the number of independent predictors of DH. Patients were stratified into 3 groups according to DH score: low risk (DH score = 0), moderate risk (DH score = 1 or 2), and high risk (DH score = 3). Estimated efficacy was analyzed for each risk-stratified population.
Results The rate of wound healing was significantly higher (57.5% vs. 37.3%, p = 0.003) and time to wound healing significantly shorter (211 days vs. 365 days; p = 0.008) in the PAA group. In a multivariate analysis, nonambulatory status, target wound depth (UT grade ≥2), and daily hemodialysis were revealed as predictors of DH. In the moderate-risk population, adjunctive PAA significantly improved the rate of wound healing (59.3% vs. 33.9%; p = 0.001). In the high-risk population, however, PAA did not affect wound healing.
Conclusions Patients who underwent PAA showed a higher rate of wound healing and shorter time to wound healing, especially in the moderate-risk population. With regard to wound healing, this aggressive strategy might become a salvage procedure for patients with critical limb ischemia presenting with pedal artery disease.
- below-the-ankle intervention
- critical limb ischemia
- endovascular therapy
- pedal artery diseases
- peripheral arterial diseases
- wound healing
Critical limb ischemia (CLI) is an unfavorable clinical consequence of peripheral arterial disease. Approximately 73% to 95% of patients lose limbs within 1 year after conservative treatment (1). Arterial revascularization, including surgical bypass and endovascular therapy (EVT), is the optimal treatment to prevent limb loss (2). Following the result of the BASIL (Bypass Versus Angioplasty in Severe Ischaemia of the Leg) trial (3), the latest American Heart Association and American College of Cardiology guidelines suggest that arterial revascularization strategies should be decided on the basis of the existence of adequate veins or patients’ life expectancy (4). However, patients with CLI rarely qualify as surgical candidates, because of their frailty, concomitant diseases, and advanced age. Percutaneous EVT, in contrast, has become popular because of its lower invasiveness and comparable limb salvage (LS) rate to surgical bypass (5–11). Recent clinical trials have reported sufficient LS, and these results have already achieved the objective performance goal (84% LS rate at 1 year), which was proposed by Conte et al. (12). Therefore, the role of EVT for patients with CLI has been significantly increasing in the current clinical setting.
However, even if major amputation is successfully avoided, achievement of complete wound healing is still challenging. Recent clinical reports have documented the apparent discrepancy between LS rate and rate of wound healing (8–11,13). Delayed or incomplete wound healing adversely affects patients’ quality of life and impedes social rehabilitation. Successful complete wound healing should be an advanced goal beyond preventing major amputation.
Several clinical trials have reported that the existence of pedal artery disease results in worse wound healing (9,13). Hence, adjunctive revascularization procedures for pedal artery disease might improve the rate of wound healing. We have already investigated the clinical outcomes of additional pedal artery angioplasty (PAA) for patients with CLI (14). Patients who underwent PAA showed a higher rate of wound healing and shorter time to wound healing. However, this result was from only a single-center, retrospective analysis with a small cohort. Further investigation with a larger number of patients is required to assess the clinical effect of PAA in patients with CLI. In the present study, we sought to investigate the clinical implications of PAA for patients with CLI presenting with pedal artery disease from the multicenter RENDEZVOUS (Retrospective Analysis for the Clinical Impact of Pedal Artery Revascularization Versus Non-Revascularization Strategy for Patients With Critical Limb Ischemia) registry.
The RENDEZVOUS registry was a multicenter registry that included patients with CLI presenting with infrapopliteal arterial disease who underwent EVT at 5 experienced cardiovascular centers in Japan. The study protocol was developed in accordance with the Declaration of Helsinki and was approved by the ethics committees of each participating institution. All patients gave written informed consent before the EVT procedure.
Enrolled patients and initial evaluation
From January 2012 to June 2014, 317 consecutive patients with CLI presenting with de novo infrapopliteal arterial disease were enrolled. In patients with bilateral CLI, only the more severe limbs were enrolled. Patients without ischemic gangrene or ulcerations (defined as Rutherford class [RC] 4) were not enrolled. Sixty patients without pedal artery disease were excluded. Finally, 257 patients with 257 limbs were retrospectively reviewed. These patients were classified into 2 groups according to whether adjunctive PAA was conducted (n = 140) or not (n = 117). The indication of arterial revascularization and primary target artery were judged by a conference of vascular specialists at each institution. Primary target artery was typically decided on the basis of the angiosome concept (15). If the angiosome direct revascularization was judged to be difficult or failed, indirect revascularization was performed as an alternative strategy. The indication of adjunctive PAA was judged by at least 2 endovascular specialists with sufficient knowledge and experience for EVT for CLI. Basic indications of adjunctive PAA were: 1) absence of visible direct flow after target artery revascularization; 2) impaired flow phenomenon due to poor pedal artery runoff; and 3) widespread wounds or limb-threatening infection, which may require a large amount of blood supply for limb salvage and wound healing.
On admission, ankle-brachial index (ABI) and skin perfusion pressure (SPP) were measured to assess hemodynamic status. Wound care specialists individually evaluated each wound. The degree of pedal artery disease in each limb was categorized according to the Kawarada classification (9): type 1 indicated patent dorsal and lateral plantar arteries, type 2 had either the dorsal (type 2A) or lateral plantar (type 2B) artery patent, and type 3 indicated that both dorsal and lateral plantar arteries were occluded. Patients with type 1 pedal artery disease were excluded from the analysis.
Dual-antiplatelet therapy (aspirin 100 mg/day and clopidogrel 75 mg/day or cilostazol 200 mg/day) was administered before EVT and continued as long as possible. All EVT procedures were performed by skilled interventional cardiologists specializing in EVT for CLI. If there were femoropopliteal lesions, they were treated first with conventional EVT techniques (optimal balloon angioplasty and provisional nitinol stenting). Stenotic infrapopliteal arteries were treated with a 0.014-inch guidewire and conventional balloon angioplasty using standard techniques (16). In cases of occlusive lesions, stiffer penetration guidewires were sometimes used. When the antegrade approach failed, a retrograde approach, such as distal puncture or the transcollateral approach, was used (16). After guidewire passage, balloon dilation was performed with an optimally sized balloon at nominal pressure for at least 3 min. Atherectomy devices, drug-coated balloons, and infrapopliteal-dedicated stents were not used, because these devices were not available in Japan.
Adjunctive PAA was performed as required. A 0.014-inch hydrophilic guidewire was advanced into the occluded pedal artery with the support of a microcatheter. When the guidewire could not cross the occlusion, a bidirectional approach from both tibial arteries was established. If needed, novel retrograde access techniques such as metatarsal artery puncture (17,18) or a transcollateral approach were used. After guidewire crossing, occluded pedal arteries were dilated with a 2.0-mm balloon for at least 3 min. A representative case of PAA is shown in Figure 1.
All patients were followed up at 1 week; 1, 2, and 3 months; and then every 3 months up to 12 months after the procedure with duplex ultrasound scanning and ABI and SPP measurements. If a patient did not return to the hospital, phone calls were made to check the limbs and general status.
Wound status evaluation and management
Wound care specialists individually followed patients’ wounds. In the settings of wound infection, surgical debridement and antibiotic injections were conducted as required. All wounds were evaluated using the Rutherford classification, University of Texas (UT) grade, and Wound, Ischemia and Foot Infection (WIfI) classification system. RC 5 was defined as minor tissue loss. RC 6 was defined as major tissue loss (wounds extending above the transmetatarsal level). Wound depth was evaluated using UT grade (19), which is defined as follows: grade 1 = superficial wound not involving the tendon, capsule, or bone; grade 2 = wound penetrating the tendon or capsule; and grade 3 = wound penetrating the bone or joint. WIfI score and clinical stage were evaluated depending on the previous recommendations (20).
CLI was defined in accordance with Inter-Society Consensus for the Management of Peripheral Arterial Disease II guidelines (2). Major amputation was defined as an above-the-ankle amputation. Wound healing was defined as the achievement of complete epithelialization of all wounds without any major amputation. In cases requiring minor amputation with skin closure, complete wound healing was defined as the removal of all skin sutures. In patients who died before complete wound healing, the date of death was defined as the cutoff date. In patients who underwent major amputation, the time to wound healing was considered to be infinite. Nonambulatory status was defined as daily wheelchair use or bedridden status. PAA was successful if at least 1 (either dorsal or plantar) artery was recanalized and defined as complete PAA if both dorsal and plantar arteries were patent after treatment. Reintervention included any EVT or bypass surgery. Angiosome-oriented revascularization was defined on the basis of angiosome mapping, which has been described in a previous report (15).
The primary efficacy outcomes were rate of wound healing at 1 year after primary treatment and time to wound healing. Overall survival rate, LS rate, amputation-free survival rate, and freedom from reintervention rate were also evaluated as secondary efficacy outcomes. The success rate of adjunctive PAA and rate of procedure-related complications were evaluated as secondary safety outcomes. Furthermore, in this report, we propose a novel delayed wound healing (DH) score, which is a scoring system for the predicted clinical efficacy of adjunctive PAA. DH score is calculated using the number of independent predictors of DH, which were demonstrated in a multivariate analysis. All enrolled patients were stratified into 3 groups according to DH score: low risk (DH score = 0), moderate risk (DH = score = 1 or 2), and high risk (DH score ≥3). The 1-year rate of wound healing was evaluated in each estimated-risk population.
Statistical analysis was performed on an intention-to-treat basis. Continuous variables are expressed as mean ± SD. Categorical variables are expressed as count (percentage). The unpaired Student t test was used to compare continuous variables, and the chi-square or Fisher exact test was used for categorical variables as appropriate. Survival analyses using the Kaplan-Meier method were applied to the primary and secondary outcomes; the groups were compared using the log-rank test. Time to wound healing, which is expressed as median (interquartile range), was compared using the Mann-Whitney U test. Univariate and multivariate Cox proportional hazards regression models were used to investigate the independent predictors of DH. Clinical variables considered to contribute to DH in previous reports were entered into the multivariate analysis. Statistical significance was defined as a p value <0.05. All statistical analyses were performed using SPSS version 19.0 (IBM, Somers, New York).
Baseline characteristics are shown in Table 1. The mean age was 73.2 years, and 175 patients (68.1%) were men. Approximately one-half of the target population (n = 132) was nonambulatory. There were no significant differences in patient baseline characteristics between the PAA group and the non-PAA group, except for the proportion of nonambulatory patients (45.7% vs. 58.1%; p = 0.048).
Baseline target limb status is shown in Tables 2 and 3. Regarding the Rutherford classification of enrolled limbs, 200 limbs (77.8%) were classified as RC 5. There were no significant differences in the target limbs’ Rutherford classification among the study centers. Limb-threating infections were observed in 123 limbs (47.9%), and 128 limbs (49.8%) underwent surgical debridement or minor amputation. There were no significant differences in target wound location, size, UT grade, and WIfI clinical stage or composite score between the PAA group and the non-PAA group. There were no significant differences in baseline ABI and SPP before revascularization between the PAA group and the non-PAA group. However, as a result of aggressive pedal artery revascularization, ABI (0.88 vs. 0.81; p = 0.040) and dorsal SPP (49.3 mm Hg vs. 41.9 mm Hg; p = 0.024) were significantly higher in the PAA group than in the non-PAA group. Angiosome direct revascularization succeeded in 155 limbs (60.3%). In the non-PAA group, because of the existence of untreated pedal artery disease, the number of patients who underwent successful angiosome direct revascularization was significantly smaller than in the PAA group (43.6% vs. 74.3%, respectively; p < 0.0001).
All patients had pedal artery disease; 144 patients (56.0%) had type 2 and 113 patients (44.0%) had type 3 pedal arteries. Type 3 pedal arteries were more frequently observed in the PAA group (52.9%) than in the non-PAA group (33.3%). Complete pedal artery revascularization was achieved in 64 patients (45.7%).
The overall LS rate was 88.5%, and the amputation-free survival rate was 73.5%. The rate of wound healing was 49.5% (Figure 2).
The rate of wound healing was significantly higher (59.3% vs. 38.1%; p = 0.003) and time to wound healing significantly shorter (211 days [interquartile range: 69.25 to 365 days] vs. 365 days [interquartile range: 86.5 to 365 days]; p = 0.008) in the PAA group compared with the non-PAA group (Figure 3).
Secondary safety outcome
Table 4 shows procedural success and complication rates. Procedure-related complications occurred in 17 cases (6.6%). Of these, limb-threating complications were observed in 3 cases (1.2%). There were no significant differences in the incidence of complications between the PAA group and the non-PAA group. The most frequently observed complication was puncture-site hematoma (2.7%). In the PAA group, the overall technical success rate of adjunctive PAA was acceptable (88.6%), and complete pedal artery revascularization was obtained in 64 cases (45.7%).
Secondary efficacy outcome
There were no significant differences in overall survival rate (85.0% vs. 80.2%; p = 0.477), LS rate (88.7% vs. 88.2%; p = 0.923), amputation-free survival rate (76.4% vs. 70.1%, p = 0.923), and rate of freedom from reintervention (57.8% vs. 64.4%; p = 0.923) (Figure 4).
Independent predictors of DH and DH score
Table 5 shows the results of univariate and multivariate analyses for DH. Three independent predictors of DH and 1 positive predictor of wound-healing were revealed: 1) nonambulatory status; 2) daily hemodialysis; and 3) wound depth UT grade ≥2. Wound location and wound size did not have significant relationships with the rate of wound healing. Adjunctive PAA positively influenced the process of wound healing. However, angiosome direct revascularization and multivessel infrapopliteal revascularization did not demonstrate efficacy for the wound-healing process.
DH score was calculated using the number of independent predictors of DH mentioned previously. The enrolled patients were divided into 3 groups according to DH score: low risk (DH score = 0), moderate risk (DH score = 1 or 2), and high risk (DH score = 3). The rate of wound healing in each risk-stratified population was evaluated using the Kaplan-Meier method. Figure 5 shows the cumulative rate of wound healing in each risk-stratified population. In the low-risk population, the PAA group showed a higher rate of wound healing (93.3% vs. 69.2%; p = 0.184), but the result was not statistically significant. In the moderate-risk population, patients who underwent PAA showed a significantly higher rate of wound healing compared with patients who did not undergo PAA (59.3% vs. 33.9%; p = 0.001). However, in the high-risk population, additional PAA did not demonstrate efficacy (29.4% vs. 35.7%; p = 0.477).
This multicenter retrospective analysis shows the clinical outcomes of adjunctive balloon angioplasty for pedal artery occlusions in patients with CLI. Patients who underwent PAA showed a significantly higher rate of wound healing and shorter time to wound healing than those without PAA. According to DH scoring, the aggressive strategy was clinically beneficial, especially in the moderate-risk population. The efficacy of pedal intervention for the low-risk population was not statistically significant, and in the high-risk population, the achievement of complete wound healing seemed extremely challenging, even by adjunctive PAA.
Incomplete wound healing or DH is still an important clinical problem in the management of patients with CLI who have undergone EVT. The latest clinical trials have shown an 86% to 92% 1-year LS rate. However, 1-year rates of wound healing have remained at 63% to 74% (7–11). There is about a 20% difference between 1-year LS rate and rate of wound healing, which shows the reality of patients who undergo EVT. About 20% of patients who undergo EVT still experience unhealed wounds, even though major amputation can be successfully avoided. Reed et al. (21) reported the importance of complete wound healing on patients’ outcomes in a retrospective study. Their results suggest that strategies to improve the rate of wound healing and to reduce time to wound healing may be beneficial in reducing major adverse limb events, which may in turn improve quality of life and health care costs. Preventing limbs loss is only the first step in the management of CLI. Complete wound healing should be the next goal after the successful prevention of major amputation.
The latest clinical trials have already reported some predictors of DH. Kawarada et al. (9) reported diabetes mellitus, wound infection, and pedal artery disease as predictors of DH. Shiraki et al. (13) reported nonambulatory status, low albumin level, RC 6, wound infection, angiosome indirect revascularization, and pedal artery disease as independent predictors of DH. Kobayashi et al. (19) reported the importance of the wound depth. In our multicenter study, nonambulatory status, wound depth (UT grade ≥2), and daily hemodialysis were investigated as independent predictors of DH. As already reported, almost all predictors of DH were related to patients’ general status or target limbs condition. EVT strategy or angiographic findings seemed not as important to the wound healing process. Therefore, the latest clinical trials have emphasized the importance of a multidisciplinary approach for the achievement of complete wound healing (22). For further improvement in patients’ clinical outcomes, several treatment strategies have been proposed. Angiosome-oriented direct revascularization, which might improve the blood supply toward target wounds, is a widespread strategy (15). However, the effectiveness of the angiosome concept is still controversial (23). In our multicenter registry, the angiosome-oriented revascularization strategy did not demonstrate efficacy. In patients with pedal artery disease, original angiosome mapping is already destroyed, and the blood supply toward the target wounds depends on the patency of the original artery-to-artery connections and newly developed collateral channels. Therefore, the angiosome concept is no longer in force for this target population. Multivessel infrapopliteal revascularization has also been suggested as an effective procedure to gain sufficient perfusion in the foot. However, this strategy is also controversial, especially in patients with pedal artery disease, because the blood supply through the percutaneously recanalized infrapopliteal artery cannot “directly” reach the target wounds, because of the existence of pedal artery disease. Direct blood supply is ultimately dependent on the condition of the pedal arteries. Re-establishment of the “destroyed pedal artery” might be necessary to enable the efficacy of these strategies. The existence of pedal artery disease was the only predictor identified in multiple clinical trials as an angiographic factor for DH. Hence, in patients with pedal artery disease, pedal artery reconstruction might become a salvaging and alternative option for improvement of the rate of wound healing after EVT.
Some physicians have already reported cases of challenging pedal artery interventions. The group of Manzi et al. has reported many challenging cases, techniques, and the importance of pedal artery intervention (17,18). We have already reported the clinical implications of additional PAA for patients with pedal artery disease (14). Patients who underwent additional PAA showed a higher rate of complete wound healing and shorter time to wound healing than patients without PAA. However, it was only a small-sample, retrospective study. Further investigation was needed to prove the efficacy of PAA.
In this multicenter study, patients who underwent PAA showed a significantly higher rate of wound healing and shorter time to wound healing than patients without PAA. The results were almost the same as those of our previous single-center study. Furthermore, DH score, which is advocated in this report, may be useful for judging the indication of adjunctive PAA. Although this scoring system does not include any quantitative parameters, it is convenient and easily evaluates the estimated effectiveness of adjunctive PAA for each target population in daily clinical settings. The present study suggests that the moderate-risk population might be able to receive a significant benefit from adjunctive PAA. Aggressive pedal artery revascularization did not show a significant benefit for the low- or high-risk population in our results. Further clinical study is needed regarding the indication of adjunctive PAA.
First, this study was a retrospective analysis. It is impossible to deny the possibility of selection bias. Further prospective investigation is needed. However, randomized studies in this cohort of patients with CLI are practically impossible. Therefore, a prospective multicenter investigation with a well-defined protocol, and with accurate and independent evaluation of angiograms and wound status, must be considered in future investigations. Especially in the indication of PAA, revascularization based on quantitative hemodynamic data (SPP, transcutaneous oximetry, or toe pressure) should be demonstrated. Second, the procedure of PAA is not yet standardized. It requires challenging techniques. The development of novel techniques and dedicated devices are required for standardization of pedal artery intervention. Third, the durability of revascularized pedal artery occlusions remains unclear. In future investigations, angiographic evaluation is needed to evaluate the patency of the recanalized pedal arteries.
Patients who underwent PAA showed a significantly higher rate of wound healing and shorter time to wound healing than patients without PAA. Nonambulatory status, wound depth, and daily hemodialysis were identified as independent predictors of DH. DH score was useful for estimating the effectiveness of PAA for each risk-stratified target population. Aggressive PAA was shown to be an appropriate strategy, especially in the moderate-risk target population. The results of this study might be a milestone for the future of EVT for patients with CLI.
WHAT IS KNOWN? Pedal artery disease is considered an independent predictor of DH after endovascular treatment. Challenging PAA might improve patient clinical outcomes, especially in the speed and extent of wound healing.
WHAT IS NEW? The results of this multicenter registry show the clinical outcomes of PAA for patients with CLI. Adjunctive PAA significantly improved the 1-year rate of wound healing, especially in the moderate-risk target population.
WHAT IS NEXT? A further prospective, multinational investigation with a well-defined protocol, and with accurate and independent evaluation, is needed to confirm the results of this multicenter registry.
The authors thank all of the medical staff members and clinical research coordinators at the participating centers, especially Kaori Nagashima (clinical research coordinator of Miyazaki Medical Association Hospital), who assisted in creating the study protocol and gaining approval from each institution’s ethics committee. The authors also express special gratitude to Prof. Giancarlo Biamino, MD, PhD, for his great help in the preparation of this report.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- ankle-brachial index
- critical limb ischemia
- delayed wound healing
- endovascular therapy
- limb salvage
- pedal artery angioplasty
- Rutherford class
- skin perfusion pressure
- University of Texas
- Received July 20, 2016.
- Revision received October 3, 2016.
- Accepted October 20, 2016.
- American College of Cardiology Foundation
- Anderson J.L.,
- Halperin J.L.,
- Albert N.M.,
- et al.
- Shiraki T.,
- Iida O.,
- Takahara M.,
- et al.
- Nakama T.,
- Watanabe N.,
- Kimura T.,
- et al.
- Kawarada O.,
- Sakamoto S.,
- Harada K.,
- et al.
- Palena L.M.,
- Manzi M.
- Kobayashi N.,
- Hirano K.,
- Nakano M.,
- et al.