2017 Volume 81 Issue 7 Pages 993-998
Background: To compare the clinical outcomes of patients undergoing repeated drug-coated balloon (DCB) treatment for femoropopliteal (FP) DCB restenosis with those of patients without repetition-DCB.
Methods and Results: From March 2013 to September 2014, 102 patients (118 affected legs) underwent DCB for symptomatic FP disease; 47 patients had restenosis, and 37 underwent reintervention over a 45-month follow-up. We compared the outcomes of repetition-DCB for DCB restenosis with those of patients without repetition. The baseline patient and lesion characteristics were similar between groups. The mean lesion length was 200.8±113.1 and 195.2±134.6 mm, P=0.894, respectively. In addition, the procedural and follow-up outcomes were not different. The rates of freedom from binary restenosis (70% vs. 14%, P=0.001) and clinically driven target lesion revascularization (CD-TLR) (78% vs. 38%, P=0.026) at 1 year were statistically different between groups. Cox regression analysis showed that repetition of DCB was the only predictor for freedom from binary restenosis (hazard ratio [HR]: 6.15, 95% confidence interval (CI) 1.60 to 23.6, P=0.008) and CD-TLR (HR: 5.37, 95% CI 1.32–22.0, P=0.019).
Conclusions: For FP DCB restenosis, repetition of DCB can potentially improve vessel patency and significantly reduce the need for reintervention compared with conventional treatment. However, these observations require further confirmation in larger scale studies.
Endovascular therapy (EVT) is the treatment of choice for most patients with symptomatic peripheral arterial disease (PAD).1 Plain old balloon angioplasty (POBA) with or without stent placement in long femoropopliteal (FP) lesions is associated with high rates of restenosis.2 Several randomized controlled trials have shown promising results using drug-coated balloons (DCB) without the requirement of metallic implants to treat symptomatic FP disease.3–6 However, Schmidt et al reported 2-year results of DCB for complex FP lesions with a higher rate of late catch-up restenosis.7 Iida et al reported that up to 50% of cases required repeated EVT (re-EVT) more than once during the 2-year follow-up among patients who had reintervention for drug-eluting stent (DES) restenosis.8 Although DCB is effective in the treatment of FP in-stent restenosis (ISR),9 information regarding the optimal strategy for FP DCB restenosis is lacking. In this study, we evaluated the efficacy and outcome of repetition of DCB for DCB restenosis during a 45-month follow-up.
The subjects for this study were derived from the Tzuchi Registry of ENDovascular Intervention for Peripheral Artery Disease (TRENDPAD), which is an ongoing, prospective, physician-initiated, single-center observational registry of patients who underwent EVT for lower limb ischemia that was started in July 2005. This database was searched to identify symptomatic patients who underwent DCB angioplasty for FP disease between March 2013 and September 2014. The details of patient enrollment and exclusion are described elsewhere.10 The local ethics committee and institutional review board approved this study (IRB 03-X27-098), and all study procedures were based on good clinical practices and the applicable laws of various governing bodies. Patients with FP restenoses after the first DCB procedure were entered into the outcome analysis to compare the efficacy between repetition (group A) and no-repetition (group B) DCB for those lesions. The indication for reintervention was clinically driven and asymptomatic patients with DCB restenosis or reocclusion did not undergo repeat revascularization. A flowchart of the study enrollment, follow-up, compliance, and analysis process is presented in Figure 1.
Flowchart of participants in a study of endovascular therapy for FP disease. DCB, drug-coated balloon; DES, drug-eluting stent; FP, femoropopliteal; re-EVT, repeated endovascular therapy.
The EVT strategy was left to the discretion of the treating physician; 3 types of DCB were used in group A, including 14 IN.PACT Admiral (3 µg/mm2, Medtronic Ireland, Galway, Ireland), 1 Lutonix (2 µg/mm2, Bard, Wexford, Ireland), and 1 RangerTM (2 µg/mm2, Boston, Wűrselen, Germany). Each treatment device uses different coating techniques and excipients. As the excipient, the Ranger DCB uses an acetyl tributyl citrate, the IN.PACT DCB uses urea and the Lutonix DCB uses a non-polymer-based polysorbate/sorbitol. The DCBs were provided in diameters of 4–7 mm and lengths of 40–150 mm. All patients received 100 mg aspirin and 300 mg clopidogrel before EVT. Unfractionated heparin (5,000–10,000 units) was administered during the procedure to maintain an activated coagulation time around 250 s. After crossing the lesion with a guidewire, predilatation with an undersized, shorter, uncoated balloon (0.5–1.0 mm smaller) was performed, followed by insertion of a DCB of appropriate size and length (balloon/vessel diameter ratio of 1:1). In cases of lesion length >15 cm, ≥2 DCBs were used with a minimum 5-mm balloon overlap at the edges. The inflation time of the DCBs was 180 s using nominal pressure to allow full drug elution for the IN.PACT Admiral and 120 s for the Lutonix and Ranger DCBs. Following repeat EVT (re-EVT), aspirin was continued indefinitely in all patients and clopidogrel was used for 3 months. The details of quantitative vascular angiography have been described previously.10
For group B, we performed POBA alone. Bailout bare metal nitinol stents (BNS) were implanted in cases of suboptimal angiographic result or flow-limiting dissection, determined by residual diameter stenosis (DS) >50%, and translesional pressure gradient ≥10 mmHg.
Binary restenosis was defined as DS >50% by angiography or peak systolic velocity ratio ≥2.4 determined by duplex ultrasound (DUS). We defined the clinically driven target lesion revascularization (CD-TLR) as reintervention performed for >50% DS within 5 mm of the target lesion after documentation of recurrent clinical symptoms after the index procedure. Risk stratification was based on the FeDCLIP (female, dialysis, critical limb ischemia, lesion length >150 mm, and poor runoff) score. The FeDCLIP score has been used for vessel patency and mortality risk stratification after superficial femoral artery EVT.11 A lesion length >150 mm was scored as 2 points, whiles female, diabetes, dialysis, critical limb ischemia and poor runoff scored 1 point each. Scores of 0–2, 3–4, and ≥5 points were classified as low-, moderate- and high-risk groups, respectively. We defined re-re-EVT as more than 1 reintervention for DCB restenosis. The primary outcome measure was the rates of freedom from binary restenosis and CD-TLR at 1 year between groups.
Follow-upVessel patency after re-EVT was regularly assessed during the follow-up period with clinical examination and noninvasive studies, including ankle or toe brachial pressure index and DUS at 1 week, 1 month and thereafter every 3 months after the re-EVT or clinical recurrence. Re-re-EVT was performed if recurrent symptoms, significant vessel stenosis (≥70%) with dampened Doppler waveform patterns by DUS, and an ankle-brachial index decrease ≥0.15 were observed. The main events (death, amputation, CD-TLR, or other vascular events) were documented at discharge and follow-up visits. If office follow-up visits were not feasible, alternative data sources included telephone interviews, medical records, local electronic medical database, and the referring physician.
Statistical AnalysisAll continuous data are expressed as mean±standard deviation and were analyzed using independent t-tests. A frequency comparison was performed using the chi-square or Fisher’s exact test. The rates of freedom from binary restenosis and CD-TLR for both groups were assessed using Kaplan-Meier curves and compared with the log-rank test. Multivariate analysis was performed using Cox proportional hazards regression, entering clinically and anatomically relevant variables to determine the independent predictors for binary restenosis and CD-TLR. All statistical analyses were conducted with the SPSS statistical package for Windows version 21.0 (SPSS, Chicago, IL, USA). A probability value <0.05 was considered statistically significant.
During the study period 47 patients had DCB restenosis, and 90% of them were in the moderate- to high-risk group based on the FeDCLIP score. After excluding 7 asymptomatic patients and 3 who underwent Viabahn (Gore, Flagstaff, AZ, USA) stent graft treatment, the remaining 37 patients were divided into 2 groups based on whether DCB was repeated. Table 1 summarizes the baseline patient and lesion characteristics of the 2 groups. No significant differences were found regarding age, sex, underlying comorbidities, and clinical presentation. The ratios of severe calcification, vessel occlusion, lesion location, and distal runoff were similar between groups. In addition, risk stratification, determined by FeDCLIP score, was similar between groups.
| Group A | Group B | P value | |
|---|---|---|---|
| No. of patients | 16 | 21 | |
| Age (years) | 70±11 | 68±11 | 0.569 |
| Male sex | 8 (50%) | 9 (43%) | 0.666 |
| Diabetes mellitus | 13 (81%) | 19 (90%) | 0.416 |
| Hypertension | 14 (88%) | 20 (95%) | 0.393 |
| CAD | 5 (31%) | 12 (57%) | 0.117 |
| CVD | 3 (19%) | 4 (19%) | 0.982 |
| CKD | 7 (44%) | 9 (43%) | 0.957 |
| Dialysis | 4 (22%) | 12 (57%) | 0.051 |
| CHF | 1 (17%) | 4 (19%) | 0.259 |
| Smoking | 8 (50%) | 11 (52%) | 0.886 |
| Hyperlipidemia | 9 (56%) | 12 (57%) | 0.957 |
| BMI (kg/m2) | 24.0±4.23 | 23.5±3.09 | 0.672 |
| Hematocrit (mg/dL) | 35.0±5.10 | 35.0±6.01 | 0.979 |
| CRP (mg/dL) | 1.36±2.48 | 1.48±2.92 | 0.897 |
| HbA1C (%) | 7.8±1.4 | 7.6±1.5 | 0.741 |
| LDL-C (mg/dL) | 88±24 | 89±31 | 0.870 |
| Albumin (mg/dL) | 3.30±0.59 | 3.11±0.52 | 0.328 |
| No. of affected limbs | 16 | 21 | |
| Target-limb ABI | 0.55±0.17 | 0.59±0.30 | 0.650 |
| Intermittent claudication | 10 (62%) | 10 (48%) | 0.368 |
| Rest pain | 2 (13%) | 2 (10%) | 0.773 |
| Non-healing ulcer | 4 (25%) | 8 (37%) | 0.399 |
| Gangrene | 0 (0%) | 1 (5%) | 0.376 |
| Combined BTK treatment | 10 (63%) | 15 (71%) | 0.565 |
| ≤1 vessel BTK runoff | 10 (63%) | 16 (76%) | 0.367 |
| ≥2 vessels BTK runoff | 6 (37%) | 5 (24%) | |
| Severe calcification | 4 (25%) | 8 (38%) | 0.213 |
| Occlusion | 5 (31%) | 11 (52%) | 0.209 |
| Location of treated lesions | 43 | 53 | |
| CFA | 1 (2.3%) | 2 (3.7%) | 0.685 |
| Proximal SFA | 10 (23.3%) | 7 (13.2%) | 0.200 |
| Middle SFA | 9 (20.9%) | 8 (15.1%) | 0.456 |
| Distal SFA | 13 (30.2%) | 18 (34.0%) | 0.698 |
| Popliteal artery | 10 (23.3%) | 18 (34.0%) | 0.251 |
| FeDCLIP risk group | 16 | 21 | |
| Low | 7 (44%) | 6 (29%) | 0.338 |
| Moderate | 5 (31%) | 11 (52%) | 0.199 |
| High | 4 (25%) | 4 (19%) | 0.663 |
Values are mean±SD or n (%). P<0.05 indicates a significant difference between groups. ABI, ankle-brachial index; BMI, body mass index; BTK, below-the-knee; CAD, coronary artery disease; CFA, common femoral artery; CKD, chronic kidney disease; CHF, congestive heart failure; CRP, C-reactive protein; CVA, cerebrovascular accident; FeDCLIP, (female, dialysis, critical limb ischemia, lesion length, runoff); HbA1C, glycated hemoglobin; LDL-C, low-density lipoprotein cholesterol; SFA, superficial femoral artery.
Table 2 shows the procedural and follow-up results. The mean lesion length in groups A and B was 200.8±113.1 and 195.2±134.6 mm, P=0.894, respectively. Both groups had a similar ratio of bailout stenting (44% vs. 52%, P=0.602), long lesion (≥15 cm) (56% vs. 48%, P=0.843), in-stent lesion (44% vs. 42%, P=0.792), and mean stent length (160.0±84.9 vs. 170.9±89.9 mm, P=0.801). The final minimal lumen diameter of the superficial femoral artery (5.02±0.65 vs. 5.16±0.80 mm, P=0.505), or popliteal artery (4.22±0.64 vs. 4.35±0.62 mm, P=0.603) and the final balloon size (5.56±0.63 vs. 5.76±0.83 mm, P=0.429) were similar between groups. The duration of the first DCB procedure to re-EVT and total follow-up period did not differ between groups (410±200 vs. 400±233 days, P=0.757 and 925±255 vs. 939±285 days, P=0.872, respectively). No patient died in either group; 2 and 3 patients in groups A and B, respectively, underwent planned minor amputation.
| Group A | Group B | P value | |
|---|---|---|---|
| No of affected limbs | 16 | 21 | |
| Mean lesion length (mm) | 200.8±113.1 | 195.2±134.6 | 0.894 |
| <150 mm | 7 (44%) | 11 (52%) | 0.603 |
| ≥150 mm | 9 (56%) | 10 (48%) | 0.843 |
| Additional stenting | 7 (44%) | 11 (52%) | 0.602 |
| Additional stent length (mm) | 160.0±84.9 | 170.9±89.9 | 0.801 |
| Atherectomy | 1 (6.3%) | 1 (4.8%) | 0.843 |
| Use of IVUS | 6 (38%) | 9 (43%) | 0.742 |
| Treated lesions | 43 | 53 | |
| ISR lesions | 19 (44%) | 22 (42%) | 0.792 |
| Non-ISR lesions | 24 (56%) | 31 (58%) | |
| RVD at distal SFA (mm) | 5.34±0.70 | 5.51±1.14 | 0.508 |
| RVD at popliteal artery (mm) | 4.59±0.71 | 4.79±0.95 | 0.545 |
| Final balloon size (mm) | 5.56±0.63 | 5.76±0.83 | 0.429 |
| Final MLD at SFA (mm) | 5.02±0.65 | 5.16±0.80 | 0.505 |
| Final MLD at popliteal artery (mm) | 4.22±0.64 | 4.35±0.62 | 0.603 |
| Total follow-up period (days) | 925±255 | 939±285 | 0.872 |
| Time to re-EVT (days) | 410±200 | 400±233 | 0.757 |
| Follow-up time after re-EVT (days) | 515±251 | 539±226 | 0.893 |
| Death | 0 | 0 | |
| Planned minor amputation | 2 (13%) | 3 (14%) | 0.875 |
ISR, in-stent restenosis; IVUS, intravascular ultrasound; MLD, minimal lumen diameter; re-EVT, repeated endovascular therapy; RVD, reference vessel diameter; SFA, superficial femoral artery.
The 1-year outcomes after re-EVT for DCB restenosis were significantly different between the groups. The rates of freedom from binary restenosis and CD-TLR in group A were much higher compared with group B (70% vs. 14%, P=0.001 and 78% vs. 38%, P=0.026, respectively) (Figure 2A,B). Three patients in group A failed to sustain vessel patency during the follow-up: 1 patient underwent iliac stenting for a new lesion and repeat DCB for focal stenosis in the proximal superficial femoral artery, and bypass surgery was suggested for the other 2 patients because of recurrent diffuse stenosis despite DCB treatment.
Kaplan-Meier curves for freedom from binary restenosis and clinically driven target lesion revascularization (CD-TLR). (A) Compared with group B (dotted line), group A (solid line) has a higher binary restenosis-free rate at 1 year (70% vs. 14%, P=0.001). (B) Compared with group B (dotted line), group A (solid line) has a higher CD-TLR-free rate at 1 year (78% vs. 38%, P=0.026). SE, standard error.
Cox regression analysis was performed to identify independent predictors of binary restenosis and CD-TLR after re-EVT for DCB restenosis. The previously reported factors affecting patency and restenosis rate, including age, sex, diabetes mellitus, critical limb ischemia, dialysis, hyperlipidemia, vessel calcification, occlusion, lesion length, C-reactive protein level, and repetition of DCB were analyzed. Only the repetition of DCB was an independent predictor for freedom from binary restenosis (hazard ratio [HR]: 6.15, 95% confidence interval (CI) 1.60–23.6, P=0.008) and CD-TLR (HR: 5.37, 95% CI 1.32–22.0, P=0.019) (Table 3).
| Factor | Binary restenosis | P value | CD-TLR | P value |
|---|---|---|---|---|
| HR (95% CI) | HR (95% CI) | |||
| Age | 0.96 (0.90–1.04) | 0.311 | 0.97 (0.89–1.04) | 0.384 |
| Female | 1.59 (0.54–4.73) | 0.404 | 2.03 (0.63–6.57) | 0.237 |
| Diabetes mellitus | 2.49 (0.33–18.8) | 0.378 | 2.46 (0.31–19.2) | 0.391 |
| Dialysis | 1.12 (0.18–7.03) | 0.891 | 2.20 (0.30–16.2) | 0.438 |
| CLI | 1.51 (0.31–7.40) | 0.611 | 1.12 (0.21–6.02) | 0.898 |
| Hyperlipidemia | 1.79 (0.47–6.80) | 0.388 | 1.14 (0.26–5.08) | 0.860 |
| Vessel calcification | 1.43 (0.23–9.06) | 0.702 | 1.16 (0.16–8.53) | 0.887 |
| Occlusion | 1.18 (0.34–4.11) | 0.796 | 1.39 (0.37–5.29) | 0.629 |
| Lesion length | 1.03 (0.98–1.08) | 0.281 | 1.02 (0.97–1.08) | 0.364 |
| Level of CRP | 1.32 (0.81–2.14) | 0.268 | 1.63 (0.67–3.95) | 0.280 |
| Repetition of DCB | 6.15 (1.60–23.6) | 0.008* | 5.37 (1.32–22.0) | 0.019* |
P<0.05 indicates a significant difference between groups. CD-TLR, clinically driven target lesion revascularization; CLI, critical limb ischemia; CRP, C-reactive protein; DCB, drug-coated balloon.
This study is the first to report an outcome analysis of repeated paclitaxel-coated balloon (PCB) vs. conventional treatment in FP DCB restenosis. We demonstrated that repetition of DCB angioplasty was effective for reducing the rates of binary restenosis and CD-TLR compared with conventional treatment.
For patients with symptomatic FP disease, DCB therapy has been shown to reduce restenosis rates, eliminating the need for a permanent metallic implant, as compared with POBA. Promising results have been proved in randomized controlled trials in both short-term and 2-year reports.3–6,12 However, DCB gradually loses its antiproliferative effect when facing complex FP lesions. Schmidt et al reported that the 2-year primary patency and CD-TLR-free rates for complex FP lesions (mean lesion length 24.0±10.2 cm) were 53.7% and 68.6%, respectively.7 Re-EVT in 30% of the participants in Schmidt’s study was necessary to maintain vessel patency, but the outcome and treatment strategy in cases of re-EVT have not been reported.7
Iida et al reported their 1-year results of re-EVT for DES-ISR of the FP artery. They showed a higher rate of restenosis after reintervention regardless of focal, diffuse or occluded lesions (53%, 74%, and 78%, respectively).13 Moreover, approximately 50% re-EVT cases required reintervention more than once during a 2-year follow-up.8 So far, the optimal reintervention strategy for DCB restenosis in FP disease remains uncertain, and outcome analysis has never been reported.
In our study, 47 patients had DCB restenosis during the follow-up, and 90% of them had moderate-to-high-risk FeDCLIP scores at first DCB angioplasty. The lesion morphology of DCB restenosis was 40%, 22%, and 38% short stenosis, diffuse stenosis, and vessel occlusion, respectively. This finding was close to that in previous studies,7,13 wherein restenosis after drug-eluting devices involved shorter and focal lesions. In the present 37 patients undergoing re-EVT, the 2-year primary patency was 58%, which was similar to a previous report.7 As well, we showed that the results of POBA or BNS for DCB restenosis were discouraging for either ISR or non-ISR lesions. The rates of freedom from binary restenosis and CD-TLR at 1 year were 14% and 38%, respectively. In contrast, we demonstrated the superiority of repeated DCB in managing DCB restenosis. The antiproliferative effect not only reduces the recurrent late lumen loss at the non-stented site, but also prevents the need for multilayer stenting of the stented site, which might compromise lumen area and increase the potential for restenosis.
Based on the experience of coronary intervention, PCB is equally effective in treating paclitaxel- or limus-eluting ISR.14 Moreover, the efficacy of the same drug platform has not been reported in endovascular intervention, and the limus-coated balloon catheter is not available in current practice. This study showed that repeat PCB angioplasty was feasible and efficient for those lesions already treated by the same drug.
Previous reports suggest that several variables (female, obesity, lesion length >15 cm, and vessel calcification) are predictors for restenosis after the first DCB treatment.7,10 After multivariable regression analysis, repetition of DCB was the only significant variable predicting restenosis and CD-TLR rates of re-EVT for DCB restenosis in this study. Temporal presentation, underlying mechanism, morphologic patterns, and tissue composition of the DCB restenosis might be different from those in de novo or BNS restenosis. These factors made POBA or BNS less efficient for treating FP DCB restenosis. Further imaging studies, such as intravascular ultrasound or optimal coherence tomography, might clarify DCB restenosis morphology and be beneficial for the decision strategy.
Study LimitationThis study has several limitations that remain to be addressed. First, the small sample size in this retrospective observational analysis makes a robust conclusion impossible. However, DCB reduced the need for re-EVT, and thus, case reduction can be expected, particularly for shorter lesions or in patients with low FeDCLIP scores. Second, wide use of DCBs for all patients with PAD was not plausible for financial reasons. Third, single-institution series are often biased towards particular patient demographics and practice patterns. Fourth, we did not routinely perform intravascular ultrasound and follow-up angiography. Thus, detailed morphology of each vessel segment was unavailable. Finally, we excluded 3 patients undergoing covered stents for DCB restenosis in the conventional treatment group because covered stent plays a role in endoluminal bypass and has a different mechanism of restenosis compared with POBA and BNS.
For FP DCB restenosis, repetition of DCB can potentially improve vessel patency and significantly reduce the need for reintervention compared with conventional treatment. However, these observations require further confirmation in larger scale studies.
The authors thank the cardiac catheterization laboratory medical staff and clinical research coordinators for participating in this study.
Nothing to declare.
