Clinical outcome of drug-coated balloon angioplasty in patients with femoropopliteal disease: A real-world single-center experience

Abstract

Background

Several multicenter industry-sponsored clinical trials reported satisfactory results in the use of drug-coated balloons (DCBs) for treatment of femoropopliteal occlusive disease. However, few single-center studies have been published to verify the outcome from real-world experience.

Methods

In thls study, 228 patients treated with DCB angioplasty (Lutonix 0.35; Bard. Tempe, Arizona) were analyzed. Perioperative major adverse events (death, amputation, target lesion thrombosis or reintervention) were calculated. Kaplan-Meier analysis was used to estimate primary patency rates (based on duplex ultrasound with or without ankle-brachial index) and limb salvage rates.

Results

Lesions treated were primarily TransAtlantic Inter-Society Consensus (TASC) type C and D lesions. Indications included claudication (Rutherford classes 2 and 3) in 40% and critical limb ischemia (CLI; Rutherford classes 4 and 5) in 60%. Lesions treated included 61% in the superficial femoral artery.15% in the popliteal artery, and 24% In both superficial femoral artery and popliteal artery. Mean follow-up was 12.2 months (range, 1–42 months). Overall perioperative morbidity and mortality rates were 13% and 1%. The perioperative major adverse event rate was 3%. Symptom relief (improvement of one Rutherford category or more) was obtained in 64%. Primary patency rates were 56% and 39% at 1 year and 2 years, respectively. Limb salvage rates were 92% and 83% at 1 year and 2 years. Patients with claudication had a lower rate of early perioperative complications (4% vs 19%; P = .001). Symptom improvement was 76% for claudication vs 49% for CLI (P < .001). Overall, major amputation rate was 0% for claudication vs 13% for CLI (P < .001). The primary patency rates at 1 year and 2 years were 59% and 41% for claudicatlon vs 54% and 37% for CLI (P = .307). The assisted primary patency rates at 1 year and 2 years were 72% and 52% for claudication vs 64% and 46% for CLI (P = .223). Primary patency rates at 1 year and 2 years were 82% and 71% for TASC A to C lesions vs 29% and 14% for TASC D lesions (P < .001). Limb salvage rates at 1 year and 2 years were 100% and 100% for claudication vs 85% and 74% for CLI (P < .001).

Conclusions

Clinical outcomes after DCB angioplasty In femoropopliteal lesions were inferior to what has been reported in previous studies. particularly for TASC D lesions. Further investigation from real-world experience with long-term follow-up is needed to confirm these results. (J Vasc Surg 2019:70:1950–9.)

Drug-coated balloons (DCBs) were developed to improve the durability of plain balloon angioplasty without the potential drawbacks of stenting. Although stents were originally advocated to prevent the elastic recoil of arteries after percutaneous transluminal angioplasty (PTA), stents have been hypothesized to increase neointimal hyperplasia by inducing trauma and increasing inflammatory reactlon. DCBs are designed to minimize neointimal hyperplasia by delivering large doses of an anti-cell-proliferating agent directly to the site of the injured vessel.^1,2

Several published and ongoing studies have reported on paclitaxel DCBs in both de novo and restenotic femoral-popliteal lesions with somewhat satisfactory results. Two DCBs, Lutonix 0.35 (Bard, Tempe, Arizona) and IN.PACT Admiral (Medtronic. Dublin, Ireland), were approved for use in the United States in the past few years.^3,4 Recently, Stellarex DCB (Philips, Amsterdam, The Netherlands) was also approved.⁵ Most of the clinical trial data on DCBs have mainly included patients with claudication and lower Rutherford classes as opposed to patients with critical limb ischemia (CLI). These multicenter clinical trials are industry sponsored and included the Lutonix Paclitaxel-Coated Balloon for the Prevention of Femoropopliteal Restenosis (LEVANT) 1, Local Taxane with Short Exposure for Reduction of Restenosis in Distal Arteries (THUNDER), Femoral Paclitaxel (FemPac), and Paclitaxel-coated Balloons in Femoral Indication to Defeat Restenosis (PACIFIER), which evaluated femoropopliteal lesions comparing DCBs with Standard PTA. More than 90% of enrolled patients in these trials had intermittent claudication (Rutherford class 3 or lower). They concluded that late lumen loss, target lesion revascularization, and restenosis significantly favored DCBs over Standard PTA.^3,6–8 There are no single-center studies published to verify the outcome from real-world experience specifically using the Lutonix 0.35 drug-eluting balloons in treatment of femoropopliteal disease. Therefore, this study analyzed the early and intermediate clinical outcome with the use of these DCBs in our medical center.

METHODS

This is a retrospective analysis of all patients undergoing endovascular intervention for femoropopliteal occlusive disease with Lutonix 0.35 DCB angioplasty who were treated between May 20, 2010, and December 19, 2017. The study was approved by the Institutional Review Board of Charleston Area Medical Center/West Virginia University and informed consent was not required. Patients’ demographics and clinical characteristics, preoperative and postoperative ankle-brachial index (ABI), and duplex ultrasound (DUS) findings were analyzed. Lesions were classified according to TransAtlantic Inter-Society Consensus (TASC) II classification (A, B. C, and D). Indications for intervention were lifestyle-limiting claudication (Rutherford classes 2 and 3) or CLI (Rutherford classes 4 and 5). Patients with no infrapopliteal runoff vessels or of limited life expectancy (<2 years), presence of acute thrombosis in the target vessel, prior vascular surgery. or thrombolysis within the past 6 weeks were exclusions for analysis.

Although there was no specific protocol for which lesions were to be treated by which therapy. our group generally tends to treat most if not all TASC A and B lesions with simple balloon angioplasty or stenting and most if not all C and D lesions with a drug-eluting balloon. During the period of this study, the mean number of endovascular therapy procedures (PTA and non-DCB or others) was 366 per year.

All procedures were performed in our circulatory dynamics laboratory (fixed imaging system) by our vascular surgeons or vascular interventionalists. Arterial access was performed through the contralateral femoral artery by a retrograde approach and in a few patients by an ipsilateral femoral antegrade or pedal approach using a 5F or 6F sheath. Most lesions were crossed with a 0.035-inch wire: for complete occlusions, a soft 0.035-inch Clidewire (Terumo Interventional Systems, Somerset, NJ) and a 4F or 5F guide catheter (Terumo) were used to cross in a subintimal plane, and if needed, a re-entry device (Outback [Cordis, Bridgewater, NJ] or Pioneer Plus re-entry catheter or Quick-Cross [Philips]) was employed to gain access to the lumen. Balloon diameter was chosen on the basis of the angiographic measurement of the nondiseased segment both proximal and distal to the lesion. After standard PTA of the lesion, DCBs were used and nominal inflations were maintained for about 2 minutes according to DCB instruction. Stents were used only if residual stenosis >30% or flow-limiting dissection after DCB angioplasty was noted. Most patients did not have vessel preparation before the use of DCB other than prolonged angioplasty (about 3 minutes), and this was left to the physician’s discretion. Some of these patients underwent debulking for calcified lesions with an atherectomy device. All patients were given a loading dose of 300 mg of clopidogrel immediately after the procedure in combination with 325 mg of aspirin. The clopidogrel was maintained as 75 mg daily for a minimum of 6 to 8 weeks. However, aspirin was continued indefinitely.

Clinical and hemodynamic follow-up

Every effort was made to observe these patients at the Vascular Center of Excellence clinic at 30 days and at 6- and 12-month intervals. The follow-up visit also included a noninvasive vascular laboratory study including ABI or color DUS examination of the treated limb. Patients who had symptom recurrence or hemodynamic failure as defined by a decrease of ABI by >0.15 underwent DUS, computed tomography angiography, or catheter angiography if reintervention was anticipated.

All noninvasive vascular laboratory testing was done in our Intersocietal Accreditation Commission-accredited vascular laboratory by registered vascular technologists. The ABI was calculated using the highest compressible ipsilateral ankle pressure from either the dorsalis pedis or posterior tibial artery divided by the highest brachial pressure. The DUS studies were performed using a linear array transducer with pulse Doppler frequencies between 3 and 10 MHz, and the Doppler signal was acquired at an angle ≤60 degrees with small sample size. DUS-derived peak systolic velocities (PSVs) of the lower extremity arteries were recorded from the common femoral artery, superficial femoral artery (SFA), popliteal artery, and runoff. Doppler spectral analysis determined the highest PSV in the treated segment, and the PSV ratio was calculated by dividing the highest PSV in the treated lesion by the PSV just proximal to the lesion. Follow-up data were collected from the electronic medical records. All death data were verified from the Social Security Death Index.

Definitions and end points

Technical success was achieved when treated lesions had <30% residual stenosis on completion angiography. Improvement in symptoms or symptom relief was used for freedom from symptoms or improvement of symptoms by one Rutherford category. Major adverse event was defined as a composite of death, major amputation, site thrombosis, or target lesion revascularization. Because most published articles on this subject used this composite end point, it was thought that it would be more appropriate to use the same end point to compare the result in our study with that of others.

Primary patency was defined as the absence of ≥50% restenosis or occlusion in the treated arterial segment according to DUS or angiography. This was defined by a decrease of the ABI by >0.15 or evidence of stenosis by DUS of PSV >300 cm/s or a ratio >3.⁹ Assisted primary patency was defined as restored patency after endovascular reintervention (PTA or stenting) for restenotic lesions. Limbs that underwent vascular bypass after occlusion of the treated segment were not considered patent. Target lesion revascularization was done for recurrent stenosis of >50% that was associated with recurrence of symptoms.

Statistical Analysis

The data analysis was performed using SAS 9.3 software (SAS Instìtute, Cary, NC). Basic descriptive statistics, such as means and standard deviations for continuous variables and proportions and frequencies for categorical variables, was used to analyze the data. Comparison of categorical variables was performed using contingency table analysis with a χ² test to determine statistically significant differences.

Kaplan-Meier analysis was used to estimate freedom rates from primary patency. assisted patency, and limb salvage. Comparison between two survival distributions was based on the log-rank test. The threshold for significance was .05.

RESULTS

This study analyzed 228 patients with a mean age of 68 years who were treated with Lutonix 0.35 drug-eluting balloon angioplasty of femoropopliteal lesions. Table I summarizes the demographics and clinical characteristics of these patients. As noted, 91 patients (40%) were treated for claudication (Rutherford classes 2 and 3). and 137 patients (60%) were treated for CLI (Rutherford classes 4 and 5); 138 patients were treated for SFA lesions (61%). 35 for popliteal artery lesions (15%), and 55 for combined femoropopliteal lesions (24%).

Table 1.

Demographics and clinical characteristics of the patients

Variable	No.	Mean	SD	Minimum	Maximum
Age, years	228	68.1	10.39	27	97
	Frequency	%
Male	104	46
Female	124	54
Diabetes mellitus	120	53
Hypertension	165	72
Hypercholesterolemia	136	60
Chronic kidney disease	55	24
Coronary artery disease	124	54
Tobacco use	62	27
Indications
Claudication (Rutherford classes 2 and 3)	91	40
CLI (Rutherford classes 4 and 5)	137	60
Location of primary treated lesion
Femoral arteries	138	61
Popliteal arteries	35	15
Combined femoral and popliteal	55	24

CLI, Critical limb ischemia; SD, standard deviation.

Table II summarizes the demographic and clinical characteristics according to indication of intervention (based on Rutherford classification). As noted in Table II. there were more femoral arteries treated in patients with claudication (70% vs 54% in patients with CLI; P = .038).

Table II.

Demographics and clinical characteristics according to indications and Rutherford class

Age, years	No.	Mean	SD	Minimum	Maximum	P value
Claudication (Rutherford classes 2 and 3)	91	65.2	8.06	45	84	.0168
CLI (Rutherford classes 4 and 5)	137	68.3	11.5	27	97
	Claudication (n = 91), No. (%)	CLI (n = 137), No. (%)	P value
Sex
Male	46 (51)	58 (42)	.2227
Female	45 (49)	79 (58)
Diabetes mellitus	44 (48)	76 (55)	.2915
Hypertension	64 (70)	101 (73)	.5748
Hypercholesterolemia	53 (58)	83 (61)	.7241
Chronic kidney disease	16 (18)	39 (28)	.0599
Coronary artery disease	52 (57)	72 (53)	.4958
Tobacco use	29 (32)	33 (24)	.196
Location of primary treated lesion
Femoral arteries	64 (70)	74 (54)	.0382
Popliteal arteries	12 (13)	23 (17)
Combined fem and pop	15 (16)	40 (29)

CLI, Critical limb ischemia; SD, standard deviation.

The mean length of all lesions was 20.2 cm (median, 19 cm; range, 3–45 cm). There were 138 limbs with a lesion ≥5 cm (47 of these were >30 cm); 119 lesions were stenotic lesions and 84 were total occlusion. The mean length for stenotic lesions was 16.9 cm (median, 15 cm; range. 3–40 cm) vs 24.9 cm for total occlusion lesions (median, 25 cm; range. 5–45 cm; P < .001). Based on TASC II classification, 5 lesions were TASC A (2.5%), 19 TASC B (9.3%), 87 TASC C (43%), and 92 TASC D (45%). All lesions were de novo (98%) lesions except in four limbs, in which both de novo and in-stent stenotic lesions were treated on the same limb; 27% of patients had one-vessel runoff, 67% had two-vessel runoff, and 6% had three-vessel runoff. All lesions were accessed through a retrograde approach (98.5%) except for three lesions, for which an antegrade approach or pedal access was used.

Early 30-day perioperative outcome

The overall technical success rate was 98.5%. Only three limbs (D lesions) failed technically because of inability to cross total occlusion. Overall, 136 patients (64%) had relief or improvement of the preoperative symptoms (76% in the claudication group vs 49% in the CLI group: P < .001). The mean preoperative ABI was 0.59 (range. 0–1.5), and the mean postoperative ABI was 0.82 (0–1.2; P < .001).

Tables III and IV summarize the perioperative complications and outcomes in patients according to indication. As noted in Table IV. there was a significantly higher early amputation rate in patients with CLI vs claudication (11.7% vs 1.1%; P = .003). One patient with claudication had a toe amputation vs 16 patients with CLI, 2 of which were major amputations. Overall, patients with CLI have a higher incidence of early perioperative complications (18.9% vs 4.4% for patients with claudication; P = .001); 3.1% of patients had a major adverse event (two deaths, two major amputations, one minor amputation in a patient with claudication, and two reinterventions for perioperative thrombosis).

Table III.

Early 30-day perioperative outcomes

	Frequency (N = 228)	%
Early amputation	17	7.5
Level of early amputation
Toe	13	5.7
Transmetatarsal amputation	2	0.9
Below-knee amputation	1	0.4
Above-knee amputation	1	0.4
Puncture site hematoma or bleeding	4	1.8
Early death	2	0.9
Myocardial infarction/death	4	1.8
Other perioperative eventsa	14	6.1
30-Day major adverse eventsb	7	3.1
All early perioperative complications including deathc	30	13.2

^aThese include four femoral pseudoaneurysms, three cases of acute renal failure, two non-ST-segment elevation myocardial infarctions, two extensive dissections, one arterial embolism, one atrial fibrillation, and one Perclose occlusion.

^bMajor adverse events included death, amputations, and drug-eluting balloon site early thrombosis.

^cSome may have more than one event.

Table IV.

Early 30-day perioperative outcome according to indications

	Claudication (n = 91), No. (%)	CLI (n = 137). No. (%)	Total	P value
Early amputation	1 (1.1)	16 (11.7)		.0029
Level of early amputation
Toe	1 (1.1)	12 (8.8)	13
Transmetatarsal amputation	0 (0)	2 (1.5)	2
Below-knee amputation	0 (0)	1 (0.7)	1
Above-knee amputation	0 (0)	1 (0.7)	1
Perioperative bleeding	1 (1.1)	3 (2.2)	4	1
Early death	0 (0)	2 (1.5)	2	.5182
Myocardial infarction/death	1 (1.1)	3 (2.2)	4	1
Other perioperative events	2 (2.2)	12 (8.8)	14	.0433
Early major adverse event	1 (1.1)	6 (4.4)	7	.2477
All early perioperative complications	4 (4.4)	26 (18.9)	30	.0014

Late clinical outcome

The mean follow-up for patients was 12.2 months (median, 10 months; range, 1–42 months). There were 20 patients who had no follow-up beyond the day of intervention. At late follow-up, 91 of 208 (44%) lesions had failed. Overall, 51 patients had restenosis or thrombosis of the treated site that needed reintervention (these included repeated ballooning with conventional balloon angioplasty. stenting, DCBs, and drug-eluting stents; atherectomy; surgical bypasses), 15 (16.5%) in the claudication group vs 36 (26%) of the CLI group (P = .082). Some of these patients needed more than one type of reintervention in the same setting. There were a total of 29 late amputations, 2 in patients with claudication (toe amputation) (2%) and 27 in the CLI group (16 [19.7%] of these were major amputations; P < .001). There were 19 late deaths, 6 (6.6%) in the claudication group and 13 (9.5%) in the CLI group (P = .439). Late symptom relief or improvement was noted in 77% in the claudication group vs 56% in the CLI group (P = .001).

The mean length for failed lesions was 25.1 cm (range, 5–45 cm) vs a mean of 16.7 cm (range, 3–45 cm) for patent lesions (P < .001). Overall, stenotic lesions had better primary patency than total occlusions (67.3% vs 39%: P < .001).

Kaplan-Meier analysis of patency rates and limb salvage

After exclusion of recurrent stenosis (four limbs) and patients who had adjunctive atherectomy for calcified lesions (nine limbs), the primary patency rates were 57% at 1 year and 39% at 2 years: 59% and 41% for the claudication group vs 54% and 37% for CLI group, respectively (P = .307: Fig 1). The primary patency for lesions that did not have adjunctive stenting at 1 year and 2 years was 51% and 41% vs 60% and 38% for lesions that had adjunctive stenting (P = .965).

Fig 1.

Overall primary patency rates: claudication vs critical limb ischemia (CLI). SE, Standard error.

The assisted primary patency rates were 68% and 48% at 1 year and 2 years: 72% and 52% for the claudication group vs 64% and 46% for CLI group, respectively (P = .223: Fig 2). The limb salvage rates were 91% and 85% at 1 year and 2 years: 100% and 100% for the claudication group vs 85% and 74% for the CLI group, respectively (P < .001; Fig 3).

Fig 2.

Overall assisted patency rates: claudication vs critical limb iischemia (CLI). SE, Standard error.

Fig 3.

Overall limb salvage rates: claudication vs critical limb ischemia (CLI). SE, Standard error.

When lesions were classified according to TASC II classification, the primary patency rates for class A to class C lesions at 1 year and 2 years were 82% and 71% in contrast to 29% and 14% for class D lesions, respectively (P < .001: Fig 4). Similar observations were noted when class A to class C lesions were compared with class D lesions according to indications (claudications vs CLI, P < .001; Fig 5).

Fig 4.

Overall primary patency rates: TransAtlantic Inter-Society Consensus (TASC) class A to class C vs class D lesions. SE, Standard error.

Fig 5.

Primary patency rates: TransAtlantic Inter-Society Consensus (TASC) class A to class C vs class D lesions: claudication vs critical limb ischemia (CLI). SE, Standard error.

DISCUSSION

Several trials that analyzed the outcome of DCBs for the treatment of SFA and popliteal lesions in patients with claudication and CLI demonstrated improved patency rates compared with conventional PTA.^3–6 However, these studies were somewhat limited by smaller sample size, heterogeneous populations of patients, and relatively shorter follow-up.

The LEVANT 1 trial randomized 101 patients to Lutonix DCB vs uncoated balloons. The primary patency rate was significantly higher with Lutonix DCB vs plain PTA at 12 months, 74% vs 57% (P < .001). It was also noted that the major adverse events as defined by death, amputation, and target lesion thrombosis or reintervention were 39% for DCB vs 46% for conventional PTA (P = .45) at 24 months.³

The largest trial using DCBs in the femoropopliteal location was the LEVANT 2 trial using Lutonix 035 DCB. There were 476 patients with Rutherford class 2, 3, and 4 disease randomly assigned to plain balloon angioplasty vs Lutonix DCB in a 1:2 ratio. This was a multicenter prospective Lutonix Global SFA Registry from a real-world population of patients. The mean total target lesion length was 101.2 ± 84 mm, 22% of the lesions were longer than 150 mm, and 31% had chronic total occlusion. The primary patency was superior in patients who had Lutonix DCB compared with conventional PTA.¹⁰ By Kaplan-Meier analysis, the primary patency rates were 85% and 76% at 12 and 24 months, respectively. The investigators also noted that 70% of patients showed significant improvement of symptoms of at least one Rutherford category.¹⁰

Our study analyzed the early and intermediate clinical outcomes with use of the Lutonix 035 DCB in treating patients with femoropopliteal vascular occlusive disease in our institution. It showed inferior overall primary patency of 57% at 1 year and 39% at 2 years. However, for TASC A to C lesions, the primary patency rates were better (82% at 1 year and 72% at 2 years), whereas these rates were unsatisfactory for D lesions (29% at 1 year and 12% at 2 years). The mean lesion length in our series was much longer (20 cm) than in both the LEVANT 1 and 2 trials; 138 lesions (66%) of the total series were >15 cm, and 47 lesions (23%) were >30 cm. In our series, 40% of lesions also had chronic total occlusion and 45% were TASC D lesions. Because of these complex lesions, 62% of our patients had adjunctive stenting after DCB use because of significant dissection that was noted primarily in patients with long C and D lesions. Patients who underwent primary balloon angioplasty and stenting in our institution during the past several years were also patients who were mainly claudicants with TASC A or B lesions. Therefore, it would be inappropriate to compare these patients to see whether the DCBs add anything to primary stenting because most of the class C and D lesions were treated with DCBs. Similarly, Scheinert et al¹¹ reported that provisional stents were implanted in 39.4% (63/160) of lesions in the DCB treatment of femoropopliteal artery disease in the long lesion imaging cohort of the IN.PACT Global Study.

Kokkinidis et al¹² also reported on adjunctive stent use in patients treated with DCBs for femoropopliteal lesions in the Excellence in Peripheral Artery Disease (XLPAD) registry. Stents were implanted in 31% of these patients. Overall, lesions treated with the stents were longer, with a mean of 150 mm vs 100 mm (P < .001). Stenting was generally more frequently used in patients with chronic total occlusion or with complex femoropopliteal lesions (66% vs 34%; P < .001). They also noted that there was no difference in periprocedural complication rates or 12-month target limb revascularization (19% vs 12%; P = .162) or 12-month amputation rates (11% vs 11%; P = .92) between lesions in which adjunctive stenting was used and lesions without stenting, respectively. However, the Lutonix balloon was used in only 26% in their study (74% were IN.PACT Admiral), and the mean length was 113.9 ± 75 mm; 22% of their lesions were in-stent restenosis, and 44% had chronic total occlusion.

A few studies, one of which is a randomized controlled trial, analyzed the result of DCB therapy for lesions longer than 20 cm.^13–16 These lesions are often heavily calcified with complete occlusions, many of which were associated with provisional stenting ranging from 10% to 50%. In most of these trials, several balloons were used to cover long and diffuse lesions, where shorter balloons may increase the chances of geographic miss, which was shown to negatively affect the outcome in the LEVANT trial.³ Dohi et al¹⁷ reported a retrospective analysis of 266 patients who were treated with DCB angioplasty for 281 de novo lesions, including the popliteal artery, with a median length of 270 mm. The primary patency was 77% at a median of 12.2 months. They thought that the outcomes after DCB angioplasty in lesions, including the popliteal artery, were acceptable compared with previous studies. However, they believed that a long-term follow-up is needed to confirm their findings. However, only 15% of their patients were treated with the Lutonix DCB and 83% with the IN.PACT DCB.

In general, the optimal therapy for patients with TASC D femoropopliteal lesions remains ill-defined. The current Society for Vascular Surgery practice guidelines for management of claudication patients recommend the following: for focal lesions (<5 cm) in the SFA that have unsatisfactory technical results with simple PTA, selective stenting is suggested; for intermediate-length lesions (5–15 cm), adjunctive use of self-expanding nitinol stents (with or without paclitaxel) to improve the midterm patency of PTA is recommended; surgical bypass is recommended as an initial strategy for patients with diffuse femoropopliteal occlusive disease (TASC D lesion), small-caliber artery (<5 mm), or extensive calcification of the SFA if they have favorable anatomy for bypass and have a low operative risk.¹⁸

A systemic review and meta-analysis of randomized controlled trials analyzing paclitaxel-coated balloon angioplasty and paclitaxel-coated stents in the femoropopliteal arterial location of 4663 patients found that all-cause death at 2 years was significantly higher in patients with paclitaxel vs control: 7.2% vs 3.8%. At 5-year follow-up, this rate was 14.7% vs 8.1%.¹⁹ This led the Food and Drug Administration to issue a warning letter about paclitaxel-coated balloons and eluting stents²⁰ and to recommend that physicians should continue to monitor patients who have been treated with these drug-eluting devices for PAD. However, a report by Secemsky et al²¹ showed that there is no evidence of increased all-cause mortality after the use of paclitaxel DCBs or drug-eluting stents in treatment of femoropopliteal artery occlusive disease compared with nondrug-coated devices. These investigators analyzed data of 16,560 patients in the U.S. Medicare database, in whìch drug-coated devices were used in 36% of patients and shown to be associated with lower cumulative incidence of all-cause mortality compared with treatment using nondrug-coated devices through 600 days (32.5% vs 34.3%: P = .007). A multivariant analysis also showed no association between drug-coated devices and all-cause mortality (P = .43).

This study has several limitations, including belng a retrospective single-center study. The study population is also heterogeneous, and there was no uniformity of protocol of using these DCBs (Lutonix) between various operators in our center. Most of the lesions included in this study were complex femoropopliteal lesions (long C and D lesions), many of which extended to include both SFA and popliteal arteries, including below-knee popliteal lesions. Follow-up was also not available for some of these patients beyond short term. It would be inappropriate to compare the result of uncoated balloons during the same period with the DCBs in our institution because they were used for two different populations. Most if not all uncoated balloons were confined to shorter lesions; the DCBs were mainly confined to complex C and D lesions. We average around 300 lower extremity endovascular interventions per year, and it would be almost impossible to collect the data for all of these during the same study period. Therefore, further investigation from real-world experience with long-term follow-up is needed to confirm the value of these drug-eluting balloons, particularly for D lesions.

CONCLUSIONS

It is not the intention of this study to indicate that DCBs are not to be used in the femoropopliteal location. However, our data suggest that selective use for shorter femoropopliteal lesions is a better choice.

ARTICLE HIGHLIGHTS.

Type of Research

Single-center retrospective analysis

Key Findings

In this study, 228 patients treated with drug-coated balloon angioplasty (Lutonix 0.35) were analyzed. The primary patency rates at 2 years were 41% for claudication and 37% for critical limb ischemia (P = .307). Primary patency rates at 2 years were 71% for TransAtlantic Inter-Society Consensus A to C lesions and 14% for D lesions (P < .001). Limb salvage rates at 2 years were 100% for claudication and 74% for critical limb ischemia (P < .001).

Take Home Message

The authors suggest selective use of Lutonix-coated balloons for shorter femoropopliteal lesions.

DISCUSSION

Dr Eva M. Rzucidlo (Florence, SC). The authors present their work entitled “Clinlcal outcome of drug-coated balloon angioplasty in patients with femoropopliteal dlsease: A real world single-center experience.” The authors should be congratulated on an excellent presentation and commended on sending in their manuscript for review in a timely fashion.

Femoropopliteal disease has become the Wild West for treatment, with limited randomized controlled trials to tell us best options for treatment with best outcomes. Treatment options range from plain old balloon angioplasty to drug-eluting balloon angioplasty, primary stenting, rescue stenting, atherctomy, and bypass surgery. With every new device, we are told that it is the best with no long-term outcomes.

For those reasons. given the authors’ large cohort and therefore considerable experience, I have the following questions:

1. The authors’ cohort contains 228 patients who have primary intervention, being drug-coated balloon (DCB) angioplasty. Their study time point was 2010 to 2017. Was this all the patients with femoropopliteal disease for the study time period? Were these consecutive patients within this retrospective review? What percentage had DCB primary angioplasty? Could the authors have a comparison cohort of other treatments during that time period that would help the reader better understand a standard treatment group and therefore compare it with the DCB angioplasty group?

2. When treating patients minimally invasively. one always needs to have consideration that the treatment will not burn any bridges if there is failure. Did the authors note any failures that led to change in available plan for bypass (eg, above-knee popliteal bypass needed to be changed to below-knee popliteal bypass)? This would be important as this would change long-term outcomes, and there are data now to show that failed endovascular treatment leads to decreased patency of future bypass options.

3. Glven the authors’ poor results with DCB for TransAtlantic Inter-Society Consensus (TASC) D lesions, has this changed the authors’ primary intervention recommendations for these patients? Would the authors consider the concept of vessel preparation before DCB vs primary bypass?

4. With such poor patency rates, have the authors considered using intravascular ultrasound to determine adequacy of primary treatment? The authors’ mean length of lesion was 24 cm. This is considerably worse than in any other study. Why choose only DCB for the primary outcomes for such extensive disease? Have these results changed the authors’ practice?

5. What was the author’s preprocedure evaluation? It appears that the authors have hemodynamic data through ABIs; however. did the patients also get CTA with runoff to determine calcification? If they did, did this determine recommendations for intervention and patient selection?

6. How did the authors choose pedal access for their primary access site? Given the authors large cohort and experience, can they elucidate on any words of wisdom for choice of patient as well as access tips?

7. There was significantly high number of patients with groin complications, including hematoma, pseudoaneurysm, dissections, and Perclose complications. Did the authors use ultrasound guidance? Do the authors have any words of wisdom to help prevent these issues?

8. Given the poor outcomes for patients with claudication, what is the author’s indication for intervention? Do your patients do cilostazol trial and walking program first?

Dr Ali F. AbuRahma. Thank you very much, Dr Rzucidlo, for discussing this paper, and I am going to try to answer some of the questions that we have data on or questions that I could potentially give some answers to based on this series and our experience.

In regard to your first question, in our institution and during this time period, we averaged roughly around 300 endovascular interventions for patients with femoropopliteal occlusive disease. Therefore, it would be almost impossible to collect data on all these patients during this time period and give you a scientific solid answer. However, over the past few years, our multidisciplinary team mostly follows this protocol: patients with TASC II A and B lesions are generally treated with simple endovascular means (ie, percutaneous transluminal angioplasty [PTA] with or without stenting); whereas patients with TASC II C and D lesions are the ones who are treated with DCBs or drug-eluting stents. This study confined data to drug-eluting balloons (Lutonix). As far as our experience with simple endovascular procedures (ie, PTA or PTA with or without stenting), we do not believe this needs to be addressed any further because extensive publications have been done in regard to this entity.

In regard to question 3, yes. Many of our team members have changed their approach of not pushing the envelope for patients with TASC D lesions, and these have been treated with primary bypass. However, some of our interventionalists stili treat these lesions, inltially, with endovascular therapy.

In regard to question 4, as indicated in the manuscript, this is relatively higher than in other reported series. However, most of our lesions had complex TASC C and D lesions, and I cannot give you an answer as to whether the failure was primarily because of a DCB or the salvage stenting.

In regard to question 5, some patients underwent preoperative computed tomography angiography that might have shown calcification. However, a good bulk of these patients went from noninvasive vascular testing (ie, ankle-brachial index or duplex ultrasound) straight to catheter-directed therapy and intervention.

In regard to question 6 about pedal access, as noted, that was done only on a few occasions and was primarily done by one of our interventionists who happened to be an interventional cardiologist.

In regard to question 7, most of our patients’ accesses have lately been using ultrasound guidance, which minimize some of these complications.

Finally, in regard to question 8, all of our claudication patients undergo a trial of medical therapy including cilostazol. unless its contraindicated.