Endovascular management of superficial femoral artery chronic total occlusions: Influence of lesion complexity and popliteal access on the efficacy of rotational atherectomy plus drug-coated balloon versus drug-coated balloon alone

Abstract

Objectives:

This study aimed to compare the safety and efficacy of rotational atherectomy combined with drug-coated balloon angioplasty (RA+DCB) versus DCB alone in patients with superficial femoral artery chronic total occlusions (SFA CTOs) and to evaluate the influence of lesion complexity and vascular access site (femoral or popliteal) on procedural success and 12-month clinical outcomes.

Material and Methods:

This single-center retrospective cohort included 98 consecutive patients (117 SFA CTOs) treated between December 2022 and January 2024. All patients presented with lifestyle-limiting claudication or chronic limb-threatening ischemia (Rutherford 3–6). The primary outcome was 12-month primary patency on duplex ultrasound; secondary outcomes were technical success and freedom from target-lesion revascularization (TLR). Multivariable logistic regression identified predictors of loss of primary patency and TLR, and Kaplan– Meier analysis compared patency across treatment groups and access sites.

Results:

Successful lesion crossing was achieved in 109 of 117 CTOs (93.2%). Technical success was similar for RA + DCB (94.3%) and DCB alone (94.6%). At 12 months, primary patency (67.3% vs. 50%, p = 0.16) and freedom from TLR (70.2% vs. 54.2%, p = 0.26) did not differ between treatment modalities, nor between femoral and popliteal access. Diabetes mellitus independently reduced the odds of primary patency (odds ratio [OR], 0.30, 95% confidence interval [CI], 0.11–0.88) and TLR-free survival (OR, 0.28, 95% CI, 0.09–0.86). Longer occlusion length predicted crossing failure (p = 0.008) but not loss of patency. No procedure-related major adverse limb events occurred.

Conclusion:

Routine addition of RA to DCB angioplasty for all SFA CTOs is not supported. RA-facilitated DCB may confer benefit only in selected heavily calcified lesions. Mid-term outcomes are driven primarily by diabetes and lesion length, rather than the access route, underscoring the importance of a risk-adapted, lesion-specific treatment strategy.

INTRODUCTION

Chronic total occlusions (CTOs) of the superficial femoral artery (SFA), in particular, present a significant therapeutic challenge, as these complex lesions often require advanced endovascular strategies to restore adequate perfusion and prevent limb-threatening ischemia.^[1] Drug-coated balloon (DCB) angioplasty has become a mainstay of treatment by delivering antiproliferative agents directly to the arterial wall to reduce neointimal hyperplasia and restenosis. However, severe calcifications increase arterial stiffness and hinder the penetration of antiproliferative drugs into the vessel wall, which in turn leads to higher restenosis rates after intervention.^[2] In particular, severe calcifications involving more than 270° of the vessel circumference adversely affect post-procedural vessel patency, thus limiting clinical success.^[3]

Rotational atherectomy (RA) has been developed as an adjunctive therapy designed to address these limitations. The principal aim of RA is vessel preparation—achieving optimal lumen expansion, improving arterial compliance, and facilitating the more effective application of adjunctive therapies such as DCBs or stents.^[4] It can effectively debulk calcified plaques, reduce the need for high-pressure balloon inflations, and potentially minimize the occurrence of flow- limiting dissections, thereby decreasing the likelihood of unplanned stent placement.^[5]

RA in combination with BA or DCB can provide superior vessel preparation, improve technical success, and enhance primary patency, particularly in calcified SFA lesions. However, its effect on reducing long-term target lesion revascularization (TLR) remains inconsistent and may depend on lesion complexity and device type. Furthermore, atherectomy procedures are technically demanding, requiring experienced operators to navigate complex anatomies while minimizing procedural complications such as vessel perforation, dissection, or device malfunction.^[1]

Another factor to consider in the management of SFA CTO is the selection of the vascular access site. However, the evidence surrounding its influence on both the technical efficacy and safety profile of atherectomy procedures remains limited. Popliteal access has become increasingly utilized for complex or flush occlusions, interventions in prior bypass grafts, or after failed antegrade attempts.^[6] The widespread use of popliteal access has been limited by concerns about access- site complications, especially the formation of arteriovenous fistulas and hematomas. However, when performed under ultrasound guidance, the direct retrograde approach can facilitate lesion crossing and reduce these complications.

The present study aims to evaluate the comparative safety and efficacy of RA + DCB versus DCB alone in the treatment of SFA CTOs. In addition, this study evaluates the influence of femoral versus popliteal access strategies on procedural success, primary patency, and limb salvage outcomes.

MATERIAL AND METHODS

Study design and population

This single-center study enrolled consecutive adult patients who underwent endovascular treatment for SFA CTOs between December 2022 and January 2024. Eligibility criteria comprised (1) the presence of an SFA CTO, (2) chronic limb- threatening ischemia, severe claudication, ulceration, or gangrene, (3) treatment with an endovascular intervention, and (4) at least 1 year of follow-up. Patients younger than 18 years, those who underwent concurrent femoral and popliteal access, and those who did not complete the 12-month clinical follow-up were excluded from the study. A total of 112 consecutive patients with 133 SFA CTO lesions were assessed for eligibility. Fourteen patients (16 SFA CTO lesions) were excluded because they did not complete the 12-month follow-up. Consequently, 98 patients with 117 SFA CTO lesions were included in the final analysis [Figure 1]. The clinical data of all patients treated in our department were prospectively recorded, and the patients were identified by a computerized search of this database and the medical records of our hospital. Up until June 2023, all consecutive patients with SFA CTOs were treated with DCB angioplasty because the RA system was not yet available at our institution. Following the introduction of the RA system into routine clinical practice, RA + DCB was performed only in lesions with successful true-lumen crossing, whereas lesions requiring subintimal recanalization were treated with DCB angioplasty alone. Therefore, treatment allocation was not randomized and was determined by the availability of the RA system and whether successful true-lumen crossing had been achieved.

Figure 1:

Flow diagram illustrating patient selection, lesion crossing outcomes, technical success, treatment allocation (rotational atherectomy plus drug-coated balloon vs. drug-coated balloon alone), and 12-month follow-up of patients with superficial femoral artery chronic total occlusions. SFA: Superficial femoral artery, CTO: Chronic total occlusion, RA: Rotational atherectomy, DCB: Drug coated balloon, TLR: Target lesion revascularization.

Ethical approval was obtained from the institutional ethics committee, and written informed consent was obtained from all participants. The study was managed without industrial financial support.

Preprocedural management

All patients received antiplatelet therapy with acetylsalicylic acid (100 mg daily) at least 1 week before the procedure, or a loading dose of 600 mg if earlier intervention was required. Patients receiving dual antiplatelet therapy (100 mg of acetylsalicylic acid and 75 mg of clopidogrel daily) continued their regimen without interruption. After vascular access, 70–100 IU/kg of heparin was administered to maintain an activated clotting time >250 seconds, with additional doses given as needed. In diabetic patients, insulin therapy was continued as usual, whereas metformin was withheld 48 h pre- and post-procedure, pending stable renal function.

Procedural technique

Access site selection was based primarily on the presence of at least a 5 cm non-occlusive segment proximal or distal to the lesion, as determined by duplex ultrasonography (DUS). Patient positioning (prone for popliteal access) and body habitus were also considered to ensure procedural feasibility and comfort. When ipsilateral access was not feasible, a contralateral crossover approach was used.

All procedures were performed under local anesthesia with 2% prilocaine. Ultrasound-guided access was performed with insertion of a 6/7F introducer sheath (Radiofocus, Intraducer II - Terumo, Tokyo, Japan) or a 6/7F, 45 cm guiding sheath (Destination - Terumo, Tokyo, Japan). A systemic heparin infusion was initiated to minimize thromboembolic complications.

Lesion crossing was initially attempted with a 0.035-inch guidewire and vertebral catheter. If unsuccessful, escalation to a 0.018-inch stiff guidewire or a 0.014-inch CTO-specific guidewire was employed. All atherectomies were performed using the RA system (Temren - Invamed, Ankara, Türkiye), with an integrated aspiration mechanism to reduce the risk of distal embolization. RA was not performed in cases where subintimal recanalization was achieved. To minimize the risk of arterial injury, the system was advanced slowly and continuously at a rate of 1–2 mm/s. Device activation was paused every 20–30 s, and the rotational speed of the atherectomy device was determined at the discretion of the operator. DCB (Extender Drug PTA Balloon – Invamed, Ankara, Türkiye) was utilized for all angioplasty procedures, with inflation times of at least 3 min and 8–10 atm. Due to institutional supply constraints, all procedures were performed using DCBs. Prolonged dilation was performed in cases of residual stenosis >30% or flow-limiting dissection, and bailout stenting was reserved for persistent complications despite these procedures.

Follow-up and endpoint events

All patients were followed at 1 month and subsequently every 3 months. DUS was performed at each follow-up visit. In cases of clinical deterioration and/or binary restenosis detected by DUS, computed tomography angiography was performed. If necessary, repeat angiography and revascularization were conducted within 1 week. Binary restenosis was defined as a peak systolic velocity ratio >2.4, corresponding to ≥50% diameter reduction.

The primary efficacy endpoint was the 12-month primary patency rate, defined as uninterrupted vessel patency without reintervention. Secondary efficacy endpoints included technical success and freedom from TLR. Technical success was defined as <30% residual stenosis and restored flow. TLR was defined as a reintervention for a target lesion with ≥50% diameter stenosis and/or clinical deterioration.

Calcification was assessed by digital subtraction angiography using the Peripheral Artery Calcification Scoring System (PACSS) and the Kockelkoren score to evaluate plaque morphology and severity.^[7,8] The DISFORM (Diameter reduction, spiral shape, flow impairment, or adverse morphology) classification was applied to describe post-angioplasty dissections.^[9] DISFORM I and II were categorized as Group A, whereas all other classifications were designated as Group B. Distal embolization was defined as the dislodgement of atherosclerotic debris or thrombotic material into the distal vascular bed during device manipulation, as evidenced by angiography.

Statistical analysis

Statistical analyses were performed using Statistical Package for the Social Sciences (SPSS) software (version 21.0; SPSS Inc., Chicago, IL, USA). Continuous variables were expressed as mean ± standard deviation or medians (ranges), and categorical variables as counts (percentages). The Shapiro– Wilk test was used to assess normality. Independent-samples t-tests or Mann–Whitney U tests were used to compare continuous variables. Categorical variables were analyzed using Chi-square or Fisher’s exact test. Univariate analysis using the Chi-square test was performed to identify potential risk factors from pre-operative clinical data. Variables with a p < 0.10 in univariate analysis and clinically relevant variables were considered for inclusion in the multivariable logistic regression model. Multivariable logistic regression analysis was performed to identify independent predictors of 12-month primary patency and freedom from TLR. Results were reported as odds ratios (ORs) with 95% confidence intervals (CIs) and p values. Kaplan–Meier analysis was used to estimate primary patency and freedom from TLR, and differences between groups were compared using the log- rank test. Patients who remained free of the study endpoint at the end of follow-up or who were lost to follow-up were treated as censored observations at the time of their last available clinical assessment. The Chi-square test was also used to assess differences between scoring groups. The accuracy and discriminative ability of the prediction model were evaluated using the receiver operating characteristic (ROC) curve and the area under the curve (AUC).

RESULTS

Procedural outcomes

The patient selection process is summarized in Figure 1. Of the 117 SFA CTO lesions included in the final analysis, successful lesion crossing was achieved in 109 lesions (93.2%). In eight cases, more than 2 crossing attempts were unsuccessful. Flow-limiting recoil was observed in 6 cases (5.1%), despite repeated DCB angioplasty in 3 cases and RA followed by repeated DCB angioplasty in 3 cases. These patients underwent femoropopliteal bypass surgery.

Technical success was achieved in 94.3% for the RA + DCB group and 94.6% for the DCB group (p = 1.00). A total of 8 procedures were performed via crossover access, with one procedure deemed unsuccessful due to elastic recoil following subintimal recanalization. 13 SFA CTOs were crossed through the subintimal pathway and managed exclusively with DCB angioplasty, achieving a technical success rate of 92.3% and no flow-limiting dissections or bailout stenting. The demographic findings of patients and target SFA CTO lesions and the statistical findings related to technical success were presented in Table 1. The cut-off value for the occlusion length to discriminate successful lesion crossing versus non-crossing was 18 cm and was statistically significant (p < 0.0001), with an AUC = 0.755 (0.667–0.830) [Figure 2a]. The sensitivity and specificity of the cut-off value were 87.5% (47.3–99.7) and 66.06% (56.4–74.9), respectively. The cutoff value for occlusion length to predict freedom from TLR in the DCB-alone group was 6 cm, which was statistically significant (p < 0.0001), with an AUC of 0.781 (95% CI: 0.635–0.889) [Figure 2b]. The sensitivity and specificity of this cutoff value were 95.24% (95% CI: 76.2–99.9) and 60% (95% CI: 38.7–78.9) for both endpoints, respectively. The ROC curve analysis for occlusion length in predicting freedom from TLR in the RA + DCB group was not statistically significant (p = 0.098).

Table 1:

Demographic findings of patients and target SFA CTO lesions

Variables	Patients (n=98)	Technical Success (n=103)	Crossing Failure (n=8)	Recoil (n=6)	p-value
Age	64.34±10.32	64.43±9.77	59.88±7.2	68.67±10.17	0.2
Male gender (%)	79 (83.7)	87 (84.5)	6 (75)	3 (50)	0.61
Hypertension (%)	55 (56.1)	61 (59.2)	4 (50)	3 (50)	0.72
Hyperlipidemia (%)	28 (28.6)	30 (29.1)	2 (25)	0 (0)	1.0
Diabetes mellitus (%)	48 (49)	48 (46.6)	4 (50)	3 (50)	0.29
Coronary artery disease (%)	51 (52)	55 (53.4)	4 (50)	2 (33.3)	1.00
Smoking (%)	82 (83.7)	87 (84.5)	7 (87.5)	5 (83.3)	1.00
Lesion characteristics (n=117)
Right SFA (%)	56 (47.9)	50 (48.5)	4 (50)	2 (33.3)	1.00
Length, cm	14.77±8.62	14.10±8.50	22.45±6.63	17.75±8.48	0.008
Proximal diameter, mm	4.66±1.01	4.71±0.98	4.37±0.92	4.12±1.43	0.35
Distal diameter, mm	4.02±1.14	4.03±1.11	4.27±1.38	3.38±1.37	0.56
TASC-II (%)
Type A	17 (14.5)	17 (16.5)	0	0	0.25
Type B	36 (30.8)	32 (31)	1 (12.5)	3 (50)
Type C	64 (54.7)	54 (52.5)	7 (87.5)	3 (50)
PACSS score (≥3) (%)	35 (29.9)	27 (26.2)	4 (50)	4 (66.7)	0.22
Kockelkoren score (<7) (%)	87 (74.4)	82 (79.6)	4 (50)	1 (16.7)	0.075
Rutherford (%)
3	29 (24.9)	25 (24.3)	2 (25)	2 (33.3)	1.00
4	46 (39.3)	41 (39.8)	3 (37.5)	2 (33.3)
5	40 (34.2)	35 (34)	3 (37.5)	2 (33.3)
6	2 (1.7)	2 (1.9)	0 (0)	0 (0)
Antegrade* (n=117) (%)	62 (52.9)	57 (55.3)	3 (37.5)	2 (33.3)	0.47

^*Antegrade is including cross procedures, TASC: Trans-Atlantic Inter-Society Consensus, PACSS: The peripheral artery calcification scoring system, Kockelkoren Score; <7 intimal dominant, ≥7 non-intimal dominant plaque formation. SFA: Superficial femoral artery, CTO: Chronic total occlusion Statistical significance was defined as p < 0.05. Continuous variables are presented as mean ± standard deviation

Figure 2:

Receiver operating characteristic (ROC) curve analysis evaluating occlusion length as a predictor of procedural and clinical outcomes in superficial femoral artery chronic total occlusions. (a) ROC curve demonstrating the predictive performance of occlusion length for successful lesion crossing. (b) ROC curve demonstrating the predictive performance of occlusion length for freedom from target lesion revascularization in the drug-coated balloon-alone group. The diagonal line represents the line of no discrimination.

Among the 103 successfully treated SFA CTOs, no statistically significant differences were found between the RA + DCB group and the DCB group regarding age, gender, comorbidities or bailout stent implantation (p > 0.05). The statistical findings related to endovascular treatment modalities are presented in Table 2. The flow- limiting dissection rate was 8.7% (9/103), and no statistically significant difference was found between the RA + DCB group and the DCB group (4% [2/50], and 13.2% [7/53], p = 0.16, respectively) and femoral and popliteal access group (8.7% [5/57], and 8.8% [4/46], p = 1.00 respectively). Bailout stent implantation was required in 9 cases, each of which exhibited a DISFORM classification score of 4. Distal embolic events occurred in 3 patients (6%) who underwent RA. In two cases accessed through the femoral artery, thrombus aspiration was successfully performed using a 6F aspiration catheter (Sofia - MicroVention, Terumo, Tokyo, Japan). In the case with popliteal access, the thrombus was aspirated directly through the 7F introducer without the need for additional intervention.

Table 2:

Intraprocedural findings of endovascular treatment modalities and access sites.

Variables	RA+DCB (n=50)	DCB (n=53)	p-value	Popliteal (n=46)	Femoral (n=57)	p-value
Lesion characteristics
Right SFA (%)	23 (46)	27 (50.9)	0.62	22 (47.8)	28 (49.1)	0.62
Length, cm	15.65±7.43	12.63±9.23	0.07	18.70±7.05	10.38±7.76	0.001
Proximal diameter, mm	4.56±1.00	4.85±0.96	0.14	4.98±0.96	4.49±0.96	0.03
Distal diameter, mm	4.28±0.99	4.09±0.89	0.26	4.27±1.07	4.01±0.92	0.05
TASC-II (%)
Type A	5 (10)	12 (22.6)	0.05	3 (6.5)	14 (24.6)	0.001
Type B	12 (24)	20 (37.7)		4 (8.7)	28 (49.1)
Type C	16 (32)	9 (17)		18 (39.1)	7 (12.3)
Type D	17 (34)	12 (22.6)		21 (47.5)	8 (14)
PACSS Score (≥3) (%)	11 (22)	16 (30.2)	0.35	9 (19.6)	18 (31.6)	0.17
Kockelkoren score (<7) (%)	41 (82)	41 (77.4)	0.56	36 (78.3)	46 (80.7)	0.76
Intraprocedural (%)
Access*/Atherectomy	26 (52)	20 (37.7)	0.15	26 (56.5)	24 (42.1)	0.15
Dissection	42 (84)	39 (73.6)	0.20	39 (84.8)	42 (73.7)	0.17
DISFORM* (%)
Group A	27 (64.3)	25 (64.1)	0.98	24 (61.5)	28 (66.7)	0.63
Group B	15 (35.7)	14 (35.9)		15 (38.5)	14 (33.3)
Bailout Stenting (%)	5 (10)	4 (7.5)	0.74	2 (3.5)	7 (15.2)	0.07
Distal Embolization (%)	3 (6)	0 (0)	0.11	1 (1.8)	2 (4.3)	0.59

^*Femoral puncture is including cross procedures, *DISFORM: Diameter reduction, spiral shape, flow impairment, or adverse morphology, TASC: Trans-Atlantic Inter-Society Consensus, PACSS: The peripheral artery calcification scoring system, Kockelkoren Score; <7 intimal dominant, ≥7 non-intimal dominant plaque formation. SFA: Superficial femoral artery, CTO: Chronic total occlusion, RA: Rotational atherectomy, DCB: Drug coated balloon, Statistical significance was defined as p < 0.05. Continuous variables are presented as mean ± standard deviation

The statistical findings comparing treatment modalities and access sites are presented in Table 3. In the popliteal access group, occlusion length and Trans-Atlantic Inter-Society Consensus II (TASC II) demonstrated statistically significant differences (p < 0.05). No significant differences were found between these approaches regarding patient demographics (p > 0.05).

Table 3:

Lesion characteristics and intraprocedural outcomes by access site and treatment modality.

Treatment modality	RA+DCB (n=50)		p-value	DCB (n=53)		p-value
Access site	Popliteal (n=26)	Femoral (n=24)		Popliteal (n=20)	Femoral (n=33)
Age	64.23±7.97	59.96±11.97	0.14	65.30±9.14	67.30±8.85	0.43
Occlusion
Right SFA (%)	10 (38.5)	13 (54.2)	0.27	12 (60)	15 (45.5)	0.31
Length, cm	19.44±4.65	11.53±7.76	0.001	17.73±9.35	9.55±7.78	0.001
Proximal diameter, mm	4.78±1.02	4.32±0.94	0.11	5.24±0.84	4.62±0.96	0.02
Distal diameter, mm	4.50±1.02	4.04±0.93	0.10	3.96±1.09	3.69±1.22	0.42
TASC-II (%)
Type A+B	2 (7.7)	15 (62.5)	0.001	5 (25)	27 (81.8)	0.001
Type C+D	24 (92.3)	9 (37.5)		15 (75)	6 (18.2)
PACSS score (≥3) (%)	3 (11.5)	8 (33.3)	0.06	6 (30)	10 (30.3)	0.98
Kockelkoren score (<7) (%)	22 (84.6)	19 (79.2)	0.72	14 (70)	27 (81.8)	0.34
Intraprocedural (%)
Dissection	23 (88.5)	19 (79.2)	0.46	16 (80)	23 (69.7)	0.41
DISFORM (%)
Group A	15 (65.2)	12 (63.2)	0.89	9 (56.3)	16 (69.6)	0.39
Group B	8 (38.5)	7 (36.8)		7 (43.8)	7 (30.4)
Access site	Popliteal (n=46)		p-value	Femoral (n=57)		p-value
Treatment modality	RA+DCB (n=26)	DCB (n=20)		RA+DCB (n=24)	DCB (n=33)
Age	64.23±7.97	65.30±9.14	0.67	59.96±11.97	67.30±8.85	0.01
Occlusion
Right SFA (%)	10 (38.5)	12 (60)	0.15	13 (54.2)	15 (45.5)	0.52
Length, cm	19.44±4.65	17.73±9.35	0.46	11.53±7.76	9.55±7.78	0.35
Proximal diameter, mm	4.78±1.02	5.24±0.84	0.11	4.32±0.94	4.62±0.96	0.26
Distal diameter, mm	4.50±1.02	3.96±1.09	0.09	4.04±0.93	3.69±1.22	0.24
TASC –II (%)
Type A+B	2 (7.7)	5 (25)	0.21	15 (62.5)	27 (81.8)	0.10
Type C+D	24 (92.3)	15 (75)		9 (37.5)	6 (18.2)
PACSS Score (≥3) (%)	3 (11.5)	6 (30)	0.15	8 (33.3)	10 (30.3)	0.81
Kockelkoren Score (<7) (%)	22 (84.6)	14 (70)	0.29	19 (79.2)	27 (81.8)	1.0
Intraprocedural (%)
Dissection	23 (88.5)	16 (80)	0.68	19 (79.2)	23 (69.7)	0.42
DISFORM (%)
Group A	15 (65.2)	9 (56.3)	0.57	12 (63.2)	16 (69.6)	0.66
Group B	8 (38.5)	7 (43.8)		7 (36.8)	7 (30.4)

DISFORM: Diameter reduction, spiral shape, flow impairment, or adverse morphology. TASC: Trans-Atlantic Inter-Society Consensus, PACSS: The peripheral artery calcification scoring system, Kockelkoren Score; <7 intimal dominant, ≥7 non-intimal dominant plaque formation, Statistical significance was defined as p < 0.05. Continuous variables are presented as mean ± standard deviation. RA: Rotational atherectomy, DCB: Drug coated balloon, SFA: Superficial femoral artery

Clinical outcomes

During the 12-month follow-up, overall mortality was 3.9% (4/103), with two deaths due to heart failure (at 2 and 11 months post-procedure) and two from multi- organ failure (approximately 6 months post-procedure). The major amputation rate was 5.8% (6/103), consisting of four above-knee and two below-knee amputations. No significant differences were found in amputation rates between treatment modalities (the RA + DCB group vs. the DCB group, p = 0.68) or access sites (femoral vs. popliteal, p = 0.69). Freedom from amputation at 12 months did not significantly differ between endovascular treatment methods or access sites (log-rank p = 0.38 and p = 0.58, respectively).

Primary patency rates at 3, 6, 9, and 12 months were 89.2%, 74%, 65.7%, and 58.6%, respectively. No significant differences were observed between the RA + DCB group and the DCB group at 3, 6, 9 and 12 months (94% vs. 84.6% p = 0.13, 76% vs. 72% p = 0.65, 71.4% vs. 60% p = 0.23, and 67.3% vs. 50%, p = 0.08, respectively). Kaplan–Meier analysis demonstrated no significant differences in overall primary patency between treatment modalities (log-rank p = 0.16; [Figure 3a]) or vascular access sites (log-rank p = 0.98; [Figure 3b]).

Figure 3:

Kaplan–Meier curves demonstrating 12-month clinical outcomes following endovascular treatment of superficial femoral artery chronic total occlusions. (a) Primary patency according to treatment modality (rotational atherectomy plus drug-coated balloon vs. drug-coated balloon alone). (b) Primary patency according to vascular access site (femoral versus popliteal). (c) Freedom from target lesion revascularization (TLR) according to treatment modality. (d) Freedom from TLR according to vascular access site. Plus (+) symbols indicate censored observations. DCB: Drug coated balloon. P_Flow denotes the p-value of the log-rank test.

Overall freedom from TLR at 3, 6, 9, and 12 months was 91%, 76.3%, 69.1%, and 62.4%, respectively. When comparing the RA + DCB group to the DCB group, no significant differences were found at 3, 6, 9 and 12 months (94% vs. 88% p = 0.49, 77.6% vs. 75% p = 0.77, 72.9% vs. 65.2% p = 0.42; and 70.2% vs. 54.2%, p = 0.11, respectively). Kaplan–Meier analysis demonstrated no significant differences in freedom from TLR between treatment modalities (log-rank p = 0.26; [Figure 3c]) or vascular access sites (log-rank p = 0.98; [Figure 3d]).

Univariate analysis identified several variables associated with 12-month primary patency and freedom from TLR [Table 4]. Variables with p < 0.10 were subsequently entered into the multivariable logistic regression model. Multivariable analysis demonstrated that diabetes mellitus was an independent predictor of reduced 12-month primary patency (OR, 0.30; 95% CI, 0.11–0.88; p = 0.029) and lower freedom from TLR (OR, 0.28; 95% CI, 0.091–0.865; p = 0.027). Subgroup analyses showed no independent predictors of primary patency or freedom from TLR in the RA + DCB group. In contrast, diabetes mellitus remained significantly associated with reduced primary patency and freedom from TLR in the DCB-alone group.

Table 4:

Univariate analysis of risk factors for 12 months primary patency and freedom from TLR.

Variables	Primary Patency		p-value	Freedom from TLR		p-value
	SG (n=58)	FG (n=41)		SG (n=58)	FG (n=35)
Age	64.34±8.86	63.56±10.89	0.70	64.34±8.86	62.74±10.73	0.44
Male (%)	47 (81)	37 (90.2)	0.20	47 (81)	31 (88.6)	0.34
Smoking (%)	53 (91.4)	32 (78)	0.06	53 (91.4)	32 (78)	0.05
DM (%)	18 (31)	26 (63.4)	0.001	18 (31)	22 (62.9)	0.003
HL (%)	9 (15.5)	17 (41.5)	0.004	9 (15.5)	14 (40)	0.008
CAD (%)	25 (43.1)	26 (63.4)	0.05	25 (43.1)	22 (23.7)	0.07
Occlusion
Length, cm	12.51±8.36	16.63±78.47	0.018	12.51±8.36	17.79±7.76	0.003
PACSS score (≥3)	12 (20.7%)	14 (34.1%)	0.13	12 (20.7%)	10 (28.6%)	0.39
Kockelkoren score (<7)	50 (86.2%)	29 (70.7%)	0.06	50 (86.2%)	27 (77.1%)	0.26
TASC-II (Type C+D)	26 (44.8%)	27 (65.9%)	0.039	26 (44.8%)	25 (71.4%)	0.013
Intraprocedural
Dissection	23 (88.5%)	19 (79.2%)	0.46	16 (80%)	23 (69.7%)	0.41
DISFORM
Group A	15 (65.2%)	12 (63.2%)	0.89	9 (56.3%)	16 (69.6%)	0.39
Group B	8 (38.5%)	7 (36.8%)		7 (43.8%)	7 (30.4%)
Variables	Primary Patency (DCB alone)		p-value	Freedom from TLR (DCB alone)		p - value
	SG (n=25)	FG (n=25)		SG (n=25)	FG (n=21)
DM (%)	6 (24)	17 (68)	0.002	6 (24)	13 (61.9)	0.009
HL (%)	3 (12)	12 (48)	0.005	3 (12)	9 (42.9)	0.018
CAD (%)	10 (40)	18 (72)	0.023	10 (40)	15 (66.7)	0.07
Occlusion
Length, cm	9.08±8.04	16.32±9.47	0.005	9.08±8.04	17.44±8.85	0.002
PACSS score (≥3) (%)	6 (24)	9 (36)	0.36	6 (24)	6 (28.6)	0.73
Kockelkoren Score (<7) (%)	21 (84)	18 (72)	0.30	21 (84)	17 (81)	1
TASC-II (Type C+D) (%)	6 (24)	14 (56)	0.021	6 (24)	13 (61.9)	0.009

DISFORM: Diameter reduction, spiral shape, flow impairment, or adverse morphology. TASC: Trans-Atlantic Inter-Society Consensus, PACSS: The peripheral artery calcification scoring system, Kockelkoren Score; <7 intimal dominant, ≥7 non-intimal dominant plaque formation. TLR: Target lesion revascularization, DCB: Drug coated balloon, DM: Diabetes mellitus, HL: Hyperlipidemia, CAD: Coronary artery disease, SG: Success group, FG: Failure group, Statistical significance was defined as p < 0.05. Continuous variables are presented as mean ± standard deviation.

DISCUSSION

In this study, the overall technical success rate for endovascular treatment of SFA CTOs was 94.3% in the RA + DCB group and 94.6% in the DCB group. True lumen crossing was achieved in 88% of lesions (109 out of 117), aligning with prior studies. Taneva et al. reported a technical success rate of 96.1% using RA + DCB angioplasty, while Zeller et al. reported 89.6% using directional atherectomy (DA) + DCB angioplasty.^[1,10] Cai et al. observed a significantly higher success rate in the DA + DCB group (95.6%) compared to DCB alone (75.5%).^[11] Wischmann et al. reported similarly high procedural success rates with REA + DCB and DCB alone in patients undergoing endovascular treatment of predominantly calcified common femoral artery lesions (95% vs. 96%).^[12] No notable differences were observed between the treatment arms at 3-, 6-, 9-month and 12-month primary patency follow-up intervals. Furthermore, the 67.3% patency rate of RA + DCB was lower than prior reports (80.5–97%), likely due to the use of the RA system with limited evidence regarding its clinical efficacy and outcomes and/or a higher- risk population with severe calcification and complex (TASC II C/D) lesions.^[1,10-14] The TLR outcomes observed in this study are consistent with previously published data, suggesting that the combination of atherectomy and DCB does not consistently reduce the need for reintervention compared to DCB alone.^[10,12,15] The major amputation rate was comparable between the two groups. In addition, the use of RA was restricted to true-lumen recanalization, excluding more complex subintimal pathways, which may have biased outcomes in favor of RA. These findings indicate that, although RA + DCB was preferentially used for longer and more complex lesions, it did not demonstrate a significant advantage over DCB alone in terms of 12-month primary patency or freedom from TLR. It should also be acknowledged that the treatment groups were not fully comparable with respect to lesion complexity. Although baseline demographic characteristics were similar, the RA + DCB group tended to include longer lesions and a higher proportion of TASC II C/D lesions. In contrast, calcification severity assessed by the PACSS and Kockelkoren classifications was comparable between the groups. These baseline differences may have attenuated the apparent benefit of RA + DCB and should be considered when interpreting the absence of statistically significant differences between treatment strategies.

Although not statistically significant, the lower dissection rate in the RA+DCB group (4% vs. 13.2%) suggests a potential benefit of vessel preparation. This pattern aligns with previous studies that also indicated lower dissection rates with atherectomy combined with DCB compared to DCB alone.^[10,11]

For immediate procedural success, longer lesion length, PACSS scores ≥3, and Kockelkoren scores ≥7 were significantly associated with higher rates of crossing failure and post-angioplasty recoil. Other anatomic variables, including proximal and distal vessel diameter, TASC-II grade, laterality of the lesion, and Rutherford classification, showed no consistent association with technical outcome. The absence of an effect from TASC-II grade may reflect its emphasis on length and anatomical distribution, rather than on plaque composition or morphological features that more directly govern wire passage and post-dilatationbehavior. Özdemir-van Brunschot et al. reported that lesion/occlusion length and higher PACSS scores were the primary predictors of crossing failure in femoropopliteal lesions, underscoring the impact of lesion complexity and calcification severity on endovascular procedural outcomes.^[16] Similarly, Qiu et al. identified calcified lesions, lesion length >20 cm, ostial SFA occlusion, and popliteal artery involvement as significant predictors of technical failure in femoropopliteal endovascular treatment.^[17]

Our findings further suggest that the PACSS and Kockelkoren classification systems provide complementary information rather than interchangeable measures of lesion complexity. While PACSS primarily reflects the overall burden and distribution of arterial calcification, the Kockelkoren classification further characterizes plaque morphology, particularly the distinction between intimal- and non- intimal-dominant calcification. In our cohort, both scoring systems were associated with procedural complexity, including crossing failure and post-angioplasty recoil, but neither independently predicted 12-month primary patency or freedom from TLR after multivariable adjustment. Therefore, their principal clinical value appears to lie in procedural planning and lesion selection rather than in predicting mid-term clinical outcomes.

In univariate analysis of the entire cohort, increasing occlusion length, the presence of DM, HL, and TASC-II C–D morphology each predicted significantly lower 12-month primary patency and freedom from TLR. However, when stratified by treatment modality, these adverse associations persisted only in the DCB-alone subgroup and were no longer evident among patients treated with RA + DCB. This suggests that adjunctive RA may attenuate the negative impact of lesion length, metabolic comorbidities and complex TASC-II anatomy on mid-term vessel patency. In addition, multivariable analysis of risk factors indicated that DM was the most significant independent predictor of TLR. These findings support a risk-adapted treatment strategy [Figure 4]. RA may be considered selectively for lesions with true-lumen crossing and intimal-dominant concentric calcification. Conversely, extensive lesions with high Kockelkoren scores or concentric non-intimal calcification may be more effectively treated by primary or percutaneous bypass. RA should be avoided in cases requiring subintimal recanalization due to limited efficacy and increased embolic risk.

Figure 4:

Proposed treatment algorithm for endovascular management of superficial femoral artery chronic total occlusions. Clinical risk factors and lesion characteristics, including occlusion length, Trans-Atlantic Inter-Society Consensus II classification, Peripheral Arterial Calcium Scoring System, and Kockelkoren score, were used to guide procedural planning and vascular access selection. Femoral access was generally preferred for shorter, less complex lesions, whereas popliteal access was favored for complex lesions or following unsuccessful antegrade crossing. Following lesion crossing, true-lumen crossing allowed treatment with rotational atherectomy plus drug-coated balloon or drug-coated balloon angioplasty according to lesion characteristics and diabetes status, whereas subintimal crossing was managed with drug-coated balloon alone, provisional stenting, or surgical bypass. TASC: Trans- Atlantic Inter-Society Consensus, PACSS: The peripheral artery calcification scoring system, RA: Rotational atherectomy, DCB: Drug coated balloon, DM: Diabetes mellitus, HL: Hyperlipidemia, CAD: Coronary artery disease.

In this study, the access site selection did not significantly affect technical success, procedural outcomes, or complication rates. Furthermore, the retrograde approach is more commonly selected in cases with longer occlusions and larger vessel diameters. Prior studies have similarly shown that popliteal access may offer higher procedural success in complex, calcified, or long-segment occlusions where femoral access is limited.^[6,18,19] In addition, using a 7F or larger introducer may also reduce distal embolization during atherectomy in calcified, high plaque burden cases. Overall, lesion features, length, location, and calcification, primarily determine access site selection, rather than differences in clinical outcomes.

This study has several limitations. As a retrospective, single- center study, it is subject to selection bias and limited generalizability. Treatment allocation was not randomized but was influenced by both the introduction of the RA system during the study period and successful true-lumen crossing, which may have introduced treatment-era bias. Although all procedures were performed by experienced endovascular operators, temporal changes in clinical practice cannot be completely excluded. In addition, the inclusion of predominantly complex lesions (TASC II C/D, PACSS ≥3) may limit the generalizability of our findings. Patency assessment relied primarily on DUS, with angiographic follow-up reserved for symptomatic patients, which may have underestimated asymptomatic restenosis rates.

CONCLUSION

The findings of this study suggest that the clinical utility of RA + DCB in SFA CTOs may be limited to specific lesion and patient subsets, rather than supporting its routine application. Lesion characteristics should be considered when selecting the optimal endovascular strategy, whereas the choice of vascular access did not significantly influence mid-term clinical outcomes. Further prospective studies are needed to define the optimal indications for atherectomy in complex femoropopliteal disease.

Funding Statement

Financial support and sponsorship: Nil.

Ethical approval:

The research/study was approved by the Institutional Review Board at the Ethics Committee of Kartal Dr. Lütfi Kırdar City Hospital, number 2024/010.99/6/30, dated July 26, 2024.

Declaration of patient consent:

Patient’s consent is not required as the patient’s identity is not disclosed or compromised.

Conflicts of interest:

There are no conflicts of interest.

Use of artificial intelligence (AI)assisted technology for manuscript preparation:

The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.