Three-year outcomes after cyanoacrylate closure of superficial truncal veins: A single-center practice cohort study

Abstract

Objective

The aim of this study was to evaluate mid-term durability, clinical outcomes, and patient-reported outcomes after cyanoacrylate closure (CAC) of superficial truncal veins in routine practice, with limb- and vein-level analyses.

Methods

This was a single-center, retrospective cohort study of adults previously treated with CAC of the great saphenous vein, small saphenous vein, anterior saphenous vein, or posterior accessory saphenous vein, with standardized follow-up assessments. Eligible patients were invited for standardized follow-up including duplex ultrasound, CEAP class, revised Venous Clinical Severity Score, EuroQol 5-dimension survey, and Aberdeen Varicose Vein Questionnaire. The primary endpoint was complete closure of the primary target vein on duplex ultrasound, defined as no ≥5-cm contiguous patency within the treated segment. Secondary endpoints included vein-level patency, new clinically significant varicose veins, adverse events, and satisfaction with symptoms and cosmetic appearance. Analyses accounted for clustering of veins/limbs within patients using generalized estimating equations.

Results

The study enrolled 89 patients (76.4% female; median age, 53 years) representing 110 limbs and 156 treated veins. The median time from treatment to follow-up was 3.7 years. Primary target vein complete closure at the limb level was 97.3% (107/110; 95% confidence interval [CI], 91.9%-99.1%). At follow-up, patency in any treated vein was present in 10.9% of limbs and 7.7% of veins. At follow-up, revised Venous Clinical Severity Score decreased from 6 (interquartile range, 4-8) to 1 (interquartile range, 0-2) (mean change, −4.5 ± 2.5; P < .001) and Aberdeen Varicose Vein Questionnaire scores improved among the subset with paired data (mean change, −7.7 ± 8.9; P < .001). New clinically significant varicose veins were present in 19.1% of limbs (95% CI, 12.7%-27.8%), often involving the anterior saphenous vein. Higher baseline body mass index was associated with vein-level patency on univariable analysis (median 33 vs 24 kg/m²; P < .001) and remained an independent predictor in exploratory multivariable models. Patient satisfaction per limb was 97% (completely/somewhat satisfied) but was significantly lower when any treated vein remained patent (completely satisfied 41.7%; P < .001). Limb-level adverse events occurred in 20.0% (95% CI, 12.9%-29.7%), with localized tenderness/phlebitis most commonly encountered; no deep venous events were observed.

Conclusions

Complete closure rates of all treated veins with CAC remain high at a median of 3.7 years post treatment with sustained clinical and patient-reported outcome improvements. Incomplete closure and recurrent varicose veins correlate with worse symptoms, visual appearance and patient satisfaction.

Article Highlights.

• •Type of Research: Single-center retrospective cohort study with standardized follow-up assessments
• •Key Findings: Among 89 patients (110 limbs, 156 veins) with median 3.7-year follow-up after cyanoacrylate closure, primary target vein closure was 97.3%. Revised Veinous Clinical Severity Score improved significantly (−4.5 ± 2.5; P < .001). Vein patency (7.7%) was associated with higher body mass index and accessory vein treatment.
• •Take Home Message: Cyanoacrylate closure provides durable mid-term anatomic closure and sustained symptom improvement in routine practice, although higher body mass index and accessory vein treatment are associated with recanalization and reduced satisfaction.

Chronic venous disorders (CVDs) are common worldwide and have substantial impact on quality of life and health care utilization.^1,2 In the United States, more than 30 million individuals have some form of CVD,³ with females and elderly patients disproportionally affected.4, 5, 6 Over the past 2 decades, treatment of CVD has shifted from surgical therapy to minimally invasive endovenous options.7, 8, 9, 10 Endothermal ablation modalities including radiofrequency (RFA) and endovenous laser therapy (EVLT), require tumescent anesthesia, whereas non-thermal non-tumescent techniques, including cyanoacrylate closure (CAC), mechanochemical ablation, and proprietary endovenous microfoam, avoid potential thermal injury and tumescent injections.

Early studies of CAC for saphenous ablation, including the eSCOPE cohort and the VeClose trial, demonstrated high short- and long-term closure rates.^11,12 The VeClose US Pivotal Trial was a multi-center, randomized trial comparing CAC with RFA (RFA, ClosureFast radiofrequency). Among 222 enrolled patients, complete closure was observed in 100% of CAC- and 87% of RFA-treated veins at 1 month. At 3 years, 94.4% of CAC- and 91.9% of RFA-treated veins had complete closure of the great saphenous vein (GSV),¹² and at 60 months, Kaplan-Meier estimates for freedom from recanalization were 91.4% and 85.2%, respectively, demonstrating noninferiority of CAC compared with RFA.¹³

The WAVES study extended these findings by evaluating CAC in a routine clinical practice including larger diameter veins and the treatment of multiple truncal veins without post-procedure compression. High closure rates were maintained at 3 months and 1 year.^14,15 One-year results demonstrated the safety and efficacy of CAC for the treatment of GSVs up to 20 mm in diameter, as well as small saphenous veins (SSVs), and/or accessory saphenous veins (ASVs), with an occlusion rate of 98%.¹⁴The objective of the current study was to evaluate mid-term clinical, duplex ultrasound (DUS), and patient-reported outcomes, including changes over time, in patients following CAC treatment in routine clinical practice in a multi-provider, single-center setting.

Institutional Review Board approval was obtained from Western IRB (WIRB Protocol #1268421), and all participants were provided informed consent after his/her eligibility was confirmed.

Methods

Study design

The New-WAVES study was a single-arm, single-center retrospective cohort study incorporating a retrospective review of index procedures and baseline characteristics, with standardized follow-up assessments performed to evaluate mid-term outcomes after CAC for symptomatic reflux of the GSV, SSV and/or accessory veins (ASV or posterior accessory saphenous vein [PASV]). Follow-up assessments were performed ≥30 months after the index CAC procedure, and all included limbs met the same minimum follow-up requirement regardless of participation in the original WAVES study.

The New-WAVES cohort differed from that of the VeClose^12,16 trial population by allowing treatment of any combination of refluxing GSV, SSV, ASV, and PASV, a wider range of patient ages, and the larger vein diameters. All study subjects that had been treated at the study site and had reached at least 30 months of follow-up after the original WAVES trial were invited to participate, whether as participants in that trial, or as treated in routine clinical practice. Limbs receiving subsequent ablation (EVLT or RFA) after the index CAC procedure were excluded.

Study subjects

Eligible subjects were adults (≥18 years) previously treated with CAC for symptomatic incompetent truncal veins. A minimum of 30 months had to have elapsed since the index CAC procedure. One or both limbs could be included if treated. At follow-up, subjects underwent a targeted physical examination, CEAP classification, revised Venous Clinical Severity Score (rVCSS), and DUS of each eligible limb. Patients completed patient-reported outcome measures including the EuroQol 5-dimension survey (EQ-5D) quality of life (QoL) survey and Aberdeen Varicose Vein Questionnaire (AVVQ). Subjects rated changes in symptoms and visual appearance compared with pre-treatment state (improved, unchanged, or worsened). For subjects not enrolled in the original WAVES study, index procedure details were abstracted by chart review. Pre-procedure baseline QoL questionnaires were available only for the patients who had been enrolled in the WAVES Study.

Adjunctive procedures, including sclerotherapy or phlebectomy, were not routinely performed at the time of the index CAC. For subjects enrolled in the original WAVES study, adjunctive procedures were not permitted within the first 3 months following index treatment per study protocol but were permitted thereafter if clinically indicated. For subjects treated outside the WAVES protocol, adjunctive treatments were performed at the discretion of the treating surgeon.

Compression therapy was not mandated following CAC and was not routinely prescribed in this cohort. In select cases, including advanced venous disease or concomitant treatments, the use of compression stockings was left to the discretion of the treating surgeon. Patients were not prohibited from wearing compression stockings if they elected to do so. The use of compression therapy was not systematically recorded.

Post-procedure imaging and outcome definitions

Follow-up DUS was performed to assess vein closure and patency. Complete closure of the primary target vein (PTV) was defined as DUS examination demonstrating closure along the entire treated target vein segment, with no discrete segment of patency ≥5 cm (assessment included color flow imaging, vein compression, and pulsed Doppler interrogation), consistent with definitions used in the VeClose and WAVES studies. In addition to this limb-level endpoint, DUS was used to assess for the presence of patency (>5 cm) in any treated vein segment, including secondary or tertiary treated veins.

Statistical analysis

The primary analytic goal of this study was descriptive characterization of mid-term clinical and patient-reported outcomes after CAC. Comparison of outcomes for subjects previously enrolled in the WAVES study with those treated contemporaneously (but not enrolled in the WAVES study), was conducted for a comparison of cohorts analysis. Categorical and continuous patient-level variables were analyzed using Fisher’s exact test and Wilcoxon rank-sum tests, respectively. A Wald test from a generalized estimating equations (GEEs) model for leg-level variables was used to account for multiple limbs per patient. There was no statistically significant difference in the clinical outcomes of WAVES vs non-WAVES enrolled subjects. Longitudinal changes in rVCSS, AVVQ, and EQ-5D among patients with paired measurements were assessed using paired t-tests (patient-level) and GEE models (leg-level). Exploratory vein-level univariable and multivariable GEE logistic regression models were used to evaluate associations between baseline characteristics (including body mass index [BMI] and treated vein type) and patency at follow-ups. Given the small number of patent veins, multivariable models were restricted to a limited number of covariates.

Results

A total of 89 treated subjects were enrolled, contributing 110 limbs, and 156 vein treatments. Of these, 32 had previously participated in the WAVES trial (ClinicalTrials.gov Identifier: NCT02585726). The interval for index CAC treatment to follow-up was 2.5 to 4.2 years (median, 3.7 years). At the time of treatment, subjects were 18 to 85 years old, (median, 53 years old), 76.4% were female, and 23.6% underwent bilateral limb treatment. The PTV was the GSV in 90.9% limbs, the SSV in 8.2%, and the ASV in 0.9%. Baseline and follow-up patient and limb level characteristics are summarized in Table I.

Table I.

Baseline characteristics of the study cohort

Characteristics	Valuea	WAVES cohortb		P-valuec	No. missingd
		Yes (n = 32)	No (n = 57)
Patient-level (N = 89 patients)
Age, years	53 (18-85)	52	53	.75
Female sex	68 (76.4)	23 (71.9)	45 (78.9)	.45
White race	83 (93.3)	30 (93.8)	53 (93.0)	>.99
BMI, kg/m2	24 (18-43)	26	24	.18
VAS (“How is your health today?”; 0-100, 100 = best)	88 (61-100)	88	N/A	–	57
Leg treated				–
Left	36 (40.4)	17 (53.1)	19 (33.3)
Right	32 (36.0)	15 (46.9)	17 (29.8)
Both	21 (23.6)	0 (0.0)	21 (36.8)
Leg-level (N = 110 legs)		(n = 32)	(n = 78)
CEAP				.015
C2	70 (63.6)	13 (40.6)	57 (73.1)
C3	23 (20.9)	10 (31.2)	13 (16.7)
C4	16 (14.5)	8 (25.0)	8 (10.3)
C5	1 (0.9)	1 (3.1)	0 (0.0)
VCSS	6 (2-14)	6	6	.34	10
Primary treated vein				.53
GSV	100 (90.9)	30 (93.8)	70 (89.7)
ASV	1 (0.9)	0 (0.0)	1 (1.3)
PASV	0 (0.0)	0 (0.0)	0 (0.0)
SSV	9 (8.2)	2 (6.2)	7 (9.0)
Veins treated (primary, secondary, or tertiary)
GSV	100 (90.9)	30 (93.8)	70 (89.7)	.53
ASV	23 (20.9)	7 (21.9)	16 (20.5)	.88
PASV	2 (1.8)	0 (0.0)	2 (2.6)	–
SSV	31 (28.2)	6 (18.8)	25 (32.1)	.17
Number of veins treated				.35
1	69 (62.7)	21 (65.6)	48 (61.5)
2	36 (32.7)	11 (34.4)	25 (32.1)
3	5 (4.5)	0 (0.0)	5 (6.4)
Pre-treatment vein diameter
GSV, mm	6.6 (4.0-14.0)	6.6	6.7	.85	6
ASV, mm	6.5 (3.6-9.9)	5.8	6.5	.55	6
PASV, mm	–	–	–	–	2
SSV, mm	5.2 (2.9-13.2)	5.5	5.2	.42	1
Total VenaSeal used, mL	1.7 (0.5-4.7)	1.7	1.6	.13	3
Total venous length treated, mm	43 (10-118)	45	40	.20	3

ASV, Anterior saphenous vein; BMI, body mass index; GSV, great saphenous vein; PASV, posterior accessory saphenous vein; SSV, small saphenous vein; VAS, Visual Analogue Scale; VCSS, Venous Clinical Severity Score.

^aValues are presented as number (%) or median (range); percentages calculated out of the number of patients or number of legs, excluding missing values.

^bValues are presented as number (%) or median (range).

^cComparison of cohorts using Fisher’s exact test or Wilcoxon rank-sum test for patient-level variables and a Wald test from a generalized estimating equations model for leg-level variables to account for multiple legs treated per patient.

^dNumber of patients or legs missing a variable or measurement; blank indicates no missingness.

The mean cyanoacrylate volume was 2.0 mL (standard deviation [SD], 0.9 mL; range, 0.5-4.7 mL), and the mean total treated vein length (all segments) was 50 cm (SD, 22.7 cm; range, 9-118 cm). DUS demonstrated complete closure of the PTV in 97.3% of limbs (95% confidence interval [CI], 91.9%-99.1%). Secondary or tertiary veins were treated in 37.3% of limbs, resulting in a total of 156 treated veins. Short-segment patency on follow-up DUS occurred predominantly in secondary or tertiary treated veins with patency observed in nine of 46 treated secondary or tertiary veins (19.6%). Overall, patency of any treated vein was present in 10.9% of limbs, and in 7.7% of treated veins (12/156).

Significant improvements from baseline to follow-up were observed in rVCSS among 100 limbs. Among the 31 patients with paired baseline and follow-up questionnaires, AVVQ scores also improved significantly (mean change, −7.7 ± 8.9; P < .001), whereas EQ-5D index scores showed a small, non-significant increase (mean change, 0.04 ± 0.12; P = .11). Baseline to follow-up changes in patient- and limb-level characteristics, including BMI, CEAP class, rVCSS, AVVQ, and EQ-5D are shown in Table II.

Table II.

Changes in patient- and limb-level characteristics between baseline and follow-up

Characteristics	No.a	Timepointb		Change
		Baseline	Follow-up	Mean ± SD	95% CI	P-valuec
Patient-level (N = 89 patients)
Age, years	89	53 (41-59)	56 (44-63)	3.6 ± 0.5	–	–
BMI, kg/m2	88	24 (22-29)	25 (22-30)	0.8 ± 2.6	0.3-1.3	.004
VAS (0-100, 100 = best)	31	90 (80-90)	90 (80-93)	0.5 ± 10.2	−3.3 to 4.2	.81
AVVQ (0-100, 0 = best)	31	15 (12-20)	6 (3-11)	−7.7 ± 8.9	−11.0 to −4.4	<.001
EQ-5D (0-1, 1 = best)	31	0.86 (0.86-0.94)	1.00 (0.84-1.00)	0.04 ± 0.12	−0.01 to 0.08	.11
Leg-level (N = 110 legs)
CEAP	110
C0		0 (0.0)	16 (14.5)
C1		0 (0.0)	37 (33.6)
C2		70 (63.6)	45 (40.9)
C3		23 (20.9)	4 (3.6)
C4		16 (14.5)	6 (5.5)
C5		1 (0.9)	2 (1.8)
VCSS	100	6 (4-8)	1 (0-2)	−4.5 ± 2.5	−5.1 to −4.0	<.001

AVVQ, Aberdeen Varicose Vein Questionnaire; BMI, body mass index; CI, confidence interval; EQ-5D, EuroQol 5-dimension survey; SD, standard deviation; VAS, Visual Analogue Scale; VCSS, Venous Clinical Severity Score.

^aNumber of patients or legs with values available at both baseline and follow-up.

^bValues are presented as median (interquartile range) or number (%); percentages calculated out of the number of patients or number of legs, excluding missing values.

^cPaired t-test (patient-level comparisons) or Wald test from a generalized estimating equations model to account for multiple legs treated per patient.

Subjects with any patent segment had a significantly higher pre-treatment BMI (median, 33 vs 24 kg/m²; P < .001). BMI increased slightly over the follow-up (mean, +0.8 kg/m²; P = .004). More of the treated veins found to be patent and refluxing were in the ASV distribution compared with the GSV/SSV distribution. Characteristics associated with patent vs closed treated veins are presented in Table III. Associations between clinical outcomes (including incomplete closure, adverse events, and clinically significant recurrent varicose veins) and patient-reported symptoms, appearance ratings, and satisfaction are summarized in Table IV. The number of treated veins that were patent on follow-up was small (n = 12), limiting the precision of multivariable analysis. In univariate models, higher baseline BMI and ASV treatment were associated with patency. When both baseline BMI and ASV were included in the same exploratory multivariable GEE model, both remained statistically significant: the odds ratio for patency was 2.8 per 1-SD increase in BMI (95% CI, 1.8-4.3; P < .001), and ASV treatment was associated with higher odds of patency compared with other treated vein segments (odds ratio, 3.8; 95% CI, 1.3-11.6; P = .018).

Table III.

Univariable comparison of veins with and without patency at follow-up (N = 156 treated veins)

Variablea	Vein closure statusb		P-valuec	No. missing
	Patent (n = 12)	Closed (n = 144)
Age, years	54 (49-58)	53 (42-58)	.48
ΔAge, years	3.7 (3.5-3.8)	3.6 (3.2-4.0)	.16
BMI, kg/m2	33 (29-38)	24 (22-28)	<.001
ΔBMI, kg/m2	1.3 (−0.3 to 1.8)	0.4 (−0.5 to 1.2)	.82	2
CEAP
C2	6 (50.0)	89 (61.8)	.43
C3	6 (50.0)	31 (21.5)	.039
C4	0 (0.0)	23 (16.0)	–
C5	0 (0.0)	1 (0.7)	–
VCSS	8 (7-9)	6 (5-8)	.16	14
Treated vein
GSV	6 (50.0)	94 (65.3)	.27
ASV	5 (41.7)	18 (12.5)	.0046
PASV	0 (0.0)	2 (1.4)	–
SSV	1 (8.3)	30 (20.8)	.33
Number of veins treated			.48
1	3 (25.0)	66 (45.8)
2	9 (75.0)	63 (43.8)
3	0 (0.0)	15 (10.4)
Pre-treatment vein diameter
GSV, mm	7.58 (6.97-7.76)	6.58 (5.30-8.11)	.027	6
ASV, mm	7.00 (6.50-8.20)	5.55 (4.50-7.55)	.10	6
PASV, mm	–	–	–	2
SSV, mm	4.30	5.30 (4.70-6.20)	–	1
Total VenaSeal, mL	2.0 (1.5-2.5)	1.8 (1.5-2.7)	.83	5
Total venous length treated, mm	40 (35-48)	50 (37-70)	.44	5

ASV, Anterior saphenous vein; AVVQ, Aberdeen Varicose Vein Questionnaire; BMI, body mass index; GSV, great saphenous vein; PASV, posterior accessory saphenous vein; SSV, small saphenous vein; VCSS, Venous Clinical Severity Score.

^aVariables are either baseline/pre-treatment or changes from baseline to follow-up (designated with the Δ symbol as a prefix).

^bValues are presented as median (interquartile range) or number (%); percentages calculated out of the number of treated veins in each column (12 or 144), excluding missing values.

^cWald test from a generalized estimating equations logistic regression model to account for multiple veins treated per patient.

Table IV.

Association of patient-reported outcomes with clinical outcomes

Variable	Any non-closurea		P-valueb	Any adverse eventa		P-valueb	Clinically significant recurrent varicose veinsa		P-valueb
	Yes (n = 12)	No (n = 98)		Yes (n = 22)	No (n = 88)		Yes (n = 21)	No (n = 89)
Symptoms			.10			.46			.001
Much better	6 (50.0)	78 (79.6)		16 (72.7)	68 (77.3)		10 (47.6)	74 (83.1)
Somewhat better	4 (33.3)	10 (10.2)		2 (9.1)	12 (13.6)		6 (28.6)	8 (9.0)
No change	1 (8.3)	8 (8.2)		3 (13.6)	6 (6.8)		2 (9.5)	7 (7.9)
Somewhat worse	1 (8.3)	1 (1.0)		1 (4.5)	1 (1.1)		2 (9.5)	0 (0.0)
Much worse	0 (0.0)	1 (1.0)		0 (0.0)	1 (1.1)		1 (4.8)	0 (0.0)
Visual appearance			.083			.95			.002
Much better	8 (66.7)	80 (81.6)		19 (86.4)	69 (78.4)		11 (52.4)	77 (86.5)
Somewhat better	3 (25.0)	13 (13.3)		1 (4.5)	15 (17.0)		8 (38.1)	8 (9.0)
No change	0 (0.0)	4 (4.1)		1 (4.5)	3 (3.4)		0 (0.0)	4 (4.5)
Somewhat worse	0 (0.0)	1 (1.0)		1 (4.5)	0 (0.0)		1 (4.8)	0 (0.0)
Much worse	1 (8.3)	0 (0.0)		0 (0.0)	1 (1.1)		1 (4.8)	0 (0.0)
Patient satisfaction			<.001			.57			.001
Completely satisfied	5 (41.7)	79 (80.6)		16 (72.7)	68 (77.3)		8 (38.1)	76 (85.4)
Somewhat satisfied	4 (33.3)	19 (19.4)		5 (22.7)	18 (20.5)		11 (52.4)	12 (13.5)
Unsatisfied but would do it again	3 (25.0)	0 (0.0)		1 (4.5)	2 (2.3)		2 (9.5)	1 (1.1)

^aValues are presented as number (%).

^bWald test from a generalized estimating equations logistic regression model to account for multiple legs treated per patient.

Patient satisfaction (“completely” or “somewhat” satisfied) was reported for 97% of limbs treated but was significantly lower in limbs with any non-closure of treated veins compared with limbs in which all treated veins remained closed (completely satisfied: 41.7% vs 80.6%; P < .001). Limbs with any non-closure also showed a trend toward less symptom improvement (“much better:” 50.0% vs 79.6%; P = .10) and less improvement in visual appearance (“much better:” 66.7% vs 81.6%; P = .083), although these differences did not reach statistical significance.

Adverse events included hypersensitivity (4.5%), treatment zone tenderness/phlebitis (7.2%), thrombosed tributary veins (5.4%), superficial thrombophlebitis (0.9%), and glue at exit site requiring local excision (0.9%). Overall limb-level adverse events occurred in 20% (95% CI, 12.9%-29.7%). There were no statistically significant differences in symptoms, appearance ratings, or satisfaction between limbs regardless of adverse event occurrences during follow-up. Limb-level outcome rates are summarized in Table V.

Table V.

Limb-level outcomes

Limb-level (N = 110 legs)	No. events	Percentage	95% CI, %a
Complete closure of the PTV	107	97.3	91.9-99.1
Any treated segments now patent	12	10.9	6.1-18.8
New clinically significant varicose veins	21	19.1	12.7-27.8
Any adverse events	22	20.0	12.9-29.7

CI, Confidence interval; PTV, primary target vein.

^aCIs calculated using generalized estimating equations logistic regression models to account for multiple legs treated per patient.

New clinically significant varicose veins developed in 19.1% of limbs (95% CI, 12.7%-27.8%). Patients with limbs with recurrent varicosities were less likely to report major symptom improvement (“much better:” 47.6% vs 83.1%; P = .001), less likely to report major improvement in visual appearance (“much better:” 52.4% vs 86.5%; P = .002), and substantially less likely to be completely satisfied (38.1% vs 85.4%; P = .001).

Discussion

This study provides mid-term outcomes of CAC for superficial truncal venous reflux in a multi-provider clinical practice. Most long-term data for CAC have been generated from prospective or industry-supported trials, and fewer studies describe outcomes in routine clinical practice. This analysis expands the evidence base by demonstrating durable anatomic closure and favorable patient-reported outcomes at a median of nearly 4 years after treatment.

Disease-specific and generic quality-of-life measures provided additional information on patient outcomes. AVVQ scores improved substantially, consistent with significant improvement in rVCSS and patient-reported symptoms. In contrast, EQ-5D index scores showed only a small, non-significant change, which is not unexpected, given that EQ-5D captures global health status rather than limb-specific venous symptoms. Our findings support the importance of including disease-specific instruments when assessing outcomes after superficial venous interventions.

The complete closure rate of 97.3% in this cohort is consistent with long-term results reported in the VeClose randomized trial, which provides the most robust comparative data for CAC. In VeClose, CAC remained noninferior to radiofrequency ablation at 3 years, with closure rates of 94.4% for CAC and 91.9% for RFA, and durability was maintained at 5 years, with freedom from recanalization of 91.4% in the CAC group.^13,15 More recent post-market data from Singapore demonstrated continued improvement in VCSS and high satisfaction at 3 years.¹⁷ These findings, together with the current study, support the long-term effectiveness of CAC outside controlled clinical trial conditions.

Higher BMI was associated with persistent venous patency or recanalization in this cohort. Although BMI-specific long-term CAC data remain limited, elevated BMI has been associated with less favorable outcomes with other superficial venous interventions. Although an analysis by Deol et al focused on endothermal ablation rather than CAC, the pattern of progressively worse rVCSS and Chronic Insufficiency Venous Interrogatory Questionnaire-20 (CIVIQ-20) outcomes with increasing BMI, particularly ≥35 kg/m²,¹⁸ is consistent with our observation that higher BMI was independently associated with vein-level patency/recanalization of treated veins. Together, these data suggest that patients with elevated BMI may benefit from shared decision-making to include realistic treatment expectations, counseling around weight management, and closer surveillance after superficial venous interventions, including CAC.

The safety profile in this study aligns with other authors’ reports. No novel adverse events associated with CAC occurred in this study cohort. Hypersensitivity reactions occurred in 4.5% of limbs, which is comparable to, or lower than, previously reported rates. In a larger cohort (379 limbs) evaluated for hypersensitivity at our practice, we found that hypersensitivity occurred in 5.8% of limbs, and that reactions were generally mild to moderate in severity.¹⁹ A separate Korean multicenter analysis by Cho et al described similar inflammatory and hypersensitivity-type responses early after CAC.²⁰ Only one exit-site granuloma was observed, and no deeper granulomatous reactions occurred. Other adverse events, including phlebitis, were self-limited and consistent with prior reports.

Patient-reported outcomes closely reflected the importance of durability of treatment. Limbs with incomplete closure demonstrated significantly lower satisfaction and showed trends toward less symptomatic and appearance improvement. In contrast, adverse events, most of which were mild and self-limited, did not significantly affect long-term symptoms, appearance ratings, or satisfaction. Clinically significant recurrent varicose veins had the most pronounced impact across all domains studied, with markedly lower rates of major symptom improvement, appearance improvement, and complete satisfaction compared with limbs without recurrence. These findings highlight the importance of varicose vein recurrence from a patient perspective, as well as the importance of recognizing and managing recurrent varicose veins, which is more likely to occur when the ASV is the treated vein.

Recurrent varicose veins were present in 19.1% of limbs. Patients with recurrent varicose veins reported lower satisfaction, which is not surprising given an expected correlation between anatomic success and symptom improvement. Accessory saphenous vein recanalization contributed to a portion of recurrences. Reflux in the ASV has likewise been implicated as a cause of recurrent varicose veins following endothermal ablation by other authors.^2,21 This perspective likely reflects the shorter treated segment and anatomic variability of ASV trunks and suggests that ASV disease may require different treatment or surveillance strategies than GSV or SSV reflux. Although we observed an association between recurrence and ASV recanalization, this study was not designed to adjudicate all mechanisms of recurrence.

The study has several limitations. Baseline data were obtained retrospectively, and although follow-up assessments were standardized, outcomes reflect late post-treatment evaluation rather than prospectively scheduled longitudinal follow-up. Additional limitations include incomplete baseline quality-of-life data for subjects who were not part of the original WAVES study, and the single-center nature of this cohort. Although this cohort includes mid-term follow-up with detailed DUS and patient-reported outcomes, the sample size limits precision for certain subgroup and multivariable analyses. Longer-term follow-up may further clarify the durability of CAC and late patterns of recanalization.

Conclusions

Overall, this study demonstrates that CAC provides durable anatomic closure and sustained patient-reported improvements at nearly 4 years in routine clinical practice without mandated postoperative compression. These findings support the mid-term effectiveness of CAC across a broad range of vein types and diameters and highlight the importance of monitoring and managing recurrence, particularly in patients with elevated BMI or ASV involvement.

Author contributions

Conception and design: KGi

Analysis and interpretation: KGi, MN, NP

Data collection: KGi, KGl, EF, RM

Writing the article: KGi, MN, RP

Critical revision of the article: KGi, KGl, EF, RM, MN, NP

Final approval of the article: KGi, KGl, EF, RM, MN, NP

Statistical analysis: MN, NP

Obtained funding: KGi

Overall responsibility: KGi

Funding

This investigator-initiated study was supported by a research grant from Medtronic. The sponsor had no role in the study design, protocol development, data collection, data analysis, or manuscript preparation. Sponsor review was limited solely to verification of accurate proprietary product nomenclature.

Disclosures

During the conduct of this work, K.D.G. received research support from Medtronic, Vesper Medical, Gore, and Boston Scientific; she serves on the scientific advisory boards for Gore, Medtronic, Boston Scientific, and Philips; and has served as a speaker for Gore, Medtronic, Boston Scientific, Philips, and Vesper Medical. R.M. receives research support from BD. The remaining authors report no conflicts.