VeClose trial 12-month outcomes of cyanoacrylate closure versus radiofrequency ablation for incompetent great saphenous veins

VeClose trial 12-month outcomes of cyanoacrylate closure versus radiofrequency ablation for incompetent great saphenous veins Nick Morrison, MD, a Kathleen Gibson, MD, b Michael Vasquez, MD, c Robert Weiss, MD, d Daniel Cher, MD, e Monte Madsen, RVT, RPhS, f and Andrew Jones, MD, g Scottsdale, Ariz; Bellevue, Wash; Buffalo, NY; Hunt Valley, Md; Palo Alto and Santa Rosa, Calif; and Bend, Ore ABSTRACT Objective: Endovenous cyanoacrylate closure (CAC) is a new U.S. Food and Drug Administration-approved therapy for treatment of clinically symptomatic venous reflux in saphenous veins. The device is indicated for the permanent closure of lower extremity superficial truncal veins, such as the great saphenous vein (GSV). Early results from a randomized trial of CAC have been reported previously. Herein we report 1-year outcomes. Methods: There were 222 subjects with symptomatic GSV incompetence randomly assigned to receive either CAC (n ¼ 108) or radiofrequency ablation (RFA; n ¼ 114). After the month 3 visit, subjects could receive adjunctive therapies aimed at treating visible varicosities and incompetent tributaries. Vein closure was assessed at day 3 and months 1, 3, 6, and 12 using duplex ultrasound. Additional study visit assessments included the Venous Clinical Severity Score; Clinical, Etiology, Anatomy, and Pathophysiology classification; EuroQol-5 Dimension; and Aberdeen Varicose Vein Questionnaire. Both time to closure and time to first reopening of the target vein were evaluated using survival curve analysis. Adverse events were evaluated at each visit. Results: Of 222 enrolled and randomized subjects, a 12-month follow-up was obtained for 192 (95 CAC and 97 RFA; total follow-up rate, 192/222 [86.5%]). By month 1, 100% of CAC subjects and 87% of RFA subjects demonstrated complete occlusion of the target vein. By month 12, the complete occlusion rate was nearly identical in both groups (97.2% in the CAC group and 97.0% in the RFA group). Twelve-month freedom from recanalization was similar in the CAC and RFA groups, although there was a trend toward greater freedom from recanalization in the CAC group (P ¼.08). Symptoms and quality of life improved equally in both groups. Most adverse events were mild to moderate and not related to the device or procedure. Conclusions: In patients with incompetent GSVs, treatment with both CAC and RFA results in high occlusion rates. Time to complete occlusion was faster with CAC, and freedom from reopening was higher after CAC. Quality of life scores improved equally with both therapies. (J Vasc Surg: Venous and Lym Dis 2017;5:321-30.) Treatment of chronic venous disorders in the United States has shifted substantially in the past decade from surgical to endovenous treatment. Proven endovenous therapies comprise those that are heat based, including radiofrequency ablation (RFA) and endovenous laser ablation (EVLA). Both have gained widespread adoption and been shown to have high target vein closure rates with good safety profiles. 1 RFA is a segmental treatment process, whereas EVLA is a continuous process requiring the practitioner to gradually retract the laser fiber at a specified rate. Both modalities require the administration of tumescent anesthesia (TA) by multiple injections to infuse a diluted lidocaine and saline mixture into the perivenous space. TA protects the surrounding tissues and provides hypoesthesia when the thermal energy is From the Morrison Vein Institute, Scottsdale a ; Lake Washington Vascular, Bellevue b ; the Venous Institute of Buffalo, Buffalo c ; MD Laser Skin & Vein, Hunt Valley d ; Wild Iris Consulting LLC, Palo Alto e ; Medtronic, Santa Rosa f ; and the Inovia Vein Specialty Center, Bend. g The work was funded by Sapheon Inc. Sapheon was directly involved in the study design, collection, conduct, analysis, and interpretation of the data with the research investigators. Sapheon has since been acquired by Covidien, which was then acquired by Medtronic. Medtronic provided biostatistical and publication coordination support. Clinical Trial registration: NCT01807585. Author conflict of interest: N.M. has been paid a consulting fee by Medtronic, received financial support as an investigator, and is on Medtronic s speakers bureau; he has received an educational grant from mediusa and is on their speakers bureau. K.G. is on the scientific advisory board of Medtronic; receives research support from Medtronic, AngioDynamics, and BTG; and is a consultant for BTG. D.C. is a consultant to Medtronic, helping with writing and statistical analysis. M.V. is on the Medtronic speakers bureau. M.M. was an employee of Sapheon Inc and is currently employed by Medtronic. Presented as a podium presentation at the 2015 European Venous Forum, St. Petersburg, Russia, July 2-4, 2015. Correspondence: Nick Morrison, MD, 6301 S McClintock Dr, Ste 115, Tempe, AZ 85283 (e-mail: nickmorrison2002@yahoo.com). The editors and reviewers of this article have no relevant financial relationships to disclose per the Journal policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 2213-333X Copyright Ó 2017 The Authors. Published by Elsevier Inc. on behalf of the Society for Vascular Surgery. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). http://dx.doi.org/10.1016/j.jvsv.2016.12.005 321

322 Morrison et al Journal of Vascular Surgery: Venous and Lymphatic Disorders May 2017 delivered. Postprocedural pain and bruising are two of themostfrequentproblemsassociatedwiththermal therapies. 2,3 Because RFA had previously been shown to be superior to EVLA with respect to postoperative ecchymosis and pain, 4,5 it was selected as the control intervention for this study. Endovenous delivery of cyanoacrylate (cyanoacrylate closure [CAC]) has become available in the United States, Europe, and elsewhere (VenaSeal; Sapheon, Morrisville, NC). Early evidence from two clinical studies supports the safety and effectiveness of CAC for treatment of great saphenous vein (GSV) reflux, including a single-center 3-year study 6-8 and a multicenter, single-arm 3-year study. 9 Previously, we reported 3-month outcomes of the VenaSeal Sapheon Closure System Pivotal Study (VeClose), a randomized trial of CAC vs RFA. 10 We present herein 12-month VeClose results, focusing on time to complete closure, durability of closure, and symptom scores and quality of life measurements. METHODS Study design. A detailed description of the VeClose trial design, eligibility criteria, and outcomes through month 3 has been previously reported. 10 The VeClose trial is a multicenter, prospective, randomized controlled trial conducted under investigational device exemption from the Food and Drug Administration at 10 study centers in the United States. All investigators were experienced with endovenous treatments and highly experienced RFA users, with backgrounds in general surgery, vascular surgery, emergency medicine, and general medicine. In total, investigators have performed >17,000 RFA procedures and >20,000 laser procedures. All sites obtained central Institutional Review Board approval before enrollment, and all participants provided written informed consent. Patients eligible for inclusion in the trial were adults with symptomatic venous reflux and varicosities (Clinical, Etiology, Anatomy, and Pathophysiology [CEAP] clinical classification of symptomatic C2-C4b) and incompetence of the GSV, with reflux time of $0.5 second assessed in the standing position with duplex ultrasound. Key exclusion criteria were hemodynamically significant reflux of the small saphenous vein or anterior accessory GSV, prior treatment of the target GSV, symptomatic peripheral arterial disease, history of deep venous thrombosis or pulmonary embolism, and aneurysm of the target GSV >12 mm in diameter. The study s objective was to compare the safety and efficacy of CAC vs RFA in patients demonstrating GSV reflux. A Data and Safety Monitoring Board was used to monitor adverse events. Assessments. Before treatment, investigators performed a detailed vascular history and targeted physical examination and completed both CEAP and Venous Clinical Severity Score (VCSS) assessments. 11 Duplex ARTICLE HIGHLIGHTS d Type of Research: Multicenter randomized controlled trial d Take Home Message: At 1 year, cyanoacrylate-based chemical ablation had similar efficacy to radiofrequency ablation thermal ablation in terms of occlusion, durability, and relief from symptoms. d Recommendation: The authors recommend that there is little difference in efficacy between cyanoacrylate treatment and radiofrequency ablation for treatment of symptomatic saphenous reflux. ultrasound of both legs was then performed. Subjects completed the EuroQol-5 Dimension (EQ-5D) quality of life survey 12 and the Aberdeen Varicose Vein Questionnaire (AVVQ). 13 Randomization. Subjects were randomized (1:1) to either CAC (VenaSeal) or RFA (ClosureFast; both now manufactured by Medtronic, Plymouth, Minn). Randomization was stratified by study site and used random block sizes of four or six; assignments were obtained through an automated telephone service connected to a passwordprotected randomization table. Roll-in (training) cases (first 2 cases performed by each site, total of 20 subjects) were treated with CAC and analyzed separately. Devices and procedures. CAC was performed as previously described. 10 Briefly, the target vein was accessed, and a 7F introducer sheath coupled with a 5F delivery catheter was advanced 5 cm caudal to the saphenofemoral junction. With proximal GSV compression using the ultrasound probe, two 0.1-mL aliquots of cyanoacrylate were delivered 1 cm apart with additional hand compression at the treated segment for 3 minutes. Subsequent 0.1-mL aliquots were delivered at 3-cm intervals along the target treatment area, and compression with the ultrasound probe and free hand was performed for 30 seconds at each treated segment. Additional aliquots of 0.1 ml of adhesive were delivered in dilated areas. After the entire length of the vein was treated, the sheath and delivery catheter were removed, and compression was applied to the entry site. A small self-adhesive bandage was applied at the access site. RFA of the target vein was performed using the ClosureFast catheter according to the manufacturer s instructions for use. Perivenous TA was used with RFA but not for CAC. Both groups used postprocedure compression stockings for 3 days continuously and 4 additional days during waking hours. Postoperative study visits. Subjects returned to the investigational center at day 3 and months 1, 3, 6, and 12. Study follow-up will continue to year 3. At follow-up visits, investigators completed VCSS and CEAP scores,

Journal of Vascular Surgery: Venous and Lymphatic Disorders Morrison et al 323 Volume 5, Number 3 and subjects completed the AVVQ and EQ-5D. In addition, Doppler ultrasound of the treated vein was performed by vascular technologists and interpreted by the investigator. The analysis reported herein uses ultrasound interpretations performed by site investigators at months 1, 3, 6, and 12. (Note that a vascular ultrasound core laboratory [VasCore, Boston, Mass] interpreted ultrasound results for the study s primary end point at month 3, with no knowledge of the sites findings at the time of the readings, and core laboratory adjudications agreed 100% with investigator readings.) Ipsilateral adjunctive target vein treatments were disallowed until after the month 3 visit to avoid confounding the study s primary end point at month 3 but were allowed thereafter. End points and statistical methods. The study s previously reported primary end point was complete closure of the target GSV at month 3. Closure was defined as Doppler ultrasound examination (including color flow, compression, and pulsed Doppler) showing closure along the entire treated target vein segment with no discrete segments of patency exceeding 5 cm. Recanalization was defined as openings along the treated segment exceeding 5 cm in length as detected by duplex ultrasound. The likelihood of closure or recanalization was evaluated in several ways. First, time to closure of the target vein was calculated for each subject as the number of days from treatment to first observation of vein closure. Time to recanalization was calculated as the number of days from treatment to first instance of recanalization after the month 1 assessment (because total vein occlusion in RFA may take several weeks, minor [residual microlumen] openings before month 1 in the RFA group were not counted as recanalization). Survival curves were compared using Kaplan- Meier analysis. Differences in the proportion of subjects with complete closure of the target vein at month 12 were assessed using a noninferiority approach with confidence limits calculated using the method of Miettinen and Nurminen. 14 The study s noninferiority margin for the primary end point at month 3 was 10%. The same noninferiority margin was used for the 12-month assessment. If noninferiority was demonstrated, superiority was tested using identical methods. Bootstrapping methods were used to calculate the distribution of the difference in long-term survival rates. Changes from baseline in VCSS, AVVQ, and EQ-5D were compared between groups using repeated-measures analysis of variance. Changes in CEAP category were compared using a Wilcoxon test. Adverse event rates were compared using Fisher exact test. All analyses were performed using R. 15 RESULTS Subject characteristics and disposition. Between March and September 2013, 242 subjects meeting eligibility criteria were enrolled and treated. Twenty subjects Table I. Demographic characteristics of VeClose trial subjects Characteristics CAC (n ¼ 108) RFA (n ¼ 114) Female 83 (77) 93 (82) Nonwhite 6 (6) 8 (7) Age, years 49.0 (27-71) 51 (26-70) Body mass index 27.0 (17.4-44.5) 27.0 (17.0-46.7) Primary symptoms Pain 33 (31) 24 (21) Aching 32 (30) 39 (34) Itching 2 (2) 5 (4) Burning 5 (5) 3 (3) Sensitivity d d Heaviness 14 (13) 16 (14) Swelling 17 (16) 18 (16) Other 4 (4) 7 (6) GSV diameter, mm Mid GSV 4.9 (1.7-9) 5.1 (2.4-11) Proximal GSV 6.3 (3-12) 6.6 (2.8-12) CEAP classification C2 61 (57) 64 (56) C3 32 (30) 36 (32) C4a 13 (12) 12 (11) C4b 2 (2) 2 (2) VCSS 5.5 (2.6) 5.6 (2.6) AVVQ 18.9 (9.0) 19.4 (9.9) EQ-5D TTO 0.935 (0.113) 0.918 (0.116) AVVQ, Aberdeen Varicose Vein Questionnaire; CAC, cyanoacrylate closure; CEAP, Clinical, Etiologic, Anatomic, and Pathophysiologic; EQ-5D, EuroQoL-5 Dimension; GSV, great saphenous vein; RFA, radiofrequency ablation; TTO, time tradeoff; VCCS, Venous Clinical Severity Score. Categorical variables are presented as number (%). Continuous variables are presented as mean (standard deviation) or median (range). (two at each site, which on the basis of experience from previous clinical trials 6,9 was established as the number of cases necessary for initial device familiarity) were roll-in cases treated with CAC, which have been reported separately. 16 Of the subsequent 222 subjects, 108 were assigned to CAC and 114 to RFA. Consistent with subjects from other superficial venous reflux clinical trials, most subjects in this study were women (79%); most (87%) subjects were in the CEAP clinical classification 2 and 3 groups; and baseline characteristics were similar between treatment groups (Table I). Twelve months into the study, 198 subjects (89%) were still participating in the study (192 of whom had a month 12 visit), and 23 (10%) had withdrawn from the study (Fig 1). After month 3, approximately 41% of subjects in both groups underwent one or more adjunctive treatments of the ipsilateral leg. Time to permanent closure. The raw proportion of subjects with complete closure of the target vein was similar at month 12 (92/95 [96.8%] in the CAC group;

324 Morrison et al Journal of Vascular Surgery: Venous and Lymphatic Disorders May 2017 Fig 1. Consolidated Standards of Reporting Trials (CONSORT) flow diagram for subjects involved in the VeClose trial. The patients progress through the VeClose trial is indicated in the diagram, including final N values for each study group. CAC, Cyanoacrylate closure; ND, not defined; RFA, radiofrequency ablation. 93/97 [95.9%] in the RFA group). Using Kaplan-Meier analysis, the number of days to initial closure of the target vein was lower in the CAC group than in the RFA group (Fig 2; P ¼.0123). By month 1, 100% of CAC subjects and 87% of RFA subjects demonstrated complete occlusion of the target vein. By the month 3 visit, the occlusionrateintherfagroupwas94%.bymonth12, the complete occlusion rate was nearly identical in both groups (97.2% in the CAC group and 97.0% in the RFA group). Time to recanalization. The likelihood of recanalization was higher in the RFA group than in the CAC group (Fig 3). By month 3, there were three recanalizations in the CAC group and nine in the RFA group. Recanalization-free survival at month 12 was 97% for the CAC group vs 90.7% for the RFA group (P ¼.08 for superiority, P <.0001 for noninferiority). Subgroup analysis for comparisons of 12-month recanalization-free survival rates was limited by the small number of failures in the CAC group. Regarding adjunctive treatments for target vein closure, one CAC subject had an open vein at month 3 and underwent both phlebectomy and sclerotherapy soon after the month 3 visit. However, examination at months 6 and 12 showed target vein patency. The subject underwent a second session of adjunctive procedures, and closure was documented 367 days after the index procedure. Another CAC subject had persistent opening despite adjunctive treatment, and the third subject had patency at the month 12 visit despite receiving adjunctive treatments before the month 12 visit. In the RFA group, there were five subjects with no closure of the target vein at month 3. Adjunctive treatments in those subjects included phlebectomy and sclerotherapy, and target vein closure occurred in three of the five RFA subjects. Symptom scores and quality of life. During the course of the 12-month follow-up, disease severity as measured by VCSS and impact on quality of life as measured by AVVQ (Fig 4) decreased rapidly and equally in both groups

Journal of Vascular Surgery: Venous and Lymphatic Disorders Morrison et al 325 Volume 5, Number 3 Fig 2. Days to complete occlusion. Shown are the days to complete occlusion of the target vein for the cyanoacrylate closure (CAC; solid line) and radiofrequency ablation (RFA; dotted line) treatment groups. Number at risk is presented along the x-axis: top line, CAC; bottom line, RFA. (P ¼.5694 and P ¼.5536, respectively). Similarly, overall quality of life as measured by EQ-5D (Fig 5) increased by a small and similar amount across groups (P ¼.1645). Adverse events. As previously reported, 10 adverse events in the first 3 months were mild (including phlebitis, paresthesia in the treatment zone, stocking irritation, access site infection, and superficial thrombophlebitis) and occurred at a similar rate between groups. Between months 3 and 12 of follow-up, one subject (originally treated with CAC) had chronic phlebitis in the index leg attributed to the procedure. There were 37 other adverse events reported between month 3 and month 12. These were either phlebitis of nontarget regions or miscellaneous events unrelated to the venous vascular system (Table II). The majority of the adverse events reported were not related to the device or the procedure (eg, pregnancy, pneumonia, Fig 3. Time to recanalization. Shown is the time to recanalization of the target vein for the cyanoacrylate closure (CAC; solid line) and radiofrequency ablation (RFA; dotted line) treatment groups. Number at risk is presented along the x-axis: top line, CAC; bottom line, RFA.

326 Morrison et al Journal of Vascular Surgery: Venous and Lymphatic Disorders May 2017 Fig 4. Change in Venous Clinical Severity Score (VCSS) and Aberdeen Varicose Vein Questionnaire (AVVQ) score. Changes in VCSS (top) and AVVQ score (bottom) are shown by time for the cyanoacrylate closure (CAC; solid line) and radiofrequency ablation (RFA; dotted line) treatment groups. Number at risk is presented at each time point. SE, Standard error. and thyroid cancer), and those adverse events that were inflammatory in nature (eg, phlebitis, tenderness in index leg, cellulitis) were seen at similar rates in both groups (four patients in each group). All of the phlebitis cases were mild. Of the events that were considered related to the device, all resolved and none were considered severe. No thrombus extensions were reported in the CAC group, but two were identified in the RFA group, both of which resolved without sequelae. DISCUSSION In a previous publication of the current trial, 10 we demonstrated CAC to be a safe and effective treatment of GSV reflux, with similar safety and effectiveness at 3 months. However, longer term data with CAC are valuable in establishing its efficacy and long-term benefit to patients, especially in comparison to standard treatments such as RFA. This trial confirms CAC to be safe and effective during a 12-month period, with persistent improvements in benefits to the patient as shown in

Journal of Vascular Surgery: Venous and Lymphatic Disorders Morrison et al 327 Volume 5, Number 3 Fig 5. Change in EuroQol-5 Dimension (EQ-5D). Subjects in the VeClose trial completed the EQ-5D quality of life survey, which measures problems with mobility, self-care, and usual activities. Solid line, Cyanoacrylate closure (CAC); dotted line, radiofrequency ablation (RFA). Number at risk is presented at each time point. SE, Standard error; TTO, time tradeoff. quality of life surveys. Twelve-month occlusion rates were high in both arms of the study (>97%), and the study successfully showed noninferiority of CAC compared with RFA at month 12 (P ¼.0015). Time to complete occlusion of the target vein was lower with CAC than with RFA, and freedom from recanalization with CAC was noninferior (P <.0001) and trends toward superiority (P ¼.08) compared with that with RFA. It is known, however, that recanalization after thermal ablation does not necessarily portend recurrent symptoms, 17 and we did not observe any symptomatic differences at month 12. Compared with RFA, potential advantages of CAC include the lack of necessity for TA, which experience tells us is often uncomfortable for patients, even though in this particular group of patients, pain scores during the procedure were similar (and low) in both groups, as reported previously. 10 This lack of a difference and low scores reflects the experience of the investigators in minimizing the pain associated with delivery of TA and the minimal pain associated with CAC as well as the difficulty in showing significant differences when most patients are at the same end of the pain numeric rating scale (range of 0-10). A much larger cohort would need to be studied to identify a statistical trend in procedural pain. Another potential advantage of CAC is that additional capital equipment, such as an RFA laser generator or tumescent pump, is unnecessary. Although both treatment groups in this study were required to wear compression stockings after the procedure to have similar recovery protocols, previous CAC trials 6,9 did not require the use of postprocedure compression for subjects treated with CAC. Three subjects treated with CAC had incomplete vein closure at some point during follow-up. Veins of two CAC subjects were open at month 12 and one was open by month 3. Possible failure mechanisms for CAC include inadequate dosing for the diameter of the target vein, junctions with refluxing tributaries or perforators, and complex anatomy with multiple junctions along the target vein. In contrast, 19 subjects treated with RFA had open target veins observed during follow-up. Possible failure mechanisms for RFA include inadequate delivery of energy to the target vein, large diameter vein or aneurysmal segments, refluxing tributaries or perforators, and complex anatomy with multiple junctions along the target vein. Twelve-month closure rates were high and nearly identical in both groups. For subjects demonstrating incomplete closure who later demonstrated complete closure, it is possible that adjunctive treatment such as phlebectomy or foam sclerotherapy aided in obtaining closure. In the CAC group, two of the three subjects had persistent opening at month 12 despite adjunctive procedures. Whether adjunctive treatments helped RFA subjects obtain closure is challenging to determine with data from the current study. In clinical practice, adjunctive procedures such as concomitant microphlebectomy and sclerotherapy are common, 18 and further study would be necessary to determine whether

328 Morrison et al Journal of Vascular Surgery: Venous and Lymphatic Disorders May 2017 Table II. Adverse events between 3 and 12 months in the VeClose trial Treatment Adverse event description Related to device Related to procedure Severity Days to adverse event CAC Stomach pains Not related Not related Mild 113 CAC Pain after sclerotherapy Not related Not related Mild 105 CAC Seroma Not related Not related Mild 182 CAC Low back pain Not related Not related Mild 288 CAC Foot pain Not related Not related Mild 184 CAC Sinus pressure Not related Not related Mild 288 CAC Pregnancy Not related Not related Mild 169 CAC Patient pregnant Not related Not related Mild 209 CAC Rash above upper lip Not related Not related Mild 163 CAC Erythema Not related Not related Mild 132 CAC Phlebitis Not related Not related Mild 105 CAC Phlebitis Not related Not related Mild 175 CAC Symptomatic venous stasis Not related Not related Mild 173 CAC Thrombosed veins after sclerotherapy Not related Not related Mild 176 CAC Tenderness over phlebectomy site Not related Not related Moderate 116 CAC Pneumonia Not related Not related Moderate 107 CAC Tenderness left leg Not related Not related Moderate 168 CAC Asthma Not related Not related Moderate 258 CAC Small bowel obstruction Not related Not related Severe 286 CAC Thyroid cancer Not related Not related Severe 232 CAC Pain in right medial thigh Not related Possibly related Moderate 221 CAC Chronic phlebitis Possibly related Definitely related Mild 92 RFA DVT nonindex leg Definitely related Definitely related Moderate 172 RFA EHIT Definitely related Definitely related Moderate 172 RFA Abdomen pain Not related Not related Mild 175 RFA Lower abdomen pain Not related Not related Mild 174 RFA Lower right-sided back pain Not related Not related Mild 269 RFA Cervical radicular pain Not related Not related Mild 277 RFA Hyperpigmentation on index (right) leg Not related Not related Mild 181 RFA Superficial phlebitis Not related Not related Mild 165 RFA Phlebitis Not related Not related Mild 209 RFA Cellulitis Not related Not related Moderate 340 RFA Left hip bone pain Not related Not related Moderate 166 RFA Non-ST-segment elevation MI Not related Not related Severe 308 RFA Left groin discomfort Possibly related Possibly related Mild 97 RFA Phlebitis Possibly related Possibly related Mild 106 CAC, Cyanoacrylate closure; DVT, deep venous thrombosis; EHIT, endovenous heat-induced thrombosis; MI, myocardial infarction; RFA, radiofrequencyablation. adjunctive treatments performed concomitantly with CAC or RFA can improve GSV closure rates in the long term. Our observation of high closure rates at month 3 (before the delivery of adjunctive treatments) suggests that such treatments may reasonably be delayed. The recanalization rate was greater after the RFA procedure than with the CAC procedure. However, as some subjects underwent additional adjunctive treatments, long-term occlusion rates were high in both groups. It is unknown if this was a result of the natural history of vein destruction with RFA compared with the immediate occlusion with CAC or if the adjunctive treatment somehow led to occlusion of more of the RFA-treated GSVs. It is believed that recanalization after RFA occurs because thrombi, a consistent byproduct of thermal ablation techniques, may commonly result in recanalization. However, successful thermal ablation is not dependent on thrombus formation; rather, it depends on vein wall destruction, and unless there has been adequate vein wall destruction, the vein will fail to occlude

Journal of Vascular Surgery: Venous and Lymphatic Disorders Morrison et al 329 Volume 5, Number 3 permanently. 19 With CAC, there is little or no venous thrombosis as the venous lumen is immediately occluded with bonding of the intima by the adhesive, and this may explain the lower recanalization rate with CAC. 9,10 As expected with occlusion of a refluxing GSV, improvements in symptom severity as measured by VCSS, AVVQ, and EQ-5D were similar and high in both groups. Phlebitis was defined broadly in our study and categorized as in the treatment zone or outside it to ensure complete reporting. The onset and details of adverse events have been previously reported. 10 These inflammatory adverse events were more common in the CAC cohort, but the difference was not statistically significant. One potential explanation includes a greater inflammatory effect of the cyanoacrylate implant (the material is designed to cause a nonspecific, mild inflammatory response in the vein to promote vein closure). Posttreatment inflammation was, in all cases, mild, self-limited, and easily treated with a short course of nonsteroidal anti-inflammatory drugs. No deep venous thrombosis or pulmonary embolism occurred in any of the subjects secondary to the target treatment per the investigators. There was one case of gastrocnemius vein thrombus in a subject treated with endovenous laser on a contralateral (nonstudy limb) small saphenous vein 5.5 months after the randomized RFA treatment. CAC does not require TA, and the treatment results in a rapid occlusion of the treated vein. The viscosity and rapid polymerization of the adhesive do not allow it to freely flow with the blood. Once it is injected, it becomes more static, thus minimizing the risk of embolization that might be seen with less viscous adhesives. Thrombus extensions at the saphenofemoral junction were observed in the prior CAC clinical trials and were similar in appearance to the endothermal heat-induced thromboses seen with the endothermal treatments. 20 These extensions were most prevalent in the initial feasibility study that was conducted with a different treatment protocol from that of the current study. 6 One of the purported advantages of nonthermal technologies is the elimination of paresthesias, especially below the knee. There were a total of six cases (three in the CAC group and three in the RFA group), and inquiry of the investigators who had these subjects in the study revealed that the paresthesia was secondary to wearing of the compression hose (determined by the fact that the compression hose were removed in response to the patient s complaints, and the paresthesia resolved in a matter of minutes). This data point was not well defined in the case report forms, and no additional information is available. Strengths of the study include its randomized design; its careful assessment of clinical outcomes, symptom scores, and ultrasound assessments at scheduled time points; and careful tracking of adjunctive procedures. All data were reviewed by external monitors and source verified. Agreement at the 3-month end points between investigator reads and core laboratory reads was high, suggesting that investigator reads at other time points were sufficiently accurate. Limitations of this trial include a modest dropout rate, with month 12 data unavailable for 13 of 108 (12.0%) subjects in the CAC group (9 withdrawn and 4 visits not done) and 19 of 114 (16.6%) subjects in the RFA group (8 withdrawn and 11 visits not done). Blinding, although potentially advantageous, was not feasible because RFA requires TA administration and CAC has characteristic findings on ultrasound. 10 However, the primary study outcome (anatomic closure) was easily judged with ultrasound and is objective. Ultrasound interpretations performed by study investigators could have introduced bias; however, the core laboratory had no knowledge of the sites findings at the time of the readings, and their findings agreed with those of the investigators (there was 100% agreement between investigator reads and core laboratory reads; k statistic was 1.0). To minimize confounding due to non-device-related postintervention factors, subjects in both groups were asked to use compression stockings after the index procedure for 7 days. This was done solely for the trial, but it was not done for two prior studies of CAC. 6,9 Whether compression stockings improve complete occlusion rates could be the subject of further study. CONCLUSIONS In this study, both CAC and RFA were associated with high rates of target vein occlusion at month 12. Both time to closure and recanalization rates were lower with CAC than with RFA. Improvements in generic and disease-specific quality of life scores were similar across both groups up to month 12; thus, the benefits of CAC appear to be as durable as those of RFA in the midterm. Follow-up of subjects will continue to year 3. AUTHOR CONTRIBUTIONS Conception and design: NM Analysis and interpretation: NM, KG, AJ, MV, RW, DC, MM Data collection: NM, KG, AJ, MV, RW, MM Writing the article: NM, KG, AJ, MV, RW, DC, MM Critical revision of the article: NM, KG, AJ, MV, RW, DC, MM Final approval of the article: NM, KG, AJ, MV, RW, DC, MM Statistical analysis: NM, DC, MM Obtained funding: NM Overall responsibility: NM REFERENCES 1. van den Bos R, Arends L, Kockaert M, Neumann M, Nijsten T. Endovenous therapies of lower extremity varicosities: a meta-analysis. J Vasc Surg 2009;49:230-9.

330 Almeida Journal of Vascular Surgery: Venous and Lymphatic Disorders May 2017 2. Anwar MA, Lane TR, Davies AH, Franklin IJ. Complications of radiofrequency ablation of varicose veins. Phlebology 2012;27(Suppl 1):34-9. 3. Dexter D, Kabnick L, Berland T, Jacobowitz G, Lamparello P, Maldonado T, et al. Complications of endovenous lasers. Phlebology 2012;27(Suppl 1):40-5. 4. Rasmussen LH, Lawaetz M, Bjoern L, Vennits B, Blemnigs A, Eklof B. Randomized clinical trial comparing endovenous laser ablation, radiofrequency ablation, foam sclerotherapy and surgical stripping for great saphenous varicose veins. Br J Surg 2011;98:1079-87. 5. Almeida JI, Kaufman J, Gockeritz O, Chopra P, Evans M, Hoheim D, et al. Radiofrequency endovenous ClosureFAST versus laser ablation for the treatment of great saphenous reflux: a multicenter, single-blinded, randomized study (RE- COVERY study). J Vasc Interv Radiol 2009;20:752-9. 6. Almeida JI, Javier JJ, Mackay EG, Bautista C, Proebstle TM. First human use of cyanoacrylate adhesive for treatment of saphenous vein incompetence. J Vasc Surg Venous Lymphat Disord 2013;1:174-80. 7. Almeida JI, Javier JJ, Mackay EG, Bautista C, Cher D, Proebstle T. Two-year follow-up of first human use of cyanoacrylate adhesive for treatment of saphenous vein incompetence. Phlebology 2015;30:397-404. 8. Almeida JI, Javier JJ, Mackay EG, Bautista C, Cher D, Proebstle TM. Three-year follow-up of first human use of cyanoacrylate adhesive for treatment of saphenous vein incompetence. J Vasc Surg Venous Lymphat Disord 2015;3:125. 9. Proebstle TM, Alm J, Dimitri S, Rasmussen L, Whiteley M, Lawson J, et al. The European multicenter cohort study on cyanoacrylate embolization of refluxing great saphenous veins. J Vasc Surg Venous Lymphat Disord 2015;3:2-7. 10. Morrison N, Gibson K, McEnroe S, Goldman M, King T, Weiss R, et al. Randomized trial comparing cyanoacrylate embolization and radiofrequency ablation for incompetent great saphenous veins (VeClose). J Vasc Surg 2015;61:985-94. 11. Vasquez MA, Munschauer CE. Venous Clinical Severity Score and quality-of-life assessment tools: application to vein practice. Phlebology 2008;23:259-75. 12. EuroQol Group. EuroQolda new facility for the measurement of health-related quality of life. Health Policy 1990;16: 199-208. 13. Garratt AM, Macdonald LM, Ruta DA, Russell IT, Buckingham JK, Krukowski ZH. Towards measurement of outcome for patients with varicose veins. Qual Health Care 1993;2:5-10. 14. Miettinen O, Nurminen M. Comparative analysis of two rates. Stat Med 1985;4:213-26. 15. R Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2013. Available at: http://www.r-project.org. 16. Kolluri R, Gibson K, Cher D, Madsen M, Weiss R, Morrison N. Roll-in phase analysis of clinical study of cyanoacrylate closure for incompetent great saphenous veins. J Vasc Surg Venous Lymphat Disord 2016;4:407-15. 17. Theivacumar NS, Dellagrammaticas D, Darwood RJ, Mavor AI, Gough MJ. Fate of the great saphenous vein following endovenous laser ablation: does re-canalisation mean recurrence? Eur J Vasc Endovasc Surg 2008;36:211-5. 18. Welch HJ. Endovenous ablation of the great saphenous vein may avert phlebectomy for branch varicose veins. J Vasc Surg 2006;44:601-5. 19. Puggioni A, Marks N, Hingorani A, Shiferson A, Alhalbouni S, Ascher E. The safety of radiofrequency ablation of the great saphenous vein in patients with previous venous thrombosis. J Vasc Surg 2009;49:1248-55. 20. Rhee SJ, Cantelmo NL, Conrad MF, Stoughton J. Factors influencing the incidence of endovenous heat-induced thrombosis (EHIT). Vasc Endovascular Surg 2013;47:207-12. Submitted Jul 15, 2016; accepted Dec 13, 2016. INVITED COMMENTARY Jose I. Almeida, MD, Miami, Fla Compared with standard surgery, endovenous thermal ablation has fewer complications, quicker recovery, and improved quality of life and cosmetic outcomes. However, endovenous thermal ablation requires tumescent anesthesia and carries a small risk of thermal-related complications, such as paresthesia, prolonged pain, and skin burn. The insertion of tumescent fluid can also be uncomfortable. Because of these drawbacks, interest in nonthermal alternatives has increased. Nonthermal alternatives require neither tumescent anesthesia nor the purchase of expensive thermal energy delivery systems. Cyanoacrylate closure (CAC), with a high-viscosity n-butyl cyanoacrylate, offers rapid polymerization and target vein fibrosis. A similar n-butyl cyanoacrylate (TRUFILL; DePuy Synthes, West Chester, Pa) is a safe, proven, liquid embolic system for treatment of arteriovenous malformations; >15,000 patients have been treated. Medical skin adhesives (Dermabond; Ethicon, Somerville, NJ) are composed of 2-octyl cyanoacrylate and behave differently. Early evidence from several feasibility studies supports the safety and effectiveness of this same proprietary high-viscosity n-butyl cyanoacrylate for treatment of great saphenous vein reflux; and the authors of this study previously reported 3-month outcomes of a randomized trial of CAC vs radiofrequency ablation (RFA). These 12-month VenaSeal Sapheon Closure System Pivotal Study (VeClose) results focus on time to complete closure, durability of closure, symptom scores, and quality of life measurements. Of 222 enrolled and randomized subjects, a 12-month follow-up was obtained for 95 CAC patients and 97 RFA patients (192/222 [86.5%]). By month 1, 100% of CAC subjects and 87% of RFA subjects demonstrated complete occlusion of the target vein. By month 12, the complete occlusion rate was nearly identical in both groups (97.2% in the CAC