Great saphenous vein patency and endovenous heat-induced thrombosis after endovenous thermal ablation with modified catheter tip positioning

Transcription

1 Great saphenous vein patency and endovenous heat-induced thrombosis after endovenous thermal ablation with modified catheter tip positioning Omar P. Haqqani, MD, Calin Vasiliu, MD, Thomas F. O Donnell, MD, and Mark D. Iafrati, MD, Boston, Mass Objective: Thrombus extension into a deep vein after superficial venous thermoablation remains a unique complication in the treatment of superficial reflux disease of the great saphenous vein (GSV). In this study, we evaluate if catheter tip positioning or vein diameter correlate with the length of proximal patent segment of GSV after ablation and more caudal catheter positioning decreases the incidence of proximal thrombus extension into the femoral vein. Methods: This was a prospective study conducted from January 2008 to November 2009 of 73 patients undergoing radiofrequency ablation (RFA). Preoperative, intraoperative, and postoperative duplex ultrasound scans were obtained using standard protocols to establish reflux and target vein diameter. Intraoperative measurements were performed from the catheter tip to the femoral vein margin. Duplex ultrasound studies were obtained between 5 and 7 days after the procedure, with 1-month follow-up. The relationship between catheter tip positioning and vein diameter with the length of the proximal patent GSV segment after ablation and the incidence of proximal thrombus extension were analyzed. Results: RFA was performed in 73 patients. Intraoperatively, the mean catheter tip positioning distance was 2.75 cm (range, cm) from the saphenofemoral junction (SFJ), with 93% of the catheters placed within 2.6 to 2.9 cm of the femoral vein. The GSV mean diameter at the SFJ was 0.90 cm (range, cm). After RFA, all GSVs were occluded, with a mean residual patent proximal GSV length of 1.17 cm (range, cm). Two patients demonstrated thrombus extension from the SFJ into the femoral vein for a 2.7% incidence of endovenous heat-induced thrombosis. Conclusions: In patients undergoing RFA for saphenous reflux, neither catheter tip positioning nor vein diameter correlates with the length of the proximal patent segment of GSV after ablation. In addition, catheter positioning does not decrease the incidence of proximal thrombus extension into the femoral vein. (J Vasc Surg 2011;54:10S-17S.) From the Department of Vascular Surgery, Cardiovascular Center, Tufts Medical Center. Competition of interest: none. Reprint requests: Omar P. Haqqani, MD, Department of Vascular Surgery, the Cardiovascular Center, Tufts Medical Center, 800 Washington St, Boston, MA ( ohaqqani@tuftsmedicalcenter.org). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest /$36.00 Copyright 2011 by the Society for Vascular Surgery. doi: /j.jvs S Endovenous thermal ablation (ETA) is a highly effective and common treatment of superficial reflux disease of the great saphenous vein (GSV) and is also used by many for small saphenous incompetence. 1-3 ETA procedures for the GSV have increased eightfold during the last 4 years in the United States. This minimally invasive approach uses laser or radio frequency energy to provide thermal energy directly to the vein wall. 4 Occlusion of the GSV occurs as a result of thermal destruction of the intima and breakdown of collagen in the media. The device produces precise tissue destruction with endothelial denudation, denaturing of the media and intramural collagen, with subsequent fibrotic seal of the vein lumen. 3 Subsequently, shrinkage and contraction of the vein wall ensues. 5 One of the unique complications of this treatment can be the formation of thrombus in the proximal GSV, which may extend for varying degrees into the femoral vein lumen. This complication has been termed endovenous heat-induced thrombosis (EHIT) and can develop with either method of thermal ablation. 6 Recent device modifications of the ETA technique for radiofrequency ablation (RFA) have provided a 7-cm-long heating element with the ClosureFast VNUS (Covidien, Mansfield, Mass), which allows higher treatment temperatures (120 C) and reduced treatment times. Each 7-cm segment of vein is heated for 20 seconds after the initial double application to the most proximal vein segment. 1 The original recommendation from the manufacturer was that the tip of the RFA ClosureFast catheter should be positioned 0.5 to 1 cm from the femoral vein; however, this distance was associated with a higher-than-expected extension of thrombus from the saphenofemoral junction (SFJ) into the femoral vein. 7 To lower the incidence of thrombus formation and extension into the femoral vein, instructions for use of RFA were modified. The current recommended distance for catheter tip positioning (CTP) from the SFJ was increased to 2 cm. 7 Currently, no prospective studies have addressed how modification in distance of the RFA ClosureFast catheter tip from the femoral vein might change the length of the occlusive process from the femoral vein or the incidence of EHIT. We evaluated if CTP or vein diameter correlate with

2 JOURNAL OF VASCULAR SURGERY Volume 54, Number 19S Haqqani et al 11S the length of proximal patent segment of the GSV after ablation and if CTP decreases the incidence of proximal thrombus extension into the femoral vein. METHODS From January 2008 to November 2009, data were collected prospectively on 73 patients undergoing RFA by two surgeons who were experienced with the ETA technique. Preoperative, intraoperative, and postoperative duplex ultrasound (DUS) scans were obtained in an Intersocietal Commission for the Accreditation of Vascular Laboratories-accredited laboratory by registered vascular laboratory technologists (RVTs) who were blinded by the results of each other s work. In our practice, the ultrasound technicians who perform preoperative and postoperative evaluations are different from the ultrasound technician who is dedicated to the ETA procedures. Selection criteria of patients for RFA. Consecutive patients who met the criteria for treatment of saphenous reflux were the basis of this report. By CEAP C classification, patients ranged from C 2 to C 5. Patients with symptomatic saphenous or tributary veins were required to have been treated with conservative therapy for at least 3 months before consideration of RFA. Medical therapy included compression support hose (20-30 mm Hg), leg evaluation, and avoidance of prolonged standing. Persistent symptoms that interfered with daily living were required for RFA treatment. Patients with lipodermatosclerosis and venous ulcers were initially treated with elastic compression (30-40 mm Hg) for a minimum of 3 months before they were eligible for RFA. Saphenous dilatation and reflux was determined by DUS imaging in our vascular laboratory following a standardized protocol. Reflux in the proximal GSV was measured with the patient supine and standing using the pneumatic cuff inflation/deflation technique. Reflux 1.5 seconds in either position was required for GSV RFA treatment eligibility. In addition, the target vein in the refluxing segments had to have a minimum diameter of 3 mm. The diameter of the GSV was measured in the standing upright position. Preoperative evaluation. Preoperative and postoperative DUS studies were performed on a GE DUS machine (GE Logiq 9, Greenville, SC). Our method has been discussed in previous publications 8 and is a modification of the van Bemmelen technique. The diameter of the GSV just below the SFJ was recorded from the preoperative DUS study. Reflux in all patients was 1.5 seconds. Intraoperative evaluation. After advancement of the RFA catheter to the region of the saphenofemoral junction under direct DUS visualization (GE Logiq E08), measurements were taken from the catheter tip to the femoral vein margin. This measurement was derived from a longitudinal line from the catheter tip projected to a curvilinear line from the superomedial margin of the femoral vein to the inferomedial margin of the femoral vein in a longitudinal view (Fig 1). Two repeat measurements were taken after Fig 1. Duplex ultrasound image demonstrates measurements derived from a longitudinal line from the catheter tip position (CTP) projected to a curvilinear line from the superomedial margin of the femoral vein (FV) to the inferomedial margin of the FV in a longitudinal view. GSV, Great saphenous vein. Table I. Table of Kabnik classification Class I II III IV Criteria Venous thrombosis to the superficial deep vein saphenofemoral junction, not extending into the deep venous system Nonocclusive venous thrombosis projecting the deep venous system, whereby the cross-sectional area of thrombus in the femoral vein is 50% As above, but with a cross-sectional area in the femoral vein of the thrombus 50% Occlusive deep vein thrombus the initial infusion of the tumescent fluid and then immediately before the initiation of thermal energy. DUS studies were obtained between 5 and 7 days after RFA treatment by an RVT, who was unaware of the intraoperative CTP. This technologist measured the distance from the edge of the occlusive process to the margin of the femoral vein by the same process as was used intraoperatively. Finally, if present, thrombus extension into the femoral vein was graded by the Kabnik classification (Table I). 6 ETA technique. In 69 patients, the ETA procedure was done in the procedure room under local anesthesia with oral diazepam sedation, and in 4 patients (5%) with tributary removal under general anesthesia in the operating room. Venous access with a micropuncture needle was obtained under DUS visualization at the below knee or periknee position. The catheter was advanced under direct DUS visualization to the saphenofemoral region. Blind positioning of the catheter was avoided, as was advancement of the catheter into the femoral vein before it was

3 12S Haqqani et al JOURNAL OF VASCULAR SURGERY December Supplement 2011 Fig 2. Distribution of length of catheter tip positioning from the saphenofemoral junction (SFJ). positioned in the proximal GSV. The catheter tip was positioned from 2.5 to 2.9 cm away from the femoral vein. J wires were used in a few patients when difficulty in advancing the catheter into the GSV was encountered. After selection of the catheter tip position, generous amounts of tumescent fluid ( ml), composed of 30 or 40 ml of 1% Xylocaine (Astra Pharmaceutical Company, Worcester, Mass) with epinephrine (depending on body weight) and 10 ml of sodium bicarbonate in 500 ml of normal saline, were infused by a Klein pump (HK Surgical, San Clemente, Calif) around the GSV. Initial infusion into the saphenofemoral region was followed by repeat DUS measurement of the CTP. Care was taken to use copious amounts of tumescent solution to compress the proximal GSV, particularly proximal to the catheter tip. Tumescent fluid in this region was administered with the patient in the Trendelenburg position. A third CTP measurement was performed before thermal treatment was initiated. The first segment of the GSV received two cycles of radiofrequency energy. The remaining segments generally received one cycle, but larger-diameter veins underwent a double application of radiofrequency energy. A completion intraoperative DUS scan was performed to assess the patency of the femoral vein and assess if the GSV was occluded by both dynamic and morphologic criteria. Those patients who underwent RFA in the operating room underwent concomitant removal of varicose tributaries that were judged by the surgeon to be too large in diameter for sclerotherapy. After RFA, direct compression was applied by gauze bolster along the course of the GSV in the thigh, and the leg was wrapped with an ankle to high inguinal area in an elastic bandage. Patients were encouraged to wear the compression dressings for at least 48 hours and replace the thigh component with compression shorts and the calf component with a class II compression stocking (20-30 mm Hg). Anticoagulation. All patients received perioperative anticoagulation with a dose of regular heparin administered subcutaneously ( units). Patients who were taking aspirin were maintained on that medication, and all other patients received aspirin postoperatively. Follow-up. Patients returned within 1 week and 1 month for follow-up. DUS imaging was performed during the initial follow-up to evaluate vein status. CEAP clinical classes were recorded. RESULTS Patient characteristics. Patients (51 women, 22 men) were a mean age of 54 years. The mean CEAP C classification was C 3. Intraoperatively, the CTP mean distance was 2.75 cm (range, cm) from the SFJ, with 93% of the catheters placed within 2.6 to 2.9 cm from the femoral vein (Fig 2). The GSV mean diameter at the SFJ was 0.90 cm (range, cm) and only 11% were 0.5 cm (Fig 3). DUS examination at 5 to 7 days after RFA treatment showed that all GSVs were occluded. The mean residual patent proximal GSV length measured from the thrombus process tip to the femoral vein was 1.17 cm (range, cm; Fig 4). The superficial epigastric vein was patent on follow-up DUS imaging in all patients. Two patients demonstrated thrombus, which extended from the SFJ into the femoral vein, for a 2.7% incidence of EHIT. On transverse views, these two patients were classified as Kabnik type II thrombi. The DUS scan showed the typical mildly echoreflective thrombus, which distinguishes EHIT from the usual echolucent characteristics of acute deep venous thrombosis (DVT) in a peripheral vein (Figs 5 and 6). In both of these cases of EHIT, the CTP was 2.6 and 2.8 cm intraoperatively at the initiation of RFA. These

4 JOURNAL OF VASCULAR SURGERY Volume 54, Number 19S Haqqani et al 13S Fig 3. Distribution of diameters of the great saphenous vein (GSV) at the saphenofemoral junction. Fig 4. Residual patent proximal great saphenous vein (GSV) length after radiofrequency ablation. two patients with thrombus extension into the femoral vein received anticoagulation or 2 weeks with low-molecularweight heparin, and a repeat DUS revealed complete resolution of thrombus. There was no correlation between the intraoperative CTP and the length of the residual patent GSV postoperatively (Fig 7). In addition, the preoperative GSV diameter failed to correlate with the length of the proximal patent GSV at follow-up, which was predominately in the 0.5- to 1.5-cm segment from the femoral vein (Fig 8). Less than 10% of the patients had a length of residual GSV 2 cm from the femoral vein. Anecdotally, on postoperative DUS imaging, several patients with a frozen valve cusp, usually on the inferior margin at the SFJ, were observed to have thrombus formation in that segment up to the valve cusp, with preservation of flow in the contralateral wall up into the superficial epigastric vein. In none of these patients did the occlusive process occupy the entire lumen of the GSV until the occlusive point. DISCUSSION This study demonstrates that after ETA by RFA, there is no correlation between the intraoperative CTP and the early postoperative length of patent proximal GSV. Moreover, our incidence of EHIT was comparable to other series that used RFA or laser. The hypothesis that the length of residual patent GSV correlated with the initial pretreatment CTP was not upheld by our study because the post-treatment residual patent GSV was highly variable. In theory, the placement of the catheter tip at a position 2.5 cm from the femoral vein would avoid or at least minimize the incidence of EHIT. Other studies with RFA

5 14S Haqqani et al JOURNAL OF VASCULAR SURGERY December Supplement 2011 Fig 5. Duplex ultrasound image shows typical mildly echoreflective thrombus, which distinguishes endovenous heat-induced thrombosis from the usual echolucent characteristics of acute deep venous thrombosis in a peripheral vein. SFJ, Saphenofemoral junction. Fig 7. Distribution of the intraoperative catheter tip position and the length of the residual patent great saphenous vein (GSV) postoperatively. SFJ, Saphenofemoral junction. Fig 6. Duplex ultrasound image shows typical mildly echoreflective thrombus, which distinguishes endovenous heat-induced thrombosis from the usual echolucent characteristics of acute deep venous thrombosis in a peripheral vein. SFJ, Saphenofemoral junction. corroborate our findings that CTP (even if at a standardized position) did not correlate with the early postablation length of patent GSV. Furthermore, there exists a paucity of data in the literature about the residual patent GSV and its potential role in the development of EHIT as well as recanalization of the GSV. Because this study was intended to assess the effect of CTP to the manufacturer s recommended distance, this study cannot inform about the effects of CTP at a shorter distance ( 2 cm) from the femoral vein. Thus, the relationship between a shorter CTP and the length of patent GSV Fig 8. Distribution of diameters of the great saphenous vein (GSV) between the preoperative GSV diameter and the length of proximal patent GSV at follow-up. cannot be determined. Moreover, the variability in the CTP in our study emphasizes the difficulty of placing a catheter at an exact and standardized position to a millimeter, as others have pointed out. However, 93% of catheter tips were placed between 2.6 and 2.9 cm from the femoral vein. Puggioni et al 9 from Mayo Clinic reported their experience with endovenous laser treatment (EVLT) and firstgeneration RFA catheters in 53 limbs as well as in 77

6 JOURNAL OF VASCULAR SURGERY Volume 54, Number 19S Haqqani et al 15S Table II. Summary of the incidence of endovenous heat-induced thrombosis in various series by type of thermal ablation Author Year Procedure Duplex FU (days) Complication rates Welch RFA 7 0 Vasquez et al RFA 4 0.2% Passman et al RFA NA 0.8% Ravi et al EVLT/RFA % Merchant et al RFA NA 0.4% Nicolaides RFA NA 1.9% Weiss and Weiss RFA 7 0 Gradman EVLT: 10,290 NA 0.15% RFA: 6275 DVT: 34%; PE, 2% Knipp et al EVLT NA DVT: 2.2%; Thrombus extension: 7.8% Hingorani et al RFA 10 (mean) 16% Mozes et al EVLT 7 (mean) 2.3% EVLT, Endovenous laser treatment; FU, follow-up; NA, no regular duplex follow-up, PE, pulmonary embolism; RFA, radiofrequency ablation. subsequent limbs with EVLT. Only 11 of 53 limbs of the RFA group and 54 of 77 of the EVLT group underwent postprocedural DUS scanning because this protocol was initiated when recent publications reported thrombotic complications after RFA. In the limbs that were actually scanned, the authors observed that the length of residual patent GSV averaged 9.5 mm, with no difference between RFA and laser. Proebstle et al 10 observed that after EVLT, the GSV was occluded to 4 cm of the femoral vein in 97% of patients, with no evidence of thrombus extension. Subsequently, the same investigators reported their experience with the newer ClosureFast RFA catheters. 1 In that prospective observational study of 194 patients with CTP just inferior to the ostium of the superficial epigastric vein, or at least 1 to 2 cm below the SFJ, the length of the patent stump at 6 months averaged cm. No early DUS measurements of the length of patent GSV were provided. One previous prospective study of ETA by laser correlated the distance of the catheter tip from the femoral vein with the length of the residual GSV stump. Pleister et al 11 examined with DUS imaging 50 patients undergoing laser ablation of the GSV with the laser tip positioned an average of 2.7 cm (range, cm) from the femoral vein. The length of the residual stump was 1.5 cm at 24 hours and 1.3 cm at 3 months. There were no instances of EHIT. Our incidence of EHIT was comparable to previous series where this was reported. The newer ClosureFast RFA catheter uses higher heating temperatures than the original catheter, and the thermal injury characteristics with the newer catheter likely differ from the original catheter. Therefore, the incidence of EHIT cannot be extended from the first generation ClosurePlus RFA catheter to the newer ClosureFast RFA catheter. The earlier Hingorani et al 12 report of a 16% incidence of extension of thrombus into the femoral vein with the first-generation RFA catheter initiated a flurry of correspondence about the ETA technique and thrombotic complications. Their 26% incidence of EHIT with the 8F catheter suggested mechanical trauma as a major cause. Puggioni et al 9 reported their experience with thrombotic complications after ETA by laser and the first-generation RFA catheter. Only 50% of the 130 limbs underwent posttreatment DUS scanning for EHIT. Thrombus protrusion into the femoral vein occurred in three limbs with EVLT (2.3%), and one patient underwent placement of a temporary inferior vena cava filter for a floating thrombus. A review of thrombotic complications after RFA for first-generation catheters or laser treatment of the GSV in series published up to 2004 by Mozes et al 13 from Mayo Clinic showed that EHIT averaged 0.3% for laser and 2.1% for RFA. No thrombus extension was reported in 7 of 9 series of EVLT or in 7 of 12 series of RFA. Two series reported one pulmonary embolus each, for a procedural incidence of 0.16% (Table II). Proebstle et al 1 conducted a prospective observational study of 252 limbs undergoing RFA with the new ClosureFast catheter and reported immediate occlusion of the vein in 100% of the 252 limbs, as assessed at 72 hours, and 99.6% at 6 months. No limb exhibited EHIT. 1 Laser treatment of the GSV is also associated with EHIT and appears in the same range as this current series (Table II). Recent studies by Lawrence et al 14 reported closure rates of 99.6% for CTP between 2.0 and 2.5 cm from the SFJ among 500 patients. Thirteen patients (2.6%) experienced thrombus extension into the femoral vein or adherent to its wall, which was treated with anticoagulation. All of these patients had thrombus retraction to the level of the saphenofemoral junction (SFJ) in an average of 16 days with anticoagulation. CTP was reported between 2.0 and 2.5 cm. No femoral DVT occurred in the series; however, there was a significantly higher rate of proximal thrombus extension in those patients with a history of DVT and those with a GSV diameter of 8 mm. Retrospective analysis by Marsh et al 15 was performed on 2470 cases of RFA and 350 of endovenous laser abla-

7 16S Haqqani et al JOURNAL OF VASCULAR SURGERY December Supplement 2011 tion. Post-RFA, DVT was identified in 17 limbs (0.7%), and 4 were EHIT (0.2%). Concomitant small saphenous vein ligation and stripping were risk factors for calf DVT (odds ratio, 3.4; 95% confidence interval, , P.036), possibly due to an older patient group with more severe disease. Four DVTs (1%) were identified after endovenous laser ablation, of which three were EHIT (0.9%). 15 Most investigators believe that EHIT is related to direct thermal damage of the endothelium in the most proximal GSV and possibly the femoral vein by the thermal catheter. 5 From their previous experience with laser ablation, Proebstle determined two measurements of thermal energy for RFA by ClosureFast: linear endovenous energy density (LEED), which is the amount of delivered energy along the vein, and endovenous fluence equivalent (EFE), which is the energy delivered to the inner vein wall surface. 16 For LEED 80 J/cm has been recommended by Timperman 17 as the target for effective closure of the GSV by laser. Proebstle et al 1 calculated that the new Closure- Fast RFA catheter delivers J/cm for the first segment of the GSV with a double application of RFA, whereas a single application for the remaining segments delivered J/cm of energy. The ClosureFast catheter delivers 120 C for two 20-second cycles and a LEED that is not dissimilar to laser. In our initial experience with ClosureFast RFA catheter position, we were influenced by the recommendations for EVLT where the catheter is positioned 1.5 to 2 cm from the femoral vein. 18 Now it is our practice, as in this present study, that the CTP is at least 2.5 cm from the femoral vein. The laser and ClosureFast RFA catheters both have a forward thrust of heat associated with the technique. This flume is directly visible on intraoperative DUS visualization of the RFA technique and most likely represents expulsion of air and saline contained within the internal channel of the catheter as vapor. Figs 9 and 10 demonstrate that this flume extends to 1.7 cm. Whether this causes the thermal damage and thrombus formation in the proximal GSV has not been settled by this study. Our technique attempts to limit the extension of this flume by the administration of generous amounts of tumescent fluid proximal around the catheter to compress, if not occlude, the proximal vein segment between the catheter tip and femoral vein. In addition, direct compression is applied proximally during the first two cycles of RFA for the proximal segment. Certainly, from our experience, compression of the proximal GSV is extremely difficult in anyone other than thin patients and is facilitated by the Trendelenburg position, which permits gravitational retraction of the pannus. Besides thermal injury, mechanical trauma can injure the distal saphenous and particularly the femoral vein. Advancement of the catheter into the femoral vein and then withdrawal, which has been frequently used with the laser technique, was not used in this series. Moreover, there was an extremely low incidence of J wire use in radiofrequency Fig 9. Duplex ultrasound image shows the forward thrust of heat, depicted as a flume associated with the ClosureFast radiofrequency ablation catheter. Fig 10. Duplex ultrasound image shows the ClosureFast radiofrequency ablation catheter with flume, likely representing expulsion of air/saline contained within the internal channel of the catheter as vapor extending to 1.7 cm. CFV, Common femoral vein; CTP, catheter tip positioning; GSV, great saphenous vein. catheter advancement, which is a potential source of intimal injury. This technique was not used in the two patients who developed EHIT. Thus, these two potential causes of direct endothelial trauma to the proximal GSV or femoral vein were minimized. We emphasize that all patients in this series received anticoagulation with intraoperative subcuta-

8 JOURNAL OF VASCULAR SURGERY Volume 54, Number 19S Haqqani et al 17S neous heparin and postoperative aspirin. How this modifies the length of patent GSV or EHIT vs no anticoagulation was not settled by the present study. CONCLUSIONS The data from our experience with 73 patients who underwent RFA for saphenous reflux suggest that neither CTP nor vein diameter correlate with the length of the proximal patent segment of GSV after ablation. In addition, catheter positioning does not decrease the incidence of proximal thrombus extension into the femoral vein. AUTHOR CONTRIBUTIONS Conception and design: TO, MI, CV Analysis and interpretation: OH, TO, CV Data collection: CV, TO Writing the article: OH, TO, MI Critical revision of the article: OH, TO, MI Final approval of the article: OH, TO, MI Statistical analysis: TO, MI, CV Obtained funding: Not applicable Overall responsibility: OH REFERENCES 1. Proebstle TM, Vago B, Alm J, Göckeritz O, Lebard C, Pichot O. Treatment of the incompetent great saphenous vein by endovenous radiofrequency powered segmental thermal ablation: first clinical experience. J Vasc Surg 2008;47: Ravi R, Trayler EA, Barrett DA, Diethrich EB. Endovenous thermal ablation of superficial venous insufficiency of the lower extremity: single-center experience with 3000 limbs treated in a 7-year period. J Endovasc Ther 2009;16: Roth SM. Endovenous radiofrequency ablation of superficial and perforator veins. Surg Clin North Am 2007;87: , xii. 4. Almeida JI, Kaufman J, Göckeritz O, Chopra P, Evans MT, Hoheim DF, et al. Radiofrequency endovenous ClosureFAST versus laser ablation for the treatment of great saphenous reflux: a multicenter, singleblinded, randomized study (RECOVERY study). J Vasc Interv Radiol 2009;20: Fan CM, Rox-Anderson R. Endovenous laser ablation: mechanism of action. Phlebology 2008;23: Kabnick LS, Agis H, Almeida J, Moritz M, Giorigio S. Endovenous heat induced thrombus (EHIT) following endovenous vein obliteration: to treat or not to treat. A new thrombotic classification. In: Third International Vein Congress; 2005.Miami, FL, April 14-16, VNUS Medical Technologies. Important update regarding the use of the VNUS Closure Fast Catheter; Available at: vnus.com/pdf/closurefast-ifu.pdf. 8. Walsh JC, Bergan JJ, Beeman S, Comer TP. Femoral venous reflux abolished by greater saphenous vein stripping. Ann Vasc Surg 1994;8: Puggioni A, Kalra M, Carmo M, Mozes G, Gloviczki P. Endovenous laser therapy and radiofrequency ablation of the great saphenous vein: analysis of early efficacy and complications. J Vasc Surg 2005; 42: Proebstle TM, Moehler T, Herdemann S. Reduced recanalization rates of the great saphenous vein after endovenous laser treatment with increased energy dosing: definition of a threshold for the endovenous fluence equivalent. J Vasc Surg 2006;44: Pleister I, Evans J, Vaccaro PS, Satiani B. Natural history of the great saphenous vein stump following endovenous laser therapy. Vasc Endovasc Surg 2008;42: Hingorani AP, Ascher E, Markevich N, et al. Deep venous thrombosis after radiofrequency ablation of greater saphenous vein: a word of caution. J Vasc Surg 2004;40: Mozes G, Kalra M, Carmo M, Swenson L, Gloviczki P. Extension of saphenous thrombus into the femoral vein: a potential complication of new endovenous ablation techniques. J Vasc Surg 2005;41: Lawrence PF, Chandra A, Wu M, Rigberg D, DeRubertis B, Gelabert H, et al. Classification of proximal endovenous closure levels and treatment algorithm. J Vasc Surg 2010;52: Marsh P, Price BA, Holdstock J, Harrison C, Whiteley MS. Deep vein thrombosis (DVT) after venous thermoablation techniques: rates of endovenous heat-induced thrombosis (EHIT) and classical DVT after radiofrequency and endovenous laser ablation in a single centrer. Eur J Vasc Endovasc Surg 2010;40: Proebstle TM, Krummenauer F, Gül D, Knop J. Nonocclusion and early reopening of the great saphenous vein after endovenous laser treatment is fluence dependent. Dermatol Surg 2004;30: Timperman PE. Prospective evaluation of higher energy great saphenous vein endovenous laser treatment. J Vasc Interv Radiol 2005;16: Cheshire N, Elias SM, Keagy B, Kolvenbach R, Leahy AL, Marston W, et al. Powered phlebectomy (TriVex) in treatment of varicose veins. Ann Vasc Surg 2002;16: Welch HJ. Endovenous ablation of the great saphenous vein may avert phlebectomy for branch varicose veins. J Vasc Surg 2006;44: Vasquez MA, Wang J, Mahathanaruk M, Buczkowski G, Sprehe E, Dosluoglu HH. The utility of the Venous Clinical Severity Score in 682 limbs treated by radiofrequency saphenous vein ablation. J Vasc Surg 2007;45: ; discussion Passman MA, Dattilo JB, Guzman RJ, Naslund TC. Combined endovenous ablation and transilluminated powered phlebectomy: is less invasive better? Vasc Endovascular Surg 2007;41: Ravi R, Rodriguez-Lopez JA, Trayler EA, Barrett DA, Ramaiah V, Diethrich EB. Endovenous ablation of incompetent saphenous veins: a large single-center experience. J Endovasc Ther 2006;13: Merchant RF, DePalma RG, Kabnick LS. Endovascular obliteration of saphenous reflux: a multicenter study. J Vasc Surg 2002;35: Nicolaides AN. Investigation of chronic venous insufficiency: A consensus statement (France, March 5-9, 1997). Circulation 2000; 102:E Weiss RA, Weiss MA. Controlled radiofrequency endovenous occlusion using a unique radiofrequency catheter under duplex guidance to eliminate saphenous varicose vein reflux: a 2-year follow-up. Dermatol Surg 2002;28: Gradman WS. Adjunctive proximal vein ligation with endovenous obliteration of great saphenous vein reflux: does it have clinical value? Ann Vasc Surg 2007;21: Knipp BS, Blackburn SA, Bloom JR, Fellows E, Laforge W, Pfeifer JR, et al. Endovenous laser ablation: venous outcomes and thrombotic complications are independent of the presence of deep venous insufficiency. J Vasc Surg 2008;48: Mozes G, Kalra M, Carmo M, Swenson L, Gloviczki P. Extension of saphenous thrombus into the femoral vein: a potential complication of new endovenous ablation techniques. J Vasc Surg 2005;41: Submitted May 2, 2011; accepted Jun 27, 2011.