2025 Volume 18 Issue 1 Article ID: oa.25-00097
Objectives: In Japan, radiofrequency ablation (RFA) was covered by insurance in 2014. Regarding the early results, there are many reports that it is excellent, but the long-term results are rarely shown. Therefore, the purpose of this study was to examine the results up to 5 years after RFA, which was the only means of endovascular treatment at my facility, and to evaluate recurrent varices after surgery (REVAS), reoperation, and complications.
Methods: In 2017, 275 consecutive cases (male 83, female 192, 67.5 ± 10.0 years old), involving 350 limbs (C2, 3, 4a, 4b, 5, 6 = 217, 18, 89, 18, 1, 7) and 354 veins (great saphenous vein [GSV]: 290, small saphenous vein [SSV]: 64), underwent RFA. Postoperative follow-up was usually performed at 3 days (100%), 1 week (100%), 1 month (99.4%), and 6 months (93.5%) after RFA. Of the 334 patients contacted 5 years later, 327 (92.4%), excluding deaths from other diseases, were interviewed about reoperation, recurrence, and neuropathy. Of these, 223 patients (63%, GSV: 180, SSV: 43) underwent an ultrasound examination.
Results: Endovenous heat-induced thrombosis (EHIT) of more than grade 2 occurred in 5.9% of cases (GSV: 6.6%, SSV: 3.1%) and regressed within 1 month in all cases using anticoagulant therapy. Neuropathy occurred in 6.3%, mainly in GSV full-length ablation cases, and 41% of these disappeared completely between 6 months and 5 years after RFA. The reoperation rate up to 5 years after RFA was 10.7% (GSV: 9.7, SSV: 15.6%). In the reoperation cases after the RFA of GSV, the sites of reflux were 14 incompetent perforating veins (IPVs), 7 deep venous junction-related, and 6 distal GSVs. In the reoperation cases after the RFA of SSV, there were 5 IPVs and 3 isolated branch varices. The time of diagnosis of recurrence by ultrasonography was within 6 months of regular follow-up in 61% of GSVs and 79% of SSVs. The occlusion rate after 5 years was 98.9% for GSV and 95.3% for SSV. Only 1 recanalized vein was symptomatic and underwent retreatment. After RFA of GSV, 80% of accessory saphenous veins were retained at 5 years.
Conclusions: 1) The 5-year results after RFA were generally good. 2) IPV played the most important role in reoperation cases by REVAS after RFA of the saphenous veins. 3) The majority of postoperative recurrences could be identified by follow-up up to 6 months. 4) Neuropathy after RFA disappeared completely in half of the cases within 5 years. 5) After GSV-RFA, accessory saphenous vein blood flow persisted in most cases. (This is a translation of Jpn J Phlebol 2024; 35: 403–408.)
In Japan, endovenous laser ablation (EVLA) and radiofrequency ablation (RFA) of the saphenous vein were approved for medical insurance coverage in 2011 and 2014, respectively. Since then, the number of varicose vein surgeries has increased dramatically. Many favorable early outcomes of these endovascular ablation procedures have already been reported worldwide, including high occlusion rates, very few complications, and improved quality of life. However, there are only a few reports of long-term outcomes, such as the 5-year occlusion rate of the great saphenous vein (GSV).1–5) This study aimed to investigate the 5-year clinical outcomes of all RFA cases treated in 2017, when RFA was the only available endovascular treatment at my facility, and to report the results with a literature review.
The study included 275 consecutive patients (83 men, 192 women, age 67.5 ± 10.0 years) who underwent RFA in 2017. In total, 350 limbs (C2, 3, 4a, 4b, 5, 6: 217, 18, 89, 18, 1, 7) involving 354 veins (GSV: 290, small saphenous vein [SSV]: 64) were treated. This does not include 23 limbs with ablation less than 13 cm, 15 Dodd perforator-derived limbs, or 4 limbs with high ligation. In addition, 22 limbs underwent high ligation and stripping for anatomical reasons at the deep venous junction or for full-length superficialization during the same period. Ablation was performed using ClosureFAST (Covidien Ireland, Dublin, Ireland) in accordance with the guidelines of the Japanese Society of Phlebology at the time6) to avoid leaving any reflux sites. Immediately after surgery, 2000 units of heparin were administered intravenously to prevent thrombotic complications, except in patients receiving controlled anticoagulation therapy. The extent of ablation is shown in Fig. 1. The concomitant surgeries included tributary varicose vein resection in 333 limbs (95.1%), foam sclerotherapy for tributary varicose veins in 7 limbs (2.0%), incompetent perforating vein (IPV) resection in 38 limbs (10.9%), and IPV foam sclerotherapy in 2 limbs (0.6%). Postoperative observations were usually performed within 3 days (100%), at approximately 1 week (100%), approximately 1 month (99.4%), and approximately 6 months (93.5%). As early complications, endovenous heat-induced thrombosis (EHIT) and neuropathy were investigated. Anticoagulant therapy was administered to patients with EHIT class 2 or higher and was terminated when the EHIT class was reduced to 1. If residual recurrent lesions were present at 6 months, the patient was given a detailed explanation of the condition and reoperation was recommended if symptoms were present. Meanwhile, if symptoms were mild or absent, the patient was instructed to visit the hospital every 6 months thereafter or to return to the clinic if symptoms appeared, after receiving an explanation of possible symptoms that may occur if the residual lesions progressed. Of the 334 patients who were contacted 5 years later, 327 (92.4%) were interviewed by telephone regarding reoperation, recurrence, and neuropathy. Of these, 225 veins (64%; GSV: 182, SSV: 43), 221 limbs (C2, 3, 4a, 4b, 5, 6: 124, 7, 73, 11, 1, 5), and 171 patients (62 men, 109 women) gave informed consent to participate in the study, after which I investigated the presence or absence of symptoms, macroscopic findings, and the 5-year progress of patients who developed early neuropathy. I also performed a free lower limb venous ultrasound examination (59.6 ± 2.6 months after RFA) outside of insurance coverage (Fig. 2).
In routine postoperative observation, EHIT of class 2 or higher was observed in 5.9% of patients (GSV 6.6%, SSV 3.1%; Table 1), and all cases regressed within 1 month with anticoagulant therapy. No cases of deep vein thrombosis or symptomatic pulmonary thromboembolism were observed. Overall, neuropathy (Table 2) occurred in 22/354 limbs (6.2%), mainly in cases where the entire GSV was ablated (Table 2); however, after 5 years, it had completely disappeared in 9 limbs (41%), and the remaining cases no longer impacted daily life. No postoperative neuropathies were observed after SSV ablation.
| Total | 2 | 3 | 4 | |
|---|---|---|---|---|
| Total (n = 354) | 21 (5.9%) | 15 (4.2%) | 6 (1.7%) | 0 (0%) |
| GSV (n = 290) | 19 (6.6%) | 13 (4.5%) | 6 (2.1%) | 0 (0%) |
| SSV (n = 64) | 2 (3.1%) | 2 (3.1%) | 0 (0%) | 0 (0%) |
GSV: great saphenous vein; SSV: small saphenous vein
| GSV | SSV | ||
|---|---|---|---|
| Range | Nerve injury (%) | Range | Nerve injury (%) |
| LT | 0/64 (0%) | MC | 0/50 (0%) |
| K | 0/50 (0%) | LC-F | 0/14 (0%) |
| UC | 1/128 (0.8%) | Total | 0/64 (0%) |
| MC | 1/14 (7.1%) | ||
| LC-F | 20/34 (58.0%) | ||
| Total | 22/290 (7.6%) | ||
LT: lower thigh; K: around the knee; UC: upper calf; MC: middle calf; LC-F: lower calf or full length; GSV: great saphenous vein; SSV: small saphenous vein
The reoperation rate up to 5 years after surgery was 10.7% (GSV 28/290, 9.7%; SSV 10/64, 15.6%). The most common sites of reflux in reoperated cases (Table 3) were IPV, deep venous junction-related, and distally residual GSV in GSV cases, and IPV and isolated side branch aneurysms in SSV cases. The timing of reoperations by the recurrence site is shown in Fig. 3. IPV was treated scattered over a 5-year period after GSV ablation; deep venous junction-related and pelvic venous varicose veins were most common 5 years later, and residual leg GSV was often performed early after surgery. Although there were no defining characteristics regarding the timing of reoperation by recurrence site after SSV ablation, the time until reoperation tended to be shorter than that after GSV ablation. The time of reoperation was defined as when reoperation was performed after symptoms related to recurrence appeared, and it differed from the time of diagnosis.
| GSV | 28 veins 33 sites |
|---|---|
| Saphenofemoral junction-related | 7 |
| Recanalization (thigh) | 0 |
| Incompetent perforating vein (thigh) | 1 |
| Pelvic venous system-related | 2 |
| Simple tributary varix | 2 |
| SSV | 2 |
| Recanalization (calf) | 0 |
| Residual distal GSV of the calf | 6 |
| IPV (calf) | 13 |
| SSV | 10 veins 11 sites |
|---|---|
| Incompetent perforating vein (thigh) | 1 |
| Saphenopopliteal junction-related | 2 |
| Proximal recanalization | 1 |
| Simple tributary varix | 3 |
| Incompetent perforating vein (calf) | 4 |
GSV: great saphenous vein; IPV: incompetent perforating vein; RFA: radiofrequency ablation; SSV: small saphenous vein
Recurrent lesions diagnosed by ultrasound after 5 years, including asymptomatic ones, were found in 41/290 (14.1%) and 12/64 (18.8%) GSVs and SSVs, respectively. The time of diagnosis was within 6 months in 28/46 sites in GSVs (61%) and 11/14 sites in SSVs (79%).
The occlusion rate after 5 years was 98.9% (180/182) in GSVs and 95.3% (41/43) in SSVs, but all recanalizations were partial, and retreatment was performed in only 1 symptomatic SSV out of 4 recanalizations. The side branches of all 182 deep vein junctions of GSVs had residual blood flow, and the residual accessory saphenous veins were found in 54%, 68%, and 80% of the cases at 1 month, 6 months, and 5 years, respectively, and increased over time. Meanwhile, deep vein junctional side branch blood flow was only observed in 2 (4.7%) of the SSVs. The length of the deep venous junction stamp was 1.2 ± 0.4 cm in the GSV and 0.4 ± 0.7 cm in the SSV.
Many studies that showed clinical outcomes 5 years after RFA have shown occlusion rates after GSV treatment; however, with the exception of the report by Morrison et al.,1) these rates were statistically calculated, and only some of the cases shown were actually investigated after 5 or more years. Table 4 compares these data1–5) with the 5-year data for the cases investigated in this study. The occlusion rate in this study was favorable at 98.9%, and there was only 1 case of reoperation due to recanalization after SSV. The reason for this is unclear; however, 1 factor is that the procedures were performed in strict compliance with the Japanese guidelines at the time.6) Cases with borderline indications at the deep venous junction were excluded from this study; in such cases, high ligation was actively combined with RFA, or RFA was replaced with vein stripping. Active treatment of accessory lesions, such as side branch varicose veins and IPVs, is also thought to be related to this favorable occlusion rate. For older types of EVLA, the RELACS Study7) and other studies have shown 5-year results, but only 1 paper8) shows long-term clinical results for the 1470-nm ring type (radial fiber) currently used in most cases in Japan, and it states that the occlusion rate after 114 months was 160/161 (99.4%) for GSV and 42/42 (100%) for SSV.
| Author | Source of ref. | n | Occlusion rate |
|---|---|---|---|
| Morrison1) | J Vasc Surg Venous Lymphat Disord 2020 | 33 | 85.2% |
| Daylan2) | J Vasc Surg Venous Lymphat Disord 2021 | 634 | 93.1% |
| Lawaetz3) | Int Angiol 2017 | 138 | 94.2% |
| Proebstle4) | Br J Surg 2015 | 295 | 94.9% |
| Baccellieri5) | J Vasc Surg Venous Lymphat Disord 2023 | 1568 | 92.8% |
| Kusagawa | NA | 182 | 98.9% |
GSV: great saphenous vein; NA: not available; RFA: radiofrequency ablation
Regarding early complications, an EHIT of class 2 or higher was observed in 5.9% of the patients, but it disappeared after 1 month of follow-up with anticoagulant therapy, and no progression to symptomatic pulmonary embolism or deep vein thrombosis was observed. Neuropathy was observed in 58.0% of cases in which full-length GSV ablation was performed for patients with obvious reflux to the lower calf, but it completely disappeared in 41% of these cases after 5 years. In the remaining cases, the patients could not clearly recall whether it had improved; however, at the very least, no cases were observed where it interfered with daily life. This is in contrast to a previous survey of 5-year results of GSV stripping,9) in which saphenous nerve damage felt immediately after stripping remained in all cases after 5 years.
The initial diagnosis of cases with recurrent lesions was made within 6 months in 61% of GSV cases and 79% of SSV cases; therefore, the usual follow-up plan at my facility was considered appropriate. Regarding the relationship between the site of recurrence and the timing of reoperation, recurrences related to the deep venous junction after GSV were treated relatively late, whereas reoperation was performed relatively early for persistent peripheral GSV. However, the number of cases was small, and further investigation is required.
In a study of reoperation cases, the reoperation rate after 5 years was 9.7% for GSV and 15.6% for SSV, which was higher than the reoperation rate after 5 years of GSV stripping in my previous study (6.8%).9) Other comparable data from RFA at 5 years or more have shown that the reoperation rate with symptoms 5 years after GSV surgery is 10%.5) In a report on EVLA, the reoperation rates after 114 months with a 1470-nm radial fiber were 21% for GSV and 5% for SSV,8) but this report did not mention recurrence due to IPV.
Table 5 shows a comparison of the reflux sites in reoperation cases 5 years after RFA and after stripping for GSV in my previous study.9) The recurrence rate related to the deep venous junction was higher with RFA. Comparing the findings of the lower limb venous ultrasound at the deep venous junction, the residual blood flow of the accessory saphenous vein was 80% after RFA and 5.9% after stripping, a substantial difference, and the length of the deep venous junction stump was 1.2 and 0.1 cm after RFA and stripping, respectively. In both RFA and stripping, the most common recurrence was from the IPV. Recurrence of side branch varicose veins alone was more common with stripping than with RFA, but this is thought to be because intraoperative ultrasound was not used during varicose vein resection with stripping, which included cases where the residual lesions had progressed. Previously, the options for treating such recurrent lesions were surgery or sclerotherapy; however, since 2018, the use of laser slim fiber has become available, and treatment options have increased in Japan, contributing to improved treatment outcomes.10)
| RFA 290 veins | Stripping 413 veins9) | |
|---|---|---|
| Saphenofemoral junction-related | 7 (2.4%) | 3 (0.7%) |
| Recanalization (thigh) | 0 (0%) | NA |
| IPV (thigh) | 1 (0.3%) | 0 (0%) |
| Pelvic venous system-related | 2 (0.7% | 0 (0%) |
| Simple tributary varix | 2 (0.7% | 8 (1.9%) |
| SSV | 2 (0.7%) | 4 (1.0%) |
| Recanalization (calf) | 0 (0%) | NA |
| Residual distal GSV of the calf | 6 (2.1%) | 4 (1.0%) |
| IPV (calf) | 13 (4.5%) | 14 (3.4%) |
| Total | 33 sites 28 limbs | 33 sites 28 limbs |
GSV: great saphenous vein; IPV: incompetent perforating vein; NA: not available; RFA: radiofrequency ablation; SSV: small saphenous vein
Regarding the analysis of recurrent varicose veins, Perrin et al. proposed the concept of recurrent varices after surgery (REVAS) at a time when surgical treatment was the norm. REVAS refers to varicose veins that exist in the lower limbs that have previously undergone varicose vein surgery. These were broadly divided into 2 categories: those in the same location as the initial surgery and those in a distant location. Those in the same location were further divided into technical failure, tactical failure, and neovascularization, and those in a distant location were divided into persistent and new.11) In addition, with the emergence of endovascular treatment, the concept of recurrent varices after thermal ablation was introduced, making the situation more complicated.12) However, endovascular treatment only adds recanalization, and recurrence after treatment for varicose veins in the lower limbs can be easily understood as follows. They are divided into 3 categories: 1) residual, 2) new, and 3) recanalization with endovascular treatment. Category 1 includes the so-called technical and tactical failures. Category 2 includes neovascularization and other new lesions that develop in any location. Regarding residual stump side branch recurrences related to the deep venous junction, these occur with incomplete high ligation even after stripping; hence, there is no need to distinguish them from those after endovenous ablation (EVA). In this case group, it was difficult to strictly classify the recurrences into Categories 1 and 2, but most of the recurrences were new or very mild lesions that progressed after surgery, except for some isolated side branch aneurysms that may have been residual. A recent study has noted that the significant risk factors for recurrence 5 years after RFA were not performing side branch varices resection, performing side branch varices sclerotherapy, not performing treatment other than RFA for the saphenous vein, and performing sclerotherapy for the IPV.5)
Among these, recanalization is a type of recurrence specific to endovascular treatment. To avoid recanalization, the cause should be considered in addition to analyzing the common risk factors. The mechanism of venous occlusion in endovascular treatment must be understood, but this is not clearly described in textbooks. Unlike normal parts, the veins in the affected area are not uniform and straight, and organic occlusion is not complete over the entire length of the treated area immediately after endovascular treatment. In reality, there is a mixture of organic occlusion and thrombotic occlusion, and in the case of cyanoacrylate closure, this also includes embolic occlusion. It is thought that thrombotic occlusion is involved in the mechanism by which occlusion occurs immediately but then recanalizes after a while. If this area were a complete dead end, it would become organized as it is. However, if side branches, perforators, or the deep vein junction were connected here, an inlet and outlet for blood flow would be created and the blood flow would enter the thrombus, gradually dissolving the thrombus and recanalizing it. Considering this thrombolytic mechanism, the blood flow in the accessory saphenous vein, which appeared to be occluded immediately after surgery, resumes over time, as observed in this study. Even if the starting position during treatment was set to 0 cm from the junction, the part that appeared to be occluded with thrombus in EHIT class 1 immediately after surgery, with no stump remaining, would dissolve due to the blood flow entering from the accessory saphenous vein. Although there were only a few cases, the timing of recanalization observed in this study was similar to the timing of recanalization in deep vein thrombosis, which ranges from 1 to 6 months, and in this respect, the usual follow-up observation plan at the facility was considered appropriate.
Although this study is a retrospective cohort study conducted at a single institution and has limitations, I hope that it will be useful as a reference for appropriate follow-up after RFA and EVA.
At my facility, good clinical outcomes were reported for up to 5 years after saphenous vein RFA, and the IPV was the most important site of recurrence requiring reoperation. Most recurrences were diagnosed by lower extremity venous ultrasound within 6 months after surgery. Postoperative neuropathy was completely asymptomatic in 41% of cases within 5 years. After RFA of the GSV, antegrade blood flow remained in the accessory saphenous vein in most cases.
There are no conflicts of interest to disclose in this study.
The personal information of the patients involved in this study was kept anonymous, and consent was obtained from each patient. The Ohta Clinic Ethics Committee approved the series of tests and surgeries (approval number OEC-K001).
This study was presented at the 2023 UIP World Congress in Miami, USA, in September 2023.
