Mechanochemical ablation for the treatment of superficial venous incompetence: A cohort study of a single centre’s early experience

Introduction

Lower limb superficial venous incompetence (SVI) is a common condition, affecting more than a third of the adult population.^1,2 The physical symptoms of uncomplicated varicose veins impact significantly upon health-related quality of life (HRQoL). ³ Disease progression can lead to irreversible skin damage in 10% of the population and venous ulcers in 1–2% of adults throughout their lifetime;^2,4–6 those patients progressing to ulceration have a profound deterioration in HRQoL. National and international guidelines based on safety, clinical and cost-effectiveness recommend endovenous thermal ablation (EVTA) as the first-line treatment modality for patients with SVI.^6–8 More recently, non-thermal non-tumescent methods such as mechanochemical ablation (MOCA) have been introduced as an alternative modality.^9,10 This treatment does not require the infiltration of tumescent local anaesthesia (TLA), and hence it is purported to reduce procedural pain. ¹¹

MOCA using the ClariVein® device (Vascular Insights, LLC, Madison, Connecticut) combines the intravenous action of a motor-powered rotating wire with simultaneous injection of liquid sclerosant. MOCA produces transmural mechanical and chemical injury to the vein wall leading to cell death and occlusion of the incompetent vein. ¹² Studies have shown this technique to be safe and effective in the management of SVI; however, there is discrepancy regarding the occlusion rates achieved using MOCA borne from inconsistent definitions in the literature.^13,14 MOCA was introduced as a new treatment for SVI in October 2014 in this tertiary academic and teaching institution in the UK serving a population of 1.2 million. This study aims to compare MOCA occlusion rates against consensus definitions of occlusion and correlate them with clinical outcomes and HRQoL.

Methods

All patients presenting to the research team with symptomatic SVI were assessed for inclusion in this prospective appraisal of MOCA. Patients were assessed and counselled for potential study participation by a consultant vascular surgeon or clinical research fellow with a special interest in venous interventions. The inclusion criteria included consenting adult patients with symptomatic unilateral primary single axis reflux suitable for endovenous treatment; no upper limit was set for vein diameter. Patients were excluded if they had recurrent disease, active thrombophlebitis, deep venous thrombosis, isolated deep venous incompetence, bilateral or multiple axis reflux, or were pregnant. This study was approved by the local trust board and established in order to assess and monitor long-term outcomes of this new treatment in keeping with the National Institute for Health and Care Excellence recommendations. ¹⁰

Interventions were carried out under ultrasound guidance using the ClariVein® device with 1.5% sodium tetradecyl sulphate (STD Pharmaceutical Products Ltd, Hereford, UK) by a vascular consultant or clinical research fellow with a special interest in the management of venous disease and formal qualification in vascular ultrasound. The technique of MOCA used in this centre has been previously published. ¹⁵ In brief, in the upright position the lowest site of reflux was marked. The patient was then positioned in the reverse Trendelenburg position where 1 ml of 1% lidocaine was used to anaesthetise the cannulation site under sterile conditions. Vein micropuncture and insertion of a 5Fr sheath was performed under ultrasound guidance followed by insertion of the MOCA catheter through the sheath. The patient was then positioned horizontally and the tip of the MOCA wire placed 2 cm from the saphenofemoral junction (SFJ). The wire was then activated for 10 s to elicit vein wall spasm followed by catheter withdrawal at a rate equivalent to 1.4 mm/s (achieved in practice by constant withdrawal of the catheter by 1 cm every 7 s) with simultaneous infusion of sclerosant at a rate of 0.2 ml/cm. ¹⁵ Concomitant ambulatory phlebectomy was carried out on patient request under TLA (100 ml of 1% lidocaine + 1:200,000 adrenaline and 10 ml of 8.4% NaHCO₃ added to 900 ml of NaCl 0.9%). ¹⁶ After treatment, patients were discharged with Clinistretch® (Haddenham Healthcare Ltd, Bucks, UK) bandaging for one week. Neither analgesia nor anticoagulation was routinely prescribed; patients were risk assessed for venous thromboembolism (VTE) using a standard pro forma widely utilised in UK NHS practice and any that were deemed to be at high risk received a single pre-procedural prophylactic dose of subcutaneous low-molecular weight heparin (LMWH). ¹⁷

Patients were assessed by members of the research team at baseline and weeks 1, 6, 26 and 52. The primary outcome was anatomical occlusion rate on duplex ultrasound (DUS). This was assessed at each visit according to the American Venous Forum consensus criteria; ¹⁸ a vein was successfully ablated if it was either absent or incompressible with no detectable flow in the entire treated length. Secondary outcomes included Venous Clinical Severity Score (VCSS), ¹⁹ HRQoL, complications, patient satisfaction, procedural pain and need for reintervention.

Periprocedural pain was measured immediately post-surgery on a 100 mm Visual Analogue Scale (VAS) from 0 (no pain) to 100 (worse pain possible). Before attending each follow-up clinic patients were asked to complete the EuroQol-5D (EQ-5D) ²⁰ and Aberdeen Varicose Vein Questionnaire (AVVQ) ²¹ to measure HRQoL. Satisfaction regarding clinical and cosmetic outcomes was measured on 100 mm VAS from completely dissatisfied to completely satisfied. Reintervention was offered after six months if the patient had symptomatic incompetent recanalisation of the treated vein, new axial reflux in the treated leg or if there were symptomatic residual or new varicosities in the treated leg. Collected data were uploaded onto a secure hospital database and analysed using IBM SPSS version 24 (IBM Corp, Armonk, New York). Normally distributed data are presented as mean (±standard deviation) and non-normally distributed data are presented as median (interquartile range). Comparative analysis was conducted using Friedman’s test with significance level set at p < 0.050.

Results

Consecutive patients meeting eligibility criteria were approached regarding participation in the study. Over a nine-month recruitment period (October 2014–June 2015) 101 patients were screened, of which 94 patients were eligible and 32 consented to participate. The most common reason for non-participation was a patient preference for endothermal ablation (62 patients). In the same period at our institution, 1178 patients underwent thermal ablation, 94 patients underwent open surgery and nine had foam sclerotherapy. Table 1 summarises the included patients’ baseline characteristics. Figure 1 shows a flow chart of the patients involved at each stage of the study.

OPEN IN VIEWER

Figure 1.

Study flow diagram.

OPEN IN VIEWER

Table 1.

Baseline characteristics.

Number of patients	32
Male:Female	14:18
Mean age (SD)	50.75 (±14.6)
Mean BMI	27.10 (±4.9)
CEAP 2:3:4:5:6	10:13:6:1:2
Median VCSS (IQR)	6 (5–8)
Median AVVQ (IQR)	13.50 (10.00–18.65)
Median EQ-5D (IQR)	0.877 (0.807–0.877)
Mean saphenofemoral junction vein diameter (SD)	7.97 mm (±2.83 mm)

AVVQ: Aberdeen Varicose Vein Questionnaire; BMI: body mass index; CEAP: Clinical aEtiology Anatomy Pathophysiology Classification; EQ-5D: EuroQol-5D; IQR: interquartile range; VCSS: Venous Clinical Severity Score.

Data are presented as mean with standard deviation (SD) if normally distributed or median with interquartile range (IQR) when not normally distributed.

All 32 patients had SFJ incompetence; in 31 patients the great saphenous vein (GSV) was the incompetent axis, while one patient had an incompetent anterior accessory saphenous vein (AASV). All 32 patients received the planned MOCA and completed one-week follow-up. The mean total volume of sclerosant used was 11 (±3) ml per patient, infused at a mean rate of 0.26 (±0.08) ml/cm. Eight patients opted to have concomitant ambulatory phlebectomy. The procedures were well tolerated with a median (IQR) periprocedural pain score of 20 (7–47) on VAS.

Technically all procedures were initially successful with a 100% occlusion rate at one week (32/32), six weeks (30/30) and six months (29/29). However, by one year anatomical occlusion on DUS was reduced to 75% (21/28). This was due to recanalisation of the treated GSV in seven patients. One (3.6%) patient had complete recanalisation whereas the other six (21.4%) had partial recanalisation. This occurred in the proximal part of the treated GSV segment (in continuity with the SFJ and deep system) in four patients and in a distal segment in two (discontinuous with the SFJ). All recanalised segments showed reflux of greater than 0.5 s duration.

VCSS scores significantly improved at one week, six weeks, six months and one year compared to baseline (Figure 2). Disease-specific HRQoL as measured by the AVVQ demonstrated significant improvement at one week, six weeks, six months and one year compared to baseline (Figure 3). Generic HRQoL as measured by the EQ-5D demonstrated significant improvements at six weeks, six months and one year when compared to baseline (Figure 4). This corresponded with high patient satisfaction, with a median score of 100 at one year on VAS for both clinical and cosmetic outcomes. Notably, patients with recanalisation reported statistically similar HRQoL scores to those with complete anatomical occlusion during follow-up; however, this study was not designed to detect such a difference.

OPEN IN VIEWER

Figure 2.

VCSS change/time p < 0.001. VCSS: Venous Clinical Severity Score.

OPEN IN VIEWER

Figure 3.

AVVQ change/time p < 0.001. AVVQ: Aberdeen Varicose Vein Questionnaire.

OPEN IN VIEWER

Figure 4.

EQ-5D change/time p = 0.001. EQ-5D: EuroQol-5D.

None of the patients in the study were deemed to be at increased risk of VTE therefore, none received prophylactic LMWH. One patient (3.1%), a 46-year-old lady, suffered a major adverse event in the form of pulmonary embolism (PE), presenting five days following GSV MOCA with chest pain and acute shortness of breath. Computed tomography pulmonary angiogram confirmed PE but no lower limb DVT was identified on two separate venous duplex scans. She was treated with six months of Apixaban and made a full recovery. Minor complications included thrombophlebitis lasting up to six weeks in (10/30) 33.3% of patients and skin staining lasting up to one year in (2/32) 6.3% of patients. There were no cases of nerve injury or infection.

Ipsilateral secondary procedures were carried out in six patients (21.4%), all taking place after six-month follow-up. Three patients underwent ambulatory phlebectomy and perforator ligation for residual symptomatic superficial varicosities. Endovenous laser ablation was performed for new reflux in a previously competent small saphenous vein in one patient and in the AASV in another. Lastly, one patient underwent laser ablation for symptomatic recanalisation and reflux of the treated GSV in the proximal segment.

Discussion

When applying the consensus definition of successful anatomical occlusion, the rate achieved at one year following MOCA in this study is 75%. This is significantly lower than has previously been reported in the literature, which ranges between 93 and 96%.^9,22,23 This is the first study to apply the consensus definition of anatomical occlusion when using MOCA. This is defined as ‘successful ablation of the target vein, as demonstrated by complete lack of flow or disappearance of vein by duplex ultrasound imaging in the entire treated segment’. ¹⁸ Studies in the literature have not stated how this outcome was measured^24,25 or defined it as ‘proximal occlusion’, ¹¹ or allowed for a ‘small amount’ of non-occluded vein in the duplex assessment of treated veins.^22,23 The application of the more stringent consensus criteria in this study is likely to have highlighted recanalised segments that other criteria would not have identified. These heterogeneous definitions of successful occlusion are a significant limitation to meaningful comparison not only between MOCA studies, but also when attempting to compare with other endovenous methods. Two systematic reviews of the literature have highlighted these differences and called for adherence to consensus criteria.^12,13

Following initial technical success, all recanalisations occurred after the six-month follow-up appointment. Evidence suggests that increased disease severity is associated with an increase in the risk of recanalisation after EVTA ²⁶ and this is postulated to be related to chronic venous insufficiency resulting in vein wall remodelling and thickening. ²⁷ Disease severity possibly also affects the recanalisation rate following MOCA as patients with more advanced disease recanalised more frequently in this study. Of the nine patients who had soft tissue damage at baseline (CEAP ≥4), five patients had recanalisation following treatment.

Transmural mechanical injury to the vein wall potentiates the penetration of sclerosant and MOCA relies on this synergistic effect between both the mechanical and chemical elements of the device. Perhaps a thicker remodelled vein wall is more resistant to mechanical injury and consequently the chemical sclerosant is less effective. Of note, larger vein diameter size did not appear to increase the risk of recanalisation in this study. Six patients had veins with a proximal diameter >10 mm, the largest of which measured 16.7 mm. All except for one remained occluded at one year, including the largest vein. A previous study also showed that a larger diameter does not increase the risk of recanalisation following MOCA. ²³

Despite the lower anatomical success in this study, the clinical results are similar to other studies with significant improvements in patient symptoms as measured by AVVQ and reduction of disease severity measured by VCSS^11,23 in the short term. Amongst patients with recanalised segments, one reported worsening symptoms and underwent thermal ablation one year after MOCA; all the others declined further treatment as their symptoms improved sufficiently. This mirrors EVTA where recanalisation did not significantly affect HRQoL gains at one year; ²⁶ however, there is little data on MOCA beyond one year. Only one cohort study has published MOCA results beyond two years, ²³ which demonstrated a deterioration in VCSS and HRQoL scores between one and three years. ²⁸ Perhaps recanalisation impacts the durability of HRQoL gains which might explain this finding.

The rate of thrombophlebitis in the current study is much higher than the previously reported rate of 10–14% in other studies of MOCA.^22,24,29 The incidence rate of this complication seems lower in those who opted for concomitant ambulatory phlebectomy, though the small sample size precludes further statistical analysis here. PE is a rare but serious complication of lower limb venous intervention. A 2010 meta-analysis reported a rate of 0.01–0.06% for PE following EVTA. ³⁰ Robust meta-analysis data for VTE rates following foam sclerotherapy are unavailable. ³¹ A Cochrane review of EVTA, surgery and foam sclerotherapy in 2014 did not give a direct figure for PE following treatment of varicose veins. ³² With regards to MOCA, a recent systematic review found two cases of PE out of 1294 patients. ¹³ The incidence rate suggested by these figures is 10 times larger than EVTA. However, MOCA is still a new treatment and to date around 70,000 procedures have been performed, ¹³ therefore it is possible that publication bias is affecting current figures. Further, the rarity of this complication precludes an accurate judgement on its incidence rate using current literature.

The findings in this study are limited by the small sample size, lack of comparator and potential for selection bias due to the non-random selection of the cohort. These limitations are unlikely to have affected the primary outcome of DUS-determined anatomical occlusion.

This study shows that MOCA offers an alternative option for patients in the treatment of SVI in terms of improved HRQoL and disease severity up to one year, despite less favourable anatomical occlusion rates when consensus definitions are applied. The impact of this recanalisation on medium- to long-term clinical and HRQoL outcomes is unknown. Future studies into MOCA should adhere to consensus definitions of outcomes in order to allow for meaningful comparison between studies and across treatment modalities. The incidence of PE and high rate of thrombophlebitis are concerning and merit specific investigation in future studies. MOCA offers some theoretical advantages over EVTA, but it remains to be seen which patients will benefit more from MOCA than the current gold standard of EVTA. Randomised clinical trials comparing these treatment modalities, along with more in-depth regression analysis of parameters favouring good technical outcome in MOCA will offer more information in this regard. In the meantime, it is incumbent upon units performing MOCA to present their early data so that meaningful comparisons can be made, and clinicians have more information with which to counsel and offer patients evidence-based tailor-made treatments.