Treatment of incompetent perforating veins using the radiofrequency ablation stylet: a pilot study

Introduction

By perforating the fascia generalis, about 150 perforating veins (PV) in the lower extremity connect the superficial with the deep venous system (reviewed by Van Neer et al.). ¹ PV larger than 1 mm diameter have valves and appear to function as pressure valves when high pressures occur in the muscular compartment and avoid an outward flow from the deep to superficial system. Incompetent PV (IPV) show reflux on ultrasound (US) examination of more than 0.5 second and often have a larger diameter (Figure 1). Because of its close association with deep or superficial venous incompetence, it is difficult to assess the contribution of isolated IPV in the development of chronic venous insufficiency (CVI).

OPEN IN VIEWER

Figure 1

Incompetent perforator vein on US

Because of the controversial role of IPV in CVI, the need to treat IPV remains somewhat unclear. ^1,2 Some authors suggest that PV are part of a compensatory mechanism in venous return and have shown that selective ligation of IPV did not improve venous haemodynamics. The increased venous limb volume diameter is correlated with the largest IPV diameter and may be responsible for and precede IPV development. ³ In clinical practice, IPV have often been treated in conjunction with the treatment of the superficial venous system. ^4,5 The most traditional therapy of IPV is surgical subfascial ligation or subfascial endoscopic perforator surgery (SEPS). ^1,6–8 Surgical ligation and, even more, Linton's procedure and its modifications leave a noticeable scar and have a high complication rate such as wound infection, nerve injury and postoperative pain, especially in patients with CVI-induced skin changes (≥C4 level of clinical, aetiological, anatomical and pathological elements [CEAP] classification). ⁹ Because the incision made in SEPS is remote from the affected skin, wound infections are less frequent; however, this technique has a slow learning curve and paratibial IPV and those of the upper leg are not eligible for this technique, which may explain why SEPS is not commonly accepted after its introduction. The percutaneous ablation of perforators (PAP) comprises techniques that are minimally invasive (i.e. no skin incision, local aesthesia and performed in an outpatient setting), are US-guided and can be easily repeated if necessary. ¹⁰ For more than 30 years, liquid sclerosant and more recently foam sclerosant have been used in the treatment of IPV with or without US guidance. US-guided sclerocompression therapy (UGST) of IPV (with foam) is a straightforward, swift and inexpensive procedure and open clinical studies suggest an efficacy of about 75–90%. ^11,12 The disadvantages of UGST of IPV are the lack of standardization (e.g. percentage and volume of sclerosant and type and duration of compression after treatment) and uncontrollable distribution of the injected sclerosant. Therefore, in addition to the IPV, it is likely that the superficial venous system may occlude. Another complication of UGST may be the inadvertent injection of foam in a satellite artery with subsequent necrosis. Two more controlled PAP options are endovenous laser ablation (EVLA) and radiofrequency ablation (RFA). EVLA has been reported to be successful in the treatment of IPV in several small case series. ¹³ To our knowledge, laser fibres and disposables used to treat saphenous varicose veins were used and no proprietary EVLA catheter or system to treat IPV is available yet. In a venous forum abstract, Whiteley et al. ¹⁴ have performed many IPV ablations by using the saphenous RF probe and reported a 93% occlusion rate after one year. Subsequently, VNUS Closure^® developed the radiofrequency stylet (RFS), which is a rigid RF catheter designed to treat IPV. ¹⁵ Although the Food and Drug Administration approved this device in 2004, relatively few studies have reported on its use, except for abstracts of the Society for Clinical Vascular Surgery (SCVS) and unpublished case series. ^16,17 One study showed 100% procedural success, a decreasing occlusion rate in time (91% were reflux free but 56% were patent after 1 year) and no complications in less than 40 IPV treated with RFS. Another open VNUS sponsored study demonstrated that 31/34 of the IPV that were treated with ‘intravascular’ RFS were occluded after three weeks. ‘Extravascular’ treatment of 63 IPV yielded ‘a significantly lower occlusion rate’. Two tibial vein thromboses occurred.

In this pilot study, the RFS procedure is described and its short-term efficacy and safety in 14 IPV after three months are reported.

Methods

Patients

In this pilot study, patients with one or more symptomatic (e.g. painful) IPV (reflux on US >0.5 second) with a diameter of at least 4 mm, with C2–4 according to the CEAP classification, were included (Table 1, Figure 2). Of the leg to be treated, the great saphenous vein (GSV) and small saphenous vein (SSV) had to be competent or successfully treated prior to the RFS procedure.

OPEN IN VIEWER

Figure 2

Diameter of incompetent perforator vein on US

OPEN IN VIEWER

Table 1

Demographic and clinical characteristics of the study population

Identification		Demographics		Phlebological status	Incompetent perforating vein (IPV)		Outcomes after three months
Patient no.	Vein no.	Gender	Age (years)	‘CEAP’ classification	Localization: distance (cm) and site	Diameter (cm)	Anatomical success†	Symptom reduction	Self-reported side-effects
1	1*	Female	67	C2	33 Right	0.5	+	+	Paresthesia
2	2*	Female	28	C3	37 Right	0.4	+	+	−
3	3*	Female	57	C4	23 Right	0.4	−	−	Paresthesia
4	4*	Female	66	C3	13 Left	0.4	+	−	−
5	5*	Female	53	C5	67 Left	0.6	+	+	−
6	6	Male	74	C2	22 Left	0.5	−	+	−
7	7	Male	77	C3	21 Right	0.4	−	−	−
8	8	Male	83	C3	37 Left	0.5	−	+	−
	9			C3	24 Left	0.7	+	+	−
9	10	Female	67	C2	20 Left	0.4	+	−	Localized pain
10	11	Female	35	C2	18 Right	0.4	+	+	−
11	12	Female	49	C4	18 Left	0.5	+	+	−
12	13	Male	75	C4	39 Right	0.4	−	−	−
	14			C4	32 Left	0.4	+	+	−

*4 cc anaesthesia and turning 90° every minute and the remaining IPV (vein no. 6–14) were treated using eight or more cc anaesthesia, ultrasound-guided and without turning every minute

^†Obliteration and absence of flow on ultrasound examination

Procedure

The procedures were performed in an outpatient setting in an academic, tertiary hospital by dermatologists specialized in (minimal invasive) phlebological procedures. The rigid RFS catheter, which has two electrodes on the distal part of the shaft (with a thermocouple) and a removable needle trocar, was used (Figure 3). Before inserting the probe, the functionality of the RFA electrodes was tested by measuring an impedance of about 100 ohm of an activated catheter in physiological water. US guidance (Phillips EnVisor HD) was used to locate and select a non-tortuous part of the IPV, which are often curvilinear and angulated, in standing position. After local lidocaine anaesthesia, with the patient now in the horizontal position, a small skin incision of about 2 mm was made to directly puncture the IPV and the tip was positioned endovenously at or just below the level of the fascia generalis (>0.5 cm from the deep venous system) under US guidance in longitudinal view (Figure 4). At 4 W power, the thermocouple on the electrodes was at about 37°C and the level of impedance was between 250 and 400 ohm. Using a syringe, (tumescent) anaesthesia was administered. In the first four sessions, we followed VNUS Closure's protocol and applied 4 cc of lidocaine solution blind around the tip of the catheter, which caused a temperature drop to about 30°C as the solution was administered around the tip. The activated RFS was rotated to four different positions, each for one minute of treatment (total treatment time of 4 minutes at 85°C), while local pressure was administered with the US probe. In subsequent sessions, this protocol was changed. Instead, after administering 8–12 cc of local anaesthesia under US guidance, the RFS remained fixed for four minutes in the same position under local pressure (Figure 5). In some cases, it was possible to repeat the procedure in the same IPV after withdrawing the catheter 1 cm. The rationale of these two deviations from the original protocol was the observation that two-fourths of patients treated in the first session presented with localized paresthesia just below the RFS-treated PV. The cause of this RFS-induced nerve injury might have been insufficient volume of tumescent anaesthesia, which should also protect the perivenous tissue by cooling it and the frequent manipulation of the catheter during the procedure. In the first sessions, we noted that the impedance relatively often increased to values above 500 (somewhere in the range of 2000) suggesting that the tip of the catheter was positioned extravascularly. When the latter occurred, the RF was inactivated and the tip was repositioned under guidance of the US and the impedance level. However, this may have played a role in the development of the observed neurological damage. After the procedure, local manual pressure was applied at the treated site for about one minute and medical elastic compression stockings were not advised.

OPEN IN VIEWER

Figure 3

Radiofrequency ablation stylet

OPEN IN VIEWER

Figure 4

Visualization of the needle inside the incompetent perforator vein

OPEN IN VIEWER

Figure 5

US visualization during treatment

Results

The demographic and clinical characteristics of the study patients are presented in Table 1. Four patients were classified as C2 of the CEAP classification, four as C3, three as C4 and one as C5. Of the included patients, six had a history of GSV therapy and four of SSV therapy prior to the RFS procedure. Of the included IPV, most connected the posterior accessory GSV with the posterior tibial veins (Cockett's PV) but paratibial PV and a Hunter's PV of the thigh were also included. All included IPV were treated using RFS (i.e. 100% procedural success, defined by a technically correct procedure with the perforator not showing flow directly after the procedure). During the procedure, patients did not report any discomfort. After three months of follow-up, nine of 14 treated IPV were occluded (64%) and the other five showed reflux on US examination. Three months after therapy, the effect of RFS on patients' symptoms varied considerably (Table 1). No deep vein thrombosis (DVT) was noted. Two patients reported localized paresthesia in the lower leg; this paresthesia entailed loss of sensibility in a 1–4 cm² area around the place of needle puncture. Six months after the procedure, these two patients reported a continuation of their paresthesia by telephone.

Discussion

RFA and EVLA are promising new techniques for the controlled ablation of IPV, but have not yet been well standardized, documented and studied. In this pilot study, a success rate of 64% of RFS in the treatment of IPV is low compared with previous open studies. ^14,16,17 Because of the relative poor reporting of the available case series, which were often abstracts, it is difficult to explain the observed difference. ^14,16,17 However, better results may be expected, due to a learning curve, as these were the first patients we treated with the RFS. In most patients, we were unable to treat IPV at two locations 1 cm apart because of the tortuous tract of the included IPV, which may explain some of the failures, and suggest that an optimal patient and IPV selection increases the success rate.

Although some surgical complications can be avoided by PAP, these techniques may be associated with specific complications such as DVT and cutaneous paresthesia, and have a relatively long learning curve. ^10,16,17 Although dermatologists who were experienced in obtaining percutaneous access to varicose veins including IPV performed the RFA procedures in this pilot, it was challenging to position the stylet because of the relatively small size and the angulated tract of the IPV and the large size of the catheter (6 French). From our experience, it seems important to pass the RFS at least a few centimetres in the IPV, to avoid the fixed electrodes being ‘pushed’ extravascularly by the contracting IPV after generating the heat. Positioning the RFS tip extravascularly and/or above the fascia decreases the likelihood of success and may increase the risk of nerve injury. ¹⁷ In 2/14 cases paresthesia was reported, which is in accordance with the available literature that suggests that between 9% and 19% of patients develop (transient) paresthesia. ¹⁶ In accordance with our observation, the likelihood of this complication is higher in the treatment of IPV below the knee because the saphenous nerve travels superficial of the saphenous compartment and is thus more susceptible to heat-induced injury. Also, along with PV run a small artery and a nerve. Ample and well-positioned tumescent anaesthesia, a long enough intravascular part of the RFS and a minimum of RFS manipulation may decrease the risk of RFS-induced paresthesia.

Indications to treat IPV are that they are symptomatic, contribute to CVI and/or induce skin changes in their close proximity. Although the role of IPV in CVI is somewhat controversial, studies showed that about two-thirds of recurrent varicose veins and limbs with skin changes have IPV in conjunction with superficial or deep venous insufficiency. In addition, several studies showed benefit of interruption of IPV. ^6,7,18 Unfortunately, there are no randomized clinical trials that assess the effect of IPV therapy compared with conservative therapy or compare different (minimally invasive) treatment options. Compared with surgery, the PAP procedures are especially indicated in patients with IPV with overlying skin changes due to CVI such as lipodermatosclerosis and leg ulcers. It could be argued that EVLA and RFA of the IPV are more likely to spare the saphenous veins because of their controlled approach compared with UGST. But because UGST is swift, inexpensive and has a low complication risk, large randomized trials are warranted to compare these three percutaneous therapies of IPV.