The frequency and clinical significance of nontarget superficial and deep vein occlusion after physician compounded foam sclerotherapy of varicose tributaries

Abstract

Objective

To evaluate the incidence and clinical relevance of silent nontarget occlusion (NTO) of superficial and deep veins occurring after ultrasound-guided foam sclerotherapy (UGFS) that can be detected by serial duplex ultrasound scan (DUS).

Methods

This retrospective analysis evaluated the medical records of patients treated with UGFS at a private clinic in Moscow, Russia from 2015 to 2017. All patients underwent serial DUS at 1 to 2 weeks and 1, 3, 6, and 12 months after UGFS.

Results

During the observation period, 268 patients were treated with UGFS, using physician compounded polidocanol foam 257 lower limbs of 196 patients (73%) with varicose veins who underwent DUS at 1 to 2 weeks after the last session of UGFS (inclusion time-point): 139 women and 57 men (mean age: 44.2 ± 12.2 years) with the following CEAP clinical class distribution: C2, 74.0%; C3, 20.0%; C4, 4.5%; and C5, 1.5%. NTO at the inclusion time-point occurred in 60 limbs (23.3%) of 57 patients (28.5%) and was symptomatic only in three limbs (1.2%). Most occlusions were localized in the untreated great saphenous vein trunk (n = 28) and the calf muscle veins (n = 23). Overall, 72%, 44%, 29%, and 10% of all limbs and 77%, 45%, 28%, and 12% of limbs with NTO were followed up by DUS at 1, 3, 6, and 12 months, respectively. There were no cases of thrombus progression or symptomatic venous thromboembolism (VTE). At six months, no deep vein occlusions persisted.

Conclusions

The frequency of nontarget vein occlusion after UGFS revealed by serial DUS may be as high as 23.3%. These occlusions tend to resolve within six months and are not associated with symptomatic VTE.

Keywords

Chronic venous disease deep vein thrombosis sclerotherapy superficial vein thrombosis varicose veins

Introduction

According to recent epidemiology data, the worldwide prevalence of primary varicose veins (VVs, clinical class C2 by the Clinical-Etiologic-Anatomic-Pathophysiologic [CEAP] classification of chronic venous disease) is 18%, and clinical classes C2 to C6 account for 42%.¹ Official Russian Federation statistics indicate that the incidence of VVs is 660 to 857 cases per 100,000 population per year^2–4 with a prevalence of 26%.⁵ Ultrasound-guided foam sclerotherapy (UGFS) is considered a safe, effective, and inexpensive method for VV ablation.^6–8 The reported incidence of serious complications is very low, and the rate of symptomatic deep vein thrombosis (DVT) and pulmonary embolism (PE) appears to be <1%.⁶ However, asymptomatic (silent) DVT is found more often (up to 3.2%) using routine duplex ultrasound scan (DUS),^9–12 but its clinical relevance is not well established. Furthermore, fewer data are available on the prevalence of superficial vein thrombosis (SVT), silent occlusion of preserved great saphenous (GSV), or small saphenous (SSV) veins after UGFS. Because symptomatic PE is a rare complication of UGFS,^9,13,14 it is reasonable to assume that silent DVT and SVT are not associated with a potential long-term threat. However, there are no qualitative prospective data on this issue.

This study evaluated the incidence and clinical relevance of silent nontarget occlusion (NTO) of superficial or deep veins that occur after UGFS with physician compounded polidocanol foam that can be detected by serial DUS.

Methods

This study is a retrospective analysis of prospectively collected data obtained from electronic medical records (EMRs) of patients treated for VV using UGFS at a private clinic (“MedSwiss”; Moscow, Russian Federation). It combines an analysis of cross-sectional data obtained at 1 to 2 weeks after the last UGFS session (inclusion time-point) and prospective data of the subsequent 12-month follow-up.

UGFS was performed to ablate the varicose tributaries in addition to the endovascular laser treatment (EVLT) or as a single approach in patients with symptomatic VVs with CEAP clinical class of ≥2. If UGFS followed EVLT, the time interval between procedures amounted 3 to 10 days due to organizational reasons: the EVLT (usually 10–15 procedures per day on Saturdays) was performed in the operating theater, and the UGFS (usually 15–20 procedures per day on Tuesdays) was done in the office at a different location. DUS was obligatory before every session of UGFS to control for efficacy, possible complications, of prior laser ablation. Based on the internal protocol, patients underwent serial DUS follow-up at 1 to 2 weeks after the last session of UGFS and then at 1, 3, 6, and 12 months with the mandatory entry of the results into the EMR. In the cross-sectional analysis, we included limbs of patients, who underwent UGFS without or in association with uncomplicated EVLT and had a visit at the inclusion time-point (one to two weeks after the last UGFS session) appropriately reported in the EMR (Figure 1). All these patients were included in the subsequent follow-up analysis based on data extracted from the EMR of those who attended at least one appointment after the inclusion time-point. Others patients, who attended only one visit at the inclusion time-point and had signs of NTO, were contacted by phone and interviewed about the possible episodes of symptomatic venous thromboembolism (VTE) over time. During this interview, patients were asked if they had any documented episodes of DVT or PE, diagnosed at other medical facilities, or any episodes of acute leg swelling, chest pain, shortness of breath, syncope, or other acute medical illness since the time of the last sclerotherapy session. Additionally, the EMRs of all patients who underwent UGFS regardless of their visit at the inclusion time-point and/or detection of NTO were examined for evidence of documented VTE of other serious complications. The sequence of treatment and follow-up are represented in Figure 1.

Figure 1.

The sequence of treatment and follow-up.

Two experienced surgeons (UGFS since 2012 and EVLT since 2013, about 200-500 interventions/year) performed all sclerotherapy procedures. The foam was produced by the classical Tessari technique using a solution of polidocanol (POD) 30 mg/ml (Aethoxysklerol, Chemische Fabrik Kreussler & Co. GmbH, Wiesbaden, Germany). Sclerosant was mixed with the saline solution in appropriate proportions to provide a lower concentration (POD 1.4 ml with 0.6 ml saline to achieve 2.0%; POD 1 ml with 1 ml saline to achieve 1.5%; POD 0.6 ml with 1.4 ml saline to achieve 1.0%). The foam was prepared using liquid sclerosant and ambient air at a ratio of 1:5 through a system of two 10-ml syringes that were connected by a three-way tap. The volume of the foam for one session was always limited to 10.0 ml. If VVs were not filled appropriately by 10 ml of the foam, a new session after seven days was scheduled. Ultrasound navigation was provided by “AixplorerUltimate” (SuperSonic Imagine, Aix-en-Provence, France) and a linear transducer “SL10-2” with a frequency of 10 МHz.

UGFS was performed only to treat varicose tributaries as an individual approach or in combination with previous laser ablation of trunks and/or perforating veins. We did not plan any sclerotherapy for the GSV or SSV without previous laser treatment.

In patients with primary varicosities, the standard treatment strategy included the initial EVLT of the incompetent GSV, SSV, anterior accessory saphenous vein (AASV), and/or Giacomini vein (GV). In the presence of GSV reflux with a competent terminal valve and the maximal diameter of <10 mm, a saphenous sparing strategy with isolated UGFS of varicose tributaries was used. Incompetent perforating veins (IPV) with a diameter of >3.5 mm were treated with laser and truncal ablation or without ablation (as a part of sparing strategy or in the absence of truncal reflux). In the case of VV recurrence, EVLT was used to treat detectable reflux in preserved trunks and perforating veins. If no reflux in trunks and/or perforating veins was observed, the isolated UGFS of varicose tributaries was used in both cases of primary and recurrent VVs.

Immediately after injection, thigh-length elastic compression stockings with a pressure of 23 to 32 mmHg (RAL GZ-387 standard) were applied, and patients walked for 30 to 40 min. We recommended patients to use the compression stockings throughout day and night for 72 h and then continue to wear them only during the daytime for one month. For symptoms and signs of persistent chronic venous disease or occurrence of uncomfortable induration in the sclerotherapy zone, patients were advised to continue to wear their compression stockings beyond one month. We did not routinely use specific prophylaxis for VTE and recommended it only for those who had a personal history of venous thrombosis or PE. In such situations, a standard prophylactic dose of low-molecular-weight heparin (LMWH, enoxaparin 40 mg subcutaneous once a day) was prescribed within the treatment period and for seven days after the last UGFS session.

The two surgeons (in Russia, only physicians are allowed to perform ultrasound investigations), involved in the laser and sclerotherapy procedures, performed all scheduled duplex scans using the same instrument. Both surgeons were trained in ultrasonography >five years before the study started and had performed >100 DUS a week. They used standard approaches and definitions for vein assessment after treatment.¹⁵ They focused on revealing NTO. For this purpose, the calf and popliteal veins were investigated in the sitting position with a moderate bend of the leg at the knee joint (approximately 45 degrees). Femoral and common femoral veins were evaluated in the supine position. GSV and SSV were assessed in the vertical position. The NTO was defined as noncompressibility or incomplete compressibility of any deep vein or superficial trunks (only GSV and SSV) that were not obliterated by prior EVLT. Perforating vein occlusion was not taken into account if it did not extend into the deep veins. All findings at the first visit were carefully recorded in the EMR. During the next scheduled visits, the full DUS by the same protocol was repeated, and surgeons referred to the primary reports to assess the dynamics of the NTO until its recanalization or involution. Recanalization was defined as restoration of vessel lumen with blood flow detected by color Doppler, and involution was defined as the absence of a typical anatomical structure of the vein in the standard point of the leg. The presence of any intraluminal masses was interpreted as a persisting occlusion. At each visit, all patients were examined clinically, with particular attention paid to the signs of DVT or SVT. All obtained information was recorded in the EMR.

The primary endpoint of the study was symptomatic and asymptomatic NTO of superficial and deep veins detected at the inclusion time-point (one to two weeks after the last session of UGFS). The secondary endpoint was a combination of symptomatic DVT, SVT, and PE detected within the next 12-month follow-up. The retrospective design of data collection and the waiver of patient informed consent were approved by the local institutional review board of the Pirogov Russian National Research Medical University.

Statistical analysis was performed using the SPSS Statistics v.19 package (IBM, Armonk, NY, USA). Data distribution was assessed using the Kolmogorov–Smirnov test. All numerical values were presented as the mean ± standard deviation (M ± σ), and relative values with their 95% confidence interval (CI) were calculated using the Wilson method. Comparison of the relative values was performed using a chi-squared test or a two-sided Fisher’s exact test. Possible predictors of nontarget venous occlusion were determined with univariate binary logistic regression analysis. The treated limb was the main unit for most of the analysis. A p < 0.05 was considered statistically significant.

Results

Data were extracted in February 2018. We identified eligible EMRs from 196 of 268 patients (73%) who underwent UGFS with physician compounded polidocanol foam between 2015 and 2017 (139 women and 57 men; mean age, 44.2 ± 12.2 years; age range, 19–72 years). Nine of the 196 patients (4.6%) had a personal history of DVT and five (2.6%) a history of SVT, so they received prophylactic LMWH. None used oral anticoagulants at the time of intervention. The 72 ineligible patients, who missed the visit at inclusion time-point, were assessed outside the main analysis. All patients attended a control DUS within 1 to 12 months after the last UGFS session, and their EMR were evaluated for evidence of VTE or other complications. No symptomatic VTE events in addition to those given below or other serious adverse events related to UGFS were reported throughout the observation period in all 268 treated patients.

Overall, 257 lower limbs in the 196 eligible patients were treated for symptomatic VVs as follows: left leg, 51.4%, right leg, 48.6%; both legs were treated in 61 of 196 patients (31%). The CEAP clinical class distribution according to the highest one was as follows: C2, 74.0%; C3, 20.0%; C4, 4.5%; and C5, 1.5%. GSV and its tributaries were affected in 85.2%, SSV and its tributaries in 13.2%, and concomitant lesion in GSV and SSV was observed in 1.6%. Recurrent VV after a prior surgery represented 6.6% of all cases. Of all limbs, UGFS followed laser ablation in 86% and was used as a single approach in 14%. Before sclerotherapy, 141 GSV trunks, 5 AASV, 27 SSV, 10 GV, and 67 IPV were treated by laser. Twenty-six limbs combined laser treatment of several different veins. Of 36 limbs that underwent UGFS as a single treatment approach, only three had recurrent VVs, and others had no axial reflux (n = 22) or utilized a saphenous sparing strategy (n = 11). The latter was used only for GSV reflux in 32 limbs (12.5%). The specification of laser ablation, the diameter of the target veins, and technical and clinical success were beyond the scope of this analysis.

Overall, we performed 300 sessions of UGFS in 257 limbs (from 1 to 3 session per limb, in average 1.2 ± 0.4 sessions) using 1.0% to 3.0% POD, as follows: 1.0% for three limbs (1.2%), 1.5% for 102 limbs (39.7%), 2.0% for 117 limbs (45.5%), and 3% for 35 limbs (13.6%). The volume of foam per session varied from 2.0 to 10.0 ml (mean, 7.5 ± 2.3 ml). Obliteration of the target veins (defined as noncompressibility with DUS along the whole length of VV) and clinical success (defined as an absence of palpable and compressible varicose tributaries) were achieved in all cases.

In 95 of 257 limbs (37%), there was a necessity to prolong the use of elastic compression stocking beyond one month due to persistent venous symptoms (n = 41) or uncomfortable induration after sclerotherapy (n = 54).

According to the cross-sectional analysis at the inclusion time-point, the NTO of superficial or deep veins was observed in 60 of 257 limbs (23.3%; 95% CI: 18.6–28.8%), and symptoms were present only in 3 cases (1.2%; 95% CI: 0.4–3.4%), which corresponded to 57 of 196 affected patients (28.5%; 95% CI: 22.6–35.2) with bilateral occlusion in 3 of 61 subjects (5.0%; 95% CI: 1.7–13.6%) who underwent treatment of both legs.

NTO localization is presented in Table 1. Occlusion of the superficial veins was found in 29 limbs (11.3%; 95% CI: 8.0–15.8%), occlusion of deep veins was found in 24 limbs (9.3%; 95% CI: 6.3–13.5%), and the combined superficial and deep vein occlusion (DVO) was present in 7 limbs (2.7%; 95% CI: 1.3–5.5%).

Table 1.

Number of affected superficial and deep veins with nontarget occlusion after ultrasound-guided foam sclerotherapy.

Affected veins	n	Prevalence among occluded veins, %	Incidence among all limbs, %
GSV alone	23	38.2	8.9
Calf muscle veins	15	25.0	5.8
Posterior tibial veins	6	10.1	2.3
SSV alone	5	8.3	1.9
GSV + calf muscle veins	4	6.7	1.6
SSV + calf muscle veins	3	5.0	1.2
GSV + SSV	1	1.7	0.4
Calf muscle veins + popliteal vein	1	1.7	0.4
Femoral vein	1	1.7	0.4
Peroneal veins	1	1.7	0.4
Total	60	100.0	23.3

GSV: great saphenous vein; SSV: small saphenous vein.

Occlusion of previously untreated GSV, alone or in combination with SSV or deep veins, was the most frequent. Overall, the GSV trunk was affected in 28 of 257 limbs (10.9%), which amounted to 47% of all NTOs. The length of the occlusion was different, and it was sometimes limited within the trunk on the calf and sometimes extended up to the saphenofemoral junction. However, the lesion never extended to the common femoral vein. Unfortunately, data on the extent of the occlusion were incomplete and could not be accurately analyzed. NTO of GSV was observed in 21 of 71 limbs (30%) with previously undetected GSV reflux (healthy vein); in 7 of 32 limbs (22%) with refluxing GSV treated with saphenous sparing strategy; in none of 141 limbs (0%) that underwent laser treatment of the GSV trunk immediately before UGFS; and in none of 13 legs (0%) that had recurrent VVs after previous GSV surgery (p < 0.001). However, there was no difference in the occlusion rate between healthy and refluxing GSV in the absence of prior laser treatment (p = 0.480).

The SSV trunk was affected in 9 of 257 limbs (3.5%) and amounted to 15% of all NTOs. There were 9 occlusions in 226 limbs (4.0%) without previous SSV reflux; no occlusion in 27 limbs (0%) that underwent SSV laser treatment immediately before UGFS; and no occlusions in 4 limbs (0%) with recurrent VVS after prior SSV surgery (p = 0.308).

NTOs of AASV or GV were not reported. Thus, the occlusion of untreated GSV appeared 7 times more often than untreated SSV: 28 of 103 limbs (27%) compared to 9 of 226 limbs (4%): relative risk of 6.7 with 95% CI of 3.3–13.8; (p < 0.001).

Among the deep veins, the most often observed occlusions were of the calf muscle veins (soleus and gastrocnemius) separately or in combination with other vessels. Overall, 23 of 60 (38%) limbs were affected. Soleus lesions never extended to other deep veins, and gastrocnemius occlusions in most cases were limited by their own junction. However, in one case, thrombus extended to the popliteal vein, and a free-floating head that was 20 mm in length had formed. We rarely observed occlusion of the tibial or peroneal veins. In one limb, we registered an asymptomatic lesion in the femoral vein in the lower third of the thigh. It presented with tightly fixed hyperechogenic parietal masses, occupying up to 50% of the vessel diameter along the 3 cm vessel length. In all cases, DVOs were connected with obliterated superficial varicose tributaries through the perforating veins. Using extracted data, we were not able to assess the diameter or other characteristics of these perforating veins, but none were mentioned in preoperative DUS protocols as clinically relevant and were not targeted for treatment.

Symptoms of venous occlusion were found in three limbs: two had a lesion of the GSV, and patients complained of low-to-moderate pain above the trunk without any symptoms of skin inflammation, and one had an occlusion of the gastrocnemius vein that manifested with local pain in the calf muscle without edema.

Specific treatment was prescribed in two cases: one patient with moderate symptoms of GSV occlusion received oral sulodexide (heparinoid with slight antithrombotic activity and good anti-inflammatory properties), and the other patient with an asymptomatic free-floating thrombus in the popliteal vein received rivaroxaban. Treatment was continued for one month with complete resolution of symptoms and resolution of the popliteal and calf thrombi without any clinical suspicion of PE. In most cases, patients did not receive any specific recommendations and continued to wear their compression stockings and were followed up with the duplex examination.

At 1 month, 185 limbs (72%) were re-assessed, and occlusion persisted in 31 of 46 (67%) limbs. At 3 months, 112 limbs (44%) were re-assessed, and occlusion persisted in 9 of 27 (33%) limbs. Of these nine limbs, all presented with occlusion of superficial veins (seven cases of GSV and two cases of SSV lesion), and only one presented with occlusion of the calf muscle veins in addition to the superficial lesion. At 6 months, 75 limbs (29%) were re-assessed, and occlusion persisted in 1 of 17 (6%) limbs, which presented as a GSV lesion. At 12 months, only 26 limbs (10%) were re-assessed, and no evidence of occlusion was found in 7 limbs (Figure 1). No propagation of deep or superficial vein lesions was detected during the observation period. The individual outcomes in patients with NTO are represented in Table 2.

Table 2.

The individual outcomes in limbs with NTO.

n	Sex, age	Type of occlusion	1 Month	1 Month	3 Months	6 Months	12 Months	EMR, phone call
1	M, 71	CMV	Resolved	Resolved	Resolved	No FU	No FU
2	F, 64	CMV	Persisted	Persisted	Resolved	No FU	No FU
3	F, 49	GSV+CMV	Persisted	Persisted	Resolved	No FU	No FU
4	F, 39	CMV	No FU	No FU	No FU	No FU	No FU	No evidence of VTE
5	M, 49	SSV+CMV	Resolved	Resolved	No FU	No FU	No FU
6	F, 68	GSV	Persisted	Persisted	Resolved	No FU	Resolved
7	M, 59	GSV	Persisted	Persisted	No FU	No FU	No FU	No evidence of VTE
8	F, 58	CMV	Resolved	Resolved	Resolved	Resolved	No FU
9	F, 46	SSV+CMV	No FU	No FU	Resolved	No FU	No FU
10	F, 68	GSV	Persisted	Persisted	Persisted	Resolved	No FU
11	M, 52	CMV	Resolved	Resolved	No FU	No FU	No FU
12	M, 52	PTV	No FU	No FU	No FU	No FU	No FU	No evidence of VTE
13	F, 35	GSV+CMV	No FU	No FU	No FU	No FU	No FU	No evidence of VTE
14	M, 31	GSV+CMV	Persisted	Persisted	No FU	No FU	No FU	No evidence of VTE
15	F, 31	CMV	Resolved	Resolved	Resolved	No FU	No FU
16	F, 62	SSV+CMV	No FU	No FU	Persisted	No FU	No FU	No evidence of VTE
17	F, 42	SSV	Persisted	Persisted	No FU	Resolved	No FU
18	F, 32	PTV	Resolved	Resolved	No FU	Resolved	No FU
19	F, 42	GSV	Persisted	Persisted	No FU	No FU	No FU	No evidence of VTE
20	F, 46	GSV	Persisted	Persisted	Persisted	No FU	No FU	No evidence of VTE
21	F, 31	CMV	No FU	No FU	No FU	No FU	No FU	No evidence of VTE
22	M, 42	CMV	No FU	No FU	Resolved	Resolved	No FU
23	M, 39	GSV	No FU	No FU	Persisted	No FU	No FU
24	F, 56	GSV+SSV	No FU	No FU	No FU	No FU	No FU	No evidence of VTE
25	M, 42	GSV	Persisted	Persisted	No FU	No FU	No FU	No evidence of VTE
26	M, 42	FV	Persisted	Persisted	No FU	No FU	No FU	No evidence of VTE
27	F, 30	SSV	Resolved	Resolved	No FU	No FU	No FU	No evidence of VTE
28	F, 30	GSV	Persisted	Persisted	No FU	No FU	No FU	No evidence of VTE
29	F, 30	CMV	Persisted	Persisted	No FU	Resolved	No FU
30	M, 48	CMV	Persisted	Persisted	Resolved	No FU	No FU
31	F, 35	CMV+PoV	Resolved	Resolved	Resolved	Resolved	No FU
32	F, 31	CMV	Persisted	Persisted	No FU	No FU	No FU	No evidence of VTE
33	F, 60	CMV	Resolved	Resolved	Resolved	Resolved	No FU
34	F, 41	GSV	Persisted	Persisted	Persisted	No FU	No FU	No evidence of VTE
35	M, 42	SSV	No FU	No FU	No FU	No FU	No FU	No evidence of VTE
36	F, 34	GSV	Persisted	Persisted	Resolved	Resolved	No FU
37	F, 41	PTV	Persisted	Persisted	Resolved	Resolved	No FU
38	F, 40	GSV	Persisted	Persisted	No FU	Resolved	No FU
39	F, 45	GSV	Resolved	Resolved	No FU	No FU	Resolved
40	M, 40	PTV	Resolved	Resolved	No FU	No FU	Resolved
41	M, 40	GSV	Resolved	Resolved	No FU	No FU	Resolved
42	F, 45	GSV+CMV	Persisted	Persisted	Persisted	Persisted	No FU	No evidence of VTE
43	F, 32	CMV	Resolved	Resolved	No FU	Resolved	No FU
44	M, 30	GSV	Persisted	Persisted	No FU	No FU	No FU	No evidence of VTE
45	F, 44	GSV	Persisted	Persisted	No FU	Resolved	No FU
46	F, 45	PeV	Persisted	Persisted	No FU	No FU	No FU	No evidence of VTE
47	F, 31	CMV	Resolved	Resolved	Resolved	No FU	No FU
48	F, 41	CMV	No FU	No FU	Resolved	No FU	No FU
49	F, 29	SSV	No FU	No FU	No FU	No FU	No FU	No evidence of VTE
50	F, 33	GSV	Persisted	Persisted	No FU	No FU	Resolved
51	F, 40	GSV	Persisted	Persisted	No FU	No FU	Resolved
52	F, 34	GSV	No FU	No FU	Persisted	No FU	No FU	No evidence of VTE
53	M, 25	SSV	Persisted	Persisted	Persisted	No FU	No FU	No evidence of VTE
54	F, 47	GSV	Persisted	Persisted	No FU	No FU	No FU	No evidence of VTE
55	M, 31	PTV	Persisted	Persisted	Resolved	Resolved	No FU
56	F, 28	PTV	Resolved	Resolved	No FU	Resolved	Resolved
57	F, 31	GSV	No FU	No FU	No FU	No FU	No FU	No evidence of VTE
58	M, 52	GSV	Persisted	Persisted	Resolved	No FU	No FU
59	M, 50	GSV	Persisted	Persisted	Resolved	No FU	No FU
60	F, 35	GSV	Persisted	Persisted	Persisted	Resolved	No FU

Limbs that presented NTO at the inclusion time-point but were not followed with DUS are marked in bold.CMV: calf muscle vein; EMR: electronic medical record; F: female; FV: femoral vein; FU: follow-up; GSV: great saphenous vein; M: male; PeV: peroneal vein; PoV: popliteal vein; PTV: posterior tibial vein; SSV: small saphenous vein; VTE: venous thromboembolism.

Despite the low rate of follow-up, 52 of 60 limbs (87%) with detected NTO at the inclusion time-point were re-examined at least once during the follow-up period. In all other cases, patients were contacted by phone, and the EMRs were inspected for possible VTE events. No clinical condition suspicious for VTE occurred over time. Thus, there were no other episodes of symptomatic DVT, SVT, or PE during the observation period except for the two symptomatic GSV thromboses and one gastrocnemius thrombosis that was observed at one to two weeks after the last session of UGFS. Thus, the symptomatic VTE event during 12 months observation (secondary endpoint) was detected in three limbs (1.2%; 95% CI: 0.4–3.4%).

Using a univariate logistic regression, we analyzed the following clinical factors as predictors of NTO: sex, age, side of the affected limb, CEAP clinical class, affected GSV or SSV venous system, primary or recurrent disease, UGFS alone or after EVLT, previous laser ablation of GSV, SSV, AASV, GV, IPV, number of UGFS sessions per limb, concentration of sclerosant, volume of foam per session, and LMWH injections associated with personal VTE history (Table 3). Of these, only previous laser ablation of GSV appeared a significant risk factor (p = 0.004). However, this was true only for superficial vein occlusion (4.3% after EVLT and 26.7% without EVLT, p < 0.0001), but not for the DVO (13.5% after EVLT and 10.3% without EVLT, p = 0.564). The personal history of VTE associated with periprocedural use of LMWH was not related to the NTO: 1 occlusion in 19 limbs (5.3%) of 14 VTE-positive patients compared to 59 occlusions in 238 limbs (24.8%) of 182 VTE-negative patients (p = 0.086).

Table 3.

The results of univariate logistic regression for the nontarget superficial or deep vein occlusion.

Possible predictor	Superficial and deep NTO in 257 limbs			NTO of the GSV in 103 limbs with never-treated GSV
Possible predictor	OR	95% CI	p	OR	95% CI	p
Sex (male)	1.084	0.581–2.022	0.800	1.576	0.614–4.045	0.344
Age	0.979	0.955–1.009	0.088	0.986	0.948–1.026	0.486
Side of the affected limb (left)	0.782	0.438–1.396	0.406	0.841	0.348–2.029	0.699
CEAP clinical class (C2)	n/a	n/a	0.999	0.967	0.103–10.653	0.812
Affected venous system (GSV)	0.843	0.370–1.922	0.921	2.587	0.882–7.588	0.083
Recurrent VVs	0.418	0.093–1.884	0.256	–	–	–
Individual history of VTE (positive)	0.169	0.022–1.290	0.086	0.215	0.026–1.753	0.151
GSV diameter	–	–	–	0.964	0.719–1.294	0.809
GSV reflux (saphenous sparing)	–	–	–	0.667	0.250–1.778	0.418
GSV laser treatment immediately before UGFS	0.416	0.230–0.753	0.004	–	–	–
AASV laser treatment immediately before UGFS	0.818	0.090–7.459	0.858	n/a	n/a	0.999
SSV laser treatment immediately before UGFS	0.931	0.358–2.452	0.884	0.615	0.186–2.028	0.424
GV laser treatment immediately before UGFS	2.274	0.620–8.341	0.215	1.380	0.321–5.939	0.665
IPV laser treatment immediately before UGFS	1.294	0.683–2.454	0.429	0.955	0.397–2.293	0.917
UGFS as a single approach	1.802	0.840–3.866	0.130	1.559	0.629–3.860	0.337
Number of UGFS sessions per limb	0.622	0.301–1.453	0.304	1.380	0.321–5.939	0.665
Concentration of POD	0.749	0.404–1.389	0.359	0.535	0.200–1.431	0.213
The volume of foam per session	0.934	0.824–1.059	0.284	0.941	0.785–1.127	0.506

AASV: anterior accessory saphenous vein; CI: confidential interval; GV: Giacomini vein; GSV: great saphenous vein; IPV: insufficient perforating vein; NTO: nontarget occlusion; OR: odds ratio; POD: polidocanol; SSV: small saphenous vein; UGFS: ultrasound-guided sclerotherapy; VTE: venous thromboembolism; VVs: various veins.

Considering the above, we performed a separate analysis for the predictors of the nontarget GSV occlusion in limbs in untreated GSV either immediately before UGFS or previously (n = 103). The same clinical predictors we assessed, except for GSV laser ablation, but included GSV diameter and use of a saphenous sparing strategy. The diameter that was routinely measured in the middle third of the thigh was further extracted from the EMR, which ranged from 2.5 to 8.7 mm (mean, 5.5 ± 1.5 mm). None of these clinical factors demonstrated the ability to predict occlusion of the GSV, including vein diameter (Table 3).

Discussion

Previous studies have shown significantly lower rates of DVO after sclerotherapy as revealed by routine DUS ranging from 1.1% to 3.2%.^9–12 However, these studies focused on a mixed population of patients who underwent less tributary treatment and more ablation of GSV or SSV trunks using liquid or foam sclerotherapy with a different DUS interval. Thus, a control DUS 8 to 30 days after foam sclerotherapy of incompetent GSV and SSV trunks may be warranted.⁹ With proper evaluation of the calf muscle veins, DVO was identified in only 10 of 1025 limbs, of which 6 presented with medial gastrocnemius vein thrombosis and 5 were asymptomatic. However, in our study, part of the calf muscle vein occlusion had already resolved within one month, suggesting a decrease in the frequency of their detection at this time-point. Follow-up with DUS 3 to 5 days after each UGFS session revealed DVO only in 16 of 489 patients treated for saphenous vein and related tributaries or nonsaphenous varices.¹⁰ All identified occlusions were asymptomatic. The authors detected only tibial and no calf muscle vein lesions, which were observed in Gillet et al.⁹ and in our study. The authors attempted to prevent the foam from spreading into the deep veins by placing the injection as far distally and away from the connections between superficial and deep veins as possible. The incidence and characteristics of DVO after UGFS of truncal veins and varicose tributaries were recently determined following DUS at 1 to 2 weeks after treatment: 17 occlusions were detected after 1166 session on 1000 limbs.¹¹ Most appeared after truncal sclerotherapy and affected the common femoral, femoral, or popliteal vein. Only one lesion followed the tributary sclerotherapy and localized in the popliteal vein and one appeared after SSV treatment in the gastrocnemius vein. The authors did not perform any particular measures to prevent the spread of the foam into the deep veins at any point except at the venous junction.

Thus, a higher incidence of detecting DVO in our study may be associated with both an accurate assessment of the calf muscle veins in the short intervals after the session and in the absolute prevalence of treatment for the varicose tributaries. It can be assumed that persistent nonobliterated GSV will drain most of the volume of the foam into the common femoral vein, where it will immediately be inactivated by blood. However, the foam may retain enough activity in the trunk to provide NTO. This may be confirmed by the higher rate of untreated GSV occlusion compared with treated GSV and untreated SSV. At the same time, there was no influence of preexisting GSV reflux on the NTO risk in the absence of truncal ablation: healthy veins were occluded as frequently, as refluxing ones treated with saphenous sparing strategy. Regardless of previous laser ablation of superficial trunks, there was no difference in the incidence of DVO, suggesting perforating veins were a second pathway for foam drainage. Interestingly, previous laser ablation of IPV performed in 26.1% of lower limbs (when recognized as a main source of reflux within the saphenous sparing strategy, or in the absence of axial reflux, or recurrent VVs treatment) did not influence the NTO, suggesting a prevalence of re-entry perforators in the zone of sclerotherapy. Sclerotherapy of tributary veins was associated with a lower risk of DVO when compared with treatment of the trunks, the only independent predictor of DVO was a volume of injected foam >10 ml.¹¹ However, in our practice, we never exceed this threshold, and we found no association between foam volume and NTO.

The number of injections may be the other reason for extensive foam spreading into deep veins. Fewer number of vein punctures (2.8 vs. 5.3 per session) was associated with more frequent foam detection in the deep veins 5 min after UGFS.¹⁶ However, there was no correlation between foam visualization and VTE events during six-month follow-up. In our practice, we usually fill the varicose tributary with the foam through 1 to 3 punctures and do not use any special maneuvers to prevent foam migration into deep veins, like digital compression or choosing of the special point for injection. Further, we do not perform focused duplex evaluation of foam migration in the deep venous system nor recommend any specific exercise to clear the deep veins from the foam. However, among deep veins occasionally scanned during the sclerotherapy procedure, the foam was visualized in all cases, usually in the calf muscle veins, tibial veins, and popliteal vein. Unfortunately, we were unable to extract and analyze the exact number of injections or other facts concerning foam migration into the deep veins. According to these data, we can assume that meticulous pre-treatment mapping and strategizing of optimal locations of injections and sites of compression of possible deep vein re-entry points may potentially reduce the incidence of NTO. However, there is no strong evidence that such preventive measures may improve clinical outcomes.

Treatment of DVOs after sclerotherapy is a complicated issue. In previous studies, LMWH in therapeutic, intermediate, or prophylactic doses was used, as was treatment with aspirin, warfarin, or nothing for various time periods.^9–11 However, in our study, most DVOs were asymptomatic and limited to the calf veins. Treatment of distal DVT is controversial, and performing serial DUS is a recommended option in such cases.^17,18 In most cases, we followed ACCP guidelines and did not treat asymptomatic patients with anticoagulation but performed serial DUS. Thus, we discovered benign characteristics of the observed occlusions that had no tendency for propagation, which usually resolved within three months. Based on this experience even in contradiction with ACCP guidelines, we avoided anticoagulation even in one asymptomatic patient with a segmental lesion of the femoral vein, and the outcomes were good. However, he had a symptomatic GSV occlusion on the contralateral limb and received sulodexide for one month.

Because there was no tendency for propagation in absence of clinical symptoms, it can be assumed that the occlusions that were present differed from spontaneous DVT or SVT and should be termed differently. The term “endovenous foam-induced thrombosis” or “EFIT” has previously been proposed.¹¹ However, more data are required to confirm the benign character of these lesions and the impracticality of their specific treatment.

The limitations of the study included the retrospective collection of previously nondesigned data, a lack of clinical (including typical adverse events of sclerotherapy) and technical information (number of injections, spreading of the foam into the deep veins), as well as low rates and short term of follow-up especially over three months. Despite the lack of follow-up, the results obtained could be supported by the fact that the medical center involved is a first-line clinic that provides medical care to outpatients having a single form of paid medical insurance. Thus, treatment of any type of complication would be carried out in this facility as an outpatient or after discharge, and the EMRs would contain all relevant information.

Conclusion

The frequency of nontarget vein occlusion after UGFS with physician compounded polidocanol foam revealed by serial DUS may be as high as 23.3%. These occlusions tend to resolve within six months and are not associated with symptomatic VTE.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

The study protocol and the waiver of patient-informed consent were approved by the local institutional review board.

Guarantor

KL.

Contributorship

KL, VB, LL, and AD researched literature and conceived the study. AV, AB, ST, and IS collected the data. KL made statistical analysis and wrote the first draft of the manuscript. All authors reviewed and edited the manuscript and approved the final version of the manuscript.

Acknowledgements

This study was presented as the plenary lecture at the 31st Annual Meeting of the American Venous Forum, Rancho Mirage, CA, 19–22 January 2019.

ORCID iD

Kirill Lobastov

References

Rabe

Guex

Puskas

, et al. Epidemiology of chronic venous disorders in geographically diverse populations: results from the Vein Consult Program. Int Angiol 2012; 31: 105–115.

The general prevalences of diseases in the total population of Russia in 2012. Statistical Materials, Part II 2013 (in Russian), http://www.rosminzdrav.ru/documents/8029-statisticheskaya-informatsiya-2012 (2013, accessed 26 October 2019).

The prevalences of diseases in the total population of Russia in 2014. Statistical Materials, Part II 2015 (in Russian), https://www.rosminzdrav.ru/documents/9479-statisticheskaya-informatsiya-za-2014 (2015, accessed 26 October 2019).

The prevalences of diseases in the total population of Russia in 2016. Statistical Materials, Part II 2017 (in Russian), https://www.rosminzdrav.ru/ministry/61/22/stranitsa-979/statisticheskie-i-informatsionnye-materialy/statisticheskiy-sbornik-2016-god (2017, accessed 26 October 2019).

Zolotukhin

Seliverstov

Shevtsov

, et al. Prevalence and risk factors for chronic venous disease in the general Russian population. Eur J Vasc Endovasc Surg 2017; 54: 752–758.

Rabe

Breu

Cavezzi

, et al. European guidelines for sclerotherapy in chronic venous disorders. Phlebology 2014; 29: 338–354.

Nicolaides

Kakkos

Eklof

, et al. Management of chronic venous disorders of the lower limbs - guidelines according to scientific evidence. Int Angiol 2014; 33: 87–208.

Wittens

Davies

Baekgaard

, et al. Editor’s choice – management of chronic venous disease: clinical practice guidelines of the European Society for Vascular Surgery (ESVS). Eur J Vasc Endovasc Surg 2015; 49: 678–737.

Gillet

Guedes

Guex

, et al. Side-effects and complications of foam sclerotherapy of the great and small saphenous veins: a controlled multicentre prospective study including 1,025 patients. Phlebology 2009; 24: 131–138.

10.

Myers

Jolley

Clough

, et al. Outcome of ultrasound-guided sclerotherapy for varicose veins: medium-term results assessed by ultrasound surveillance. Eur J Vasc Endovasc Surg 2007; 33: 116–121.

11.

Kulkarni

Messenger

Slim

, et al. The incidence and characterization of deep vein thrombosis following ultrasound-guided foam sclerotherapy in 1000 legs with superficial venous reflux. J Vasc Surg Venous Lymphat Disord 2013; 1: 231–238.

12.

Gillet

Lausecker

Sica

, et al. Is the treatment of the small saphenous veins with foam sclerotherapy at risk of deep vein thrombosis? Phlebology 2014; 29: 600–607.

13.

Guex

Allaert

Gillet

, et al. Immediate and midterm complications of sclerotherapy: report of a prospective multicenter registry of 12,173 sclerotherapy sessions. Dermatol Surg 2005; 31: 123–128.

14.

Guex

Schliephake

Otto

, et al. The French polidocanol study on long-term side effects: a survey covering 3,357 patient years. Dermatol Surg 2010; 36(Suppl 2): 993–1003.

15.

De Maeseneer

Pichot

Cavezzi

, et al. Duplex ultrasound investigation of the veins of the lower limbs after treatment for varicose veins - UIP consensus document. Eur J Vasc Endovasc Surg 2011; 42: 89–102

16.

Yamaki

Nozaki

Sakurai

, et al. Multiple small-dose injections can reduce the passage of sclerosant foam into deep veins during foam sclerotherapy for varicose veins. Eur J Vasc Endovasc Surg 2009; 37: 343–348.

17.

Guyatt

Akl

Crowther

, et al. Executive summary: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest 2012; 141: 7S–47S.

18.

Kearon

Akl

Ornelas

, et al. Antithrombotic therapy for VTE disease: CHEST guideline and expert panel report. Chest 2016; 149: 315–352.