Clinical risk factors to predict deep venous thrombosis post-endovenous laser ablation of saphenous veins

Introduction

Lower extremity venous insufficiency or venous incompetence, is a common medical condition that affects 25–30% of adult women and about 15% of men in Western society. ¹ A majority of patients with varicose veins have insufficiency of the truncal veins in the legs. Most frequently the great saphenous vein (GSV) is involved, but incompetence of the small saphenous vein (SSV) is not uncommon and can occur in up to 20% of individuals with varicose veins. ² Classic symptoms and signs of venous insufficiency are aching, discomfort, oedema and muscle cramps. Associated complications include eczema, lipodermatosclerosis, atrophy blanche, superficial thrombophlebitis and venous ulcerations. Chronic venous insufficiency has a great impact on patients’ health-related quality of life, which is comparable to other common diseases and is associated with considerable healthcare costs. ³

The treatment of varicose veins reduces the symptoms and complications of chronic venous insufficiency and improves health-related quality of life. Over the past decade, endovenous laser ablation therapy (EVLA) of the saphenous veins has been proven safe, with long-term results that are comparable or superior to traditional surgical stripping and high ligation.^4–6 Although technical success of EVLA is close to 100%, post-EVLA complications – such as postprocedural pain, ecchymosis, tenderness and phlebitis – are common.^7,8 One of the more concerning side-effects is deep venous thrombosis (DVT), which has been reported in up to 7.7%^4,5,9,10 of those who have undergone EVLA. The purpose of this study is to examine risk factors for developing DVT post-EVLA.

Methods

Results

Data from 360 consecutive EVLA procedures in 353 patients (247 women, 106 men) were collected and reviewed. Mean age (± SD) was 58 ± 14 years.

Follow-up at one week was 100%, and the technical success rate at one week was 100%. Technical success was defined as ultrasonographic evidence of great or small saphenous vein incompressibility suggesting occlusion without colour flow. A total of nineteen DVTs were found on follow-up DUS: 18 involved either the saphenofemoral or saphenopopliteal junctions as endothermal heat-induced thrombosis (EHIT) type II or III, ¹¹ and one case of deep system (popliteal vein) involvement. By definition, ¹¹ EHIT type I involves thrombosis to the level of the superficial-deep junction and is deemed clinically insignificant, EHIT type II involves thrombus extension into the deep venous system with cross-sectional area ≤50%, EHIT type III involves thrombus extension into the deep venous system with cross-sectional area ≥50%, and EHIT type IV is total occlusion of the deep vein.

Demographics and clinical history of the patients developed DVT were compared with those who did not. Among the categorical variables were higher prevalences of HLD, VTE, SVT and female gender in those who developed DVT post-EVLA (Table 1). Of the continuous variables, only age and glomerular filtration rate were statistically significant: P = 0.001 and P = 0.006, respectively (Table 2). To determine which variables were the strongest predictors of DVT when other potential predictors were included in the model, a step-wise logistic regression analysis was performed. When all variables were considered, only history of SVT (P = 0.002), female gender (P = 0.048) and age >66 (P = 0.007) were significant predictors of the development of DVT (Table 3). To further define the impact of age on the development of DVT following EVLA, a second step-wise regression was performed using age thresholds in place of the continuous variable of age. This model yielded the greatest predicative power, with an age threshold of >66 years (P = 0.007). The resulting model yielded odds ratios for history of SVT, female gender and age >66 years as 3.6, 2.6 and 4.1, respectively (Table 3).

OPEN IN VIEWER

Table 1

Categorical variables

	Female gender	DM	HTN	HLD	CAD	Stroke/TIA	PAD	VTE	SVT
No DVT	291	58	192	154	33	20	37	24	30
DVT	14	2	16	15	3	0	5	4	60
Percent (%)	4.8	3.4	8.3	9.7	9.1	0	13.5	16.7	200
P value	0.05	0.35	0.07	0.019	0.5	0.25	0.07	0.048	0.004

DM, diabetes mellitus; HTN, hypertension; HLD, hyperlipidaemia; CAD, coronary artery disease; TIA, transient ischaemic attack; PAD, peripheral artery disease; VTE, venous thromboembolism; SVT, superficial venous thrombosis; DVT, deep vein thrombosis

OPEN IN VIEWER

Table 2

Continuous variables

	Age (years)	Weight (kg)	GFR (mL/min)	Distance of laser tip to either SFJ or SPJ (cm)	Total distance of ablation (cm)
No DVT
Average	58.24299	90.1875	75.2482	3.32891	29.98442
SD	13.78598	22.07871	16.18595	1.550298	11.00482
DVT
Average	69.45	83.55	63.23529	3.424	31.025
SD	12.81231	15.35707	15.61461	1.093042	13.35199
P value	0.001	0.08	0.006	0.7	0.7

SFJ, saphenofemoral junction; SPJ, saphenopopliteal junction; DVT, deep vein thrombosis; SD, standard deviation; GFR, glomerular filtration rate

OPEN IN VIEWER

Table 3

Clinical risk factors to predict deep vein thrombosis post-endovenous laser therapy

Odds ratio	P value
Age >66 years	4.1	0.007
Female gender	2.6	0.048
History of SVT	3.6	0.002

SVT, superficial venous thrombosis

Discussion

Minimally invasive techniques for the treatment of saphenous vein incompetence have largely supplanted more invasive vein treatments due to the relative simplicity of the procedures and high patient satisfaction. The ultimate goal is to cause fibrotic changes to the incompetent saphenous vein, leading to diminution.

EVLA accomplishes this goal by releasing thermal energy to both the blood and the venous wall, causing localized tissue damage that ends in collagenization. It has demonstrated high technical success rates with minimal complications. Nonetheless, the most serious complication, DVT, has been reported in up to 7.7% of post-EVLA patients.^4,5,9,10 Our analysis found age >66 years, female gender, and a history of SVT to be clinical factors that predict DVT post-EVLA. Prior studies on EVLA and its complications have shown that age >50 years^11,12 and SSV ablation ⁷ are possible risk factors predicting DVT formation.

Differences in DVT rates may be related to variability in technique, the timing of follow-up DUS, or the definition of DVT. The criteria we have used – any involvement of the saphenofemoral or saphenopopliteal junctions (EHIT type II, III, IV) ¹¹ and/or any deep veins – are very conservative.

In any event, our technical success was 100% at one week, and our incidence of DVT after EVLA was 5.27%. This incidence rate falls within the reported rate of DVT post-EVLA: up to 7.7%, as previously described.^4,5,9,10 All of the DVTs were found within one week of the index procedure. Postprocedural symptoms in patients with DVT were indistinguishable from those without DVT. The diagnosis of DVT was made solely from routine DUS. None of the patients with DVT had any clinically obvious adverse outcomes, such as post-thrombotic syndrome, pulmonary embolism or death.

The limitations of our study include the retrospective nature of analysis and single-center experience, so the ability to generalize may be limited.

Our data demonstrate that EVLA of the GSV and SSV is feasible and safe and has excellent clinical outcomes. The incidence of DVT was not negligible, although no patients had an adverse outcome related to DVT or anticoagulation. The lack of distinguishing symptoms in patients with DVT compared with those without highlights the importance of early postprocedure DUS to demonstrate this complication. Further investigation may ascertain whether prophylactic anticoagulation is warranted in some patients.