Population-Based Analysis of Venous Thrombotic Events Following Saphenous Ablation

Abstract

Background: Venous thrombotic events following great and small saphenous vein ablation procedures for varicose veins have been reported. Current knowledge of these events is based on single institution studies with small sample sizes. Methods: The National Surgical Quality Improvement Program (NSQIP) database (2005-2009) was used to identify a total of 3,874 patients who underwent radiofrequency ablation (RFA) or endovenous laser ablation (EVLA) of the great and small saphenous veins with or without stab phlebectomy. Outcome variables included clinically documented postoperative deep vein thrombosis (DVT) or pulmonary embolism (PE). Bivariate and multivariate logistic regression analyses were performed to identify factors associated with venous thrombotic events after ablation procedures. Results: A total of 2,897 (74.8%) patients underwent RFA and 977 (25.2%) underwent EVLA for lower extremity varicose veins. Patients who underwent RFA were more likely to be older (53.8 y vs 51.8 y, P .0001), obese (42.8% vs 38%, P .009), diabetic (8.5% vs 6.4%, P .01) hypertensive (31.9% vs 26.8%, P .002) and to have undergone procedures involving both veins (24% vs 4%, P .0001). Patients who underwent EVLA were more likely to have received general anesthesia (56.9% vs 40.8%, P .0001), to have undergone concomitant stab phlebectomy (44.9% vs 31.7%, P .0001), and had longer operative times (63.6 min vs. 57.3 min, P .0001). The incidence of DVT (1.74% vs 1.52%, P .63) and pulmonary embolus (0.07% vs 0%, P 1) were similar between EVLA and RFA. No significant predictors of DVT in the postoperative period were identified on bivariate or multivariate analyses. In the subgroup of patients who underwent ablation procedures only (no stab phlebectomy, N 2514), there was a trend toward higher incidence of DVT in patients undergoing EVLA (2.6% vs 1.4%, P .057). After adjusting for patient demographics, patients presenting with lower extremity ulcers were 2.4 times more likely to develop DVT compared to those without ulcers (OR 2.4, 95% CI 1.01-6.11, P .04). In the multivariate model when only ablation procedures were performed, EVLA was associated with an 83% increase in odds of DVT compare to RFA, although not statistically significant (OR 1.83, 95% CI 0.95-3.52, P .06). Conclusions: The incidence of venous thrombotic events after saphenous ablation is low. However, given that patients with lower extremity ulcers experienced an increased risk of DVT, care should be taken to ensure that the ablation catheter is positioned an appropriate distance from the sapheno-femoral or sapheno-popliteal junction, and postprocedural preventative measures such as early ambulation and lower extremity compression should be emphasized. The finding of a trend toward increased venous thrombotic events in patients undergoing EVLA warrants further investigation in a large patient population.