BACKGROUND Varicose veins recurrence rate remained almost unchanged despite the constant technological advancement in its treatment. The aim of this study is to evaluate the variable accessory saphenous vein (ASV) anatomy at the sapheno-femoral junction (SFJ) as a possible risk factor for recurrent varicose vein (RVV) after great saphenous vein (GSV) radiofrequency thermal ablation (RTA). METHODS Two-hundred consecutive patients affected by chronic venous disease (mean age 52.4±10.3 years; 187 women; CEAP C2-C6; 25.2±1.4), underwent to RTA from 2014 to 2016, at our Institute. Preoperatively all patients underwent duplex-ultrasound scanning, reporting the anatomical site, extension of reflux and the ASV anatomy at the SFJ. Duplex ultrasound and physical examination was performed postoperatively at 1, 6 and 12 months, and yearly thereafter. RESULTS Patients were divided in two groups based on the anatomical site of reflux: group A (N.=187) including GSV and SFJ, group B (N.=82) including SFJ reflux. There was no preoperative statistical difference between the two groups. At a mean follow-up of 29.7±2.4 months, a freedom from recurrent varicose vein and GSV recanalization was: 100% and 100% at 1 month, 95.9% and 99.1% at 1 year, 93.7% and 96.7% at 3 years, respectively. A higher rate of RVV was documented for patients in group A at 3-year of follow-up (P=0.042). Cox regression analysis found, among five potential predictors of outcome, that direct confluence of ASV in SFJ (HR=1.561; 95% CI: 1.0-7.04; P=0.032) was a negative predictors of 1-year RVV. CONCLUSIONS Sapheno-femoral junction morphology may affect recurrent varicose veins formation. In particular, a concomitant incompetence of the accessory saphenous vein or its directly confluence into the SFJ could represent an indication to simultaneous treatment by non-surgical techniques (RTA or laser) and avoid surgical ligation.