A network meta-analysis on the efficacy and safety of thermal and nonthermal endovenous ablation treatments.

OBJECTIVE

We assessed the mid-term efficacy and safety of thermal and nonthermal endovenous ablation for the treatment of lower limb superficial venous insufficiency.

METHODS

We performed a systematic review in accordance with the PRISMA (preferred reporting items for systematic reviews and meta-analyses) statement and a Bayesian network meta-analysis. The primary end points were great saphenous vein (GSV) closure and venous clinical severity score (VCSS) improvement. A meta-regression using GSV diameter as a covariate was undertaken for the two primary end points.

RESULTS

We included 14 studies and 4177 patients, with a mean follow-up of 25.7 months. Radiofrequency ablation (RFA; odds ratio [OR], 3.99; 95% confidence interval [CI], 1.82-10.53), cyanoacrylate ablation (CAC; OR, 3.09; 95% CI, 1.35-8.37), and endovenous laser ablation (EVLA; OR, 2.72; 95% CI, 1.23-7.38) displayed increased odds for GSV closure compared with mechanochemical ablation (MOCA). MOCA inferiority compared with RFA (mean difference [MD], 0.96; 95% CI, 0.71-1.20), EVLA (MD, 0.94; 95% CI, 0.61-1.24), and CAC (MD, 0.89; 95% CI, 0.65-1.15) was also depicted regarding VCSS improvement. EVLA resulted in an increased risk of postoperative paresthesia compared with MOCA (risk ratio [RR], 9.61; 95% CI, 2.32-62.29), CAC (RR, 7.90; 95% CI, 2.44-38.16), and RFA (RR, 6.96; 95% CI, 2.31-28.04). Although the overall analysis identified nonstatistically significant differences for Aberdeen varicose vein questionnaire score improvement, thrombophlebitis, ecchymosis, and pain, further investigation revealed an increase pain profile for EVLA at 1470 nm compared with RFA (MD, 3.22; 95% CI, 0.93-5.47) and CAC (MD, 3.04; 95% CI, 1.05-4.97). A sensitivity analysis displayed a persistent underperformance of MOCA compared with RFA (OR, 4.33; 95% CI, 1.15-55.54) for GSV closure and both RFA (MD, 0.99; 95% CI, 0.22-1.77) and CAC (MD, 0.84; 95% CI, 0.08-1.65) regarding VCCS improvement. Although no regression model reached statistical significance, the GSV closure regression model revealed a trend for considerably decreased efficacy for both CAC and MOCA with larger GSV diameters compared with RFA and EVLA.

CONCLUSIONS

Although our analysis has produced skepticism regarding the efficacy of MOCA in the mid-term period for VCSS improvement and GSV closure rates, CAC showed equivalent results compared with both RFA and EVLA. Additionally, CAC displayed a decreased risk of postprocedural paresthesia and pigmentation and induration compared with EVLA. Also, both RFA and CAC had an improved pain profile compared with EVLA 1470 nm. The potential underperformance of nonthermal, nontumescent ablation modalities in ablating large GSVs necessitates further research.