Network Meta-Analysis of Randomized Controlled Trials Evaluating Long-Term Outcomes of Interventional Approaches for Great Saphenous Vein Insufficiency

Fig 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of the study selection process. CEAP, Clinical-Etiology-Anatomy-Pathophysiology classification; CENTRAL, Cochrane Central Register of Controlled Trials.

Fig 2. Pooled Kaplan-Meier curves showing the cumulative freedom from failure across time.

Fig 3. Forest plot of Cox-proportional hazards model for freedom from failure across time. Error bars represent Holm-Bonferroni-adjusted 95% confidence intervals. P values shown are adjusted using the Holm-Bonferroni method to control the family-wise type I error rate. The global P value was obtained from the log-rank test. Model performance is summarized using the Akaike Information Criterion (AIC) and the concordance index (C-index).

Fig 4. Summary results of frequentist network meta-analysis of time to return to normal activities outcomes. A, Network Graph, the width of the line indicates the number of direct pairwise comparisons. The width of the node indicates the number of samples contained in the treatment; B, Forest plot; C, Pairwise forest plot, treatment effects are expressed as mean differences with corresponding 95% confidence intervals (CIs). High ligation and stripping (HL&S) used as the reference treatment. Negative mean difference (MD) values indicate shorter time to return to normal activities compared with HL&S and vice versa. P scores represent the relative ranking of treatments within the network meta-analysis, with higher values indicating better performance; D, Funnel plot, each point represents an individual study, plotted according to its effect estimate and corresponding standard error. Symmetry of the plot suggests a low risk of small-study effects; E, Net heat plot, a quadratic heatmap with an “ash box” in the middle, indicating the importance of a treatment comparison to estimating other treatment comparisons. The colored background indicates the inconsistency of the treatment in the row…

Fig 5. Summary results of network meta-analysis of Venous Clinical Severity Score (VCSS) at 6 months outcomes. A, Network graph, the width of the line indicates the number of direct pairwise comparisons. The width of the node indicates the number of samples contained in the treatment; B, Forest plot; C, Pairwise forest plot, treatment effects are expressed as mean differences with corresponding 95% confidence intervals. High ligation and stripping (HL&S) used as the reference treatment. Negative mean difference (MD) values indicate lower VCSS at 6 months compared with HL&S and vice versa. P scores represent the relative ranking of treatments within the network meta-analysis, with higher values indicating better performance; D, Funnel plot, each point represents an individual study, plotted according to its effect estimate and corresponding standard error. Symmetry of the plot suggests a low risk of small-study effects; E, Net heat plot, a quadratic heatmap with an “ash box” in the middle, indicating the importance of a treatment comparison to estimating other treatment comparisons. The colored background indicates the inconsistency of the treatment in the row that can contribute to the…

Fig 6. League table of frequentist random-effects model for time to return to normal activities and Venous Clinical Severity Score (VCSS) at 6 months outcomes. Comparison should be read from left to right, with outcome estimates shown at the intersection of the column-defining treatment and the row-defining treatment. Results are presented as mean difference [95% confidence interval (CI)]. For all outcomes, values below 0 indicate a favorable effect. Interventions are listed in alphabetical order, with high ligation and stripping (HL&S) as the reference intervention. The highlighted column indicates treatments demonstrating better outcomes compared with HL&S.