Comparative efficacy of endovascular interventions for femoropopliteal in-stent restenosis: A systematic review and meta-analysis.

BACKGROUND: To evaluate the comparative efficacy and safety of all available endovascular interventions versus standard percutaneous transluminal angioplasty (PTA/POBA) for femoropopliteal in-stent restenosis (ISR), incorporating randomized and non-randomized evidence. METHODS: PubMed, Embase, and Cochrane Library were searched from inception to May 2026. Twenty-four studies were included: nine randomized controlled trials (RCTs, n = 942) for quantitative meta-analysis and fifteen additional studies (13 observational/registry studies and 2 contextual RCTs with mixed populations; n = 1,740) for qualitative synthesis. Effect sizes are expressed as odds ratios (ORs) with 95% confidence intervals (CIs); heterogeneity was assessed by Q and I2 statistics. A random-effects model (REML estimation with Hartung-Knapp-Sidik-Jonkman [HKSJ] confidence intervals) was pre-specified as the primary pooling approach for all outcomes, given anticipated clinical and methodological heterogeneity; fixed-effects (Mantel-Haenszel) estimates are reported only as sensitivity analyses. RESULTS: Drug-coated balloon (DCB) angioplasty significantly reduced target lesion revascularization (TLR) at 12 months versus PTA/POBA (OR 0.164; 95% CI 0.103-0.263; P < 0.001; I2 = 1.7%; 6 RCTs, n = 475, excluding one trial with a mismatched 24-month timepoint that is now reported separately - see Results) and recurrent ISR at 6 months (OR 0.223; 95% CI 0.129-0.385; P < 0.001; I2 = 0%). Excimer laser atherectomy (ELA)-based strategies with adjunctive DCB were associated with lower odds of TLR versus PTA in the only two available RCTs (OR 0.396; 95% CI 0.248-0.630; P < 0.001; I2 = 0%), although these two trials used non-identical outcome definitions (freedom-from-TLR and recurrent ISR, respectively) and this estimate should be interpreted as exploratory (see Discussion). Covered stent grafts achieved 74.8% primary patency versus 28.0% for PTA at 12 months in a single small RCT. Paclitaxel-eluting stents yielded 78.8%-85.5% primary patency at 12 months in single-arm registry data. Combining ELA with DCB provided freedom-from-TLR of 72.5%-90.9% versus ELA with conventional balloon in three retrospective comparative studies; the only randomized comparison against DCB alone (INTACT) found essentially identical 18-month ISR rates (30.0% vs 30.2%). Cryoplasty was inferior to conventional PTA. No significant difference in all-cause mortality was observed between active treatment and PTA within the available follow-up (pooled OR 0.725; 95% CI 0.276-1.904; median follow-up 12 months, maximum 24 months); no trial followed patients beyond 24 months. CONCLUSION: DCB angioplasty is supported by consistent randomized evidence as first-line treatment for femoropopliteal ISR, reducing target lesion revascularization and recurrent restenosis compared with PTA/POBA. Evidence for laser atherectomy combined with DCB in complex Tosaka II-III lesions, paclitaxel-eluting stents as salvage therapy, and covered stent grafts derives predominantly from small trials, single registries, or retrospective comparisons, and should be regarded as hypothesis-generating rather than established practice pending dedicated randomized confirmation. Device selection for complex lesions should currently be individualized, and reported mortality equivalence should be interpreted as reassuring only within the observed 12-24-month follow-up window.