An indirect comparison by Bayesian network meta-analysis of drug-coated devices versus saphenous vein graft bypass in femoropopliteal arterial occlusive disease

J Vasc Surg. 2021 Aug;74(2):478-486.e11. doi: 10.1016/j.jvs.2020.11.054. Epub 2021 Feb 16.


Objective: To compare the efficacy and safety between drug-coated devices (DCDs) and bypass surgery with saphenous vein graft (BSV) in femoropopliteal arterial occlusive disease.

Methods: A Bayesian network meta-analysis and indirect comparison were performed. Randomized controlled trials of BSV, bypass surgery with prosthetic graft, bare metal stents, endoluminal bypass (covered stent), percutaneous transluminal angioplasty, and DCDs treating femoropopliteal arterial occlusive disease were collected. The primary end point was target lesion revascularization/target vessel revascularization, and secondary end points were all-cause mortality, limb salvage, and early complications (PROSPERO registry number: CRD42019136530).

Results: Forty-two trials and 6867 patients were included. The comparison of DCDs and BSV revealed no significant difference in the 1-year target lesion revascularization/target vessel revascularization (DCDs vs BSV: odds ratio [OR], 0.60; 95% credible interval [CrI], 0.16-2.39). Total early complications from BSV were significantly higher than those from DCDs (DCDs vs BSV: OR, 0.14; 95% CrI, 0.05-0.45), and the main complications of BSV were not death related. There was also no significant difference in systemic early complications (DCDs vs BSV: OR, 0.19; 95% CrI, 0.00-7.82) and 1-year amputation rate (DCDs vs BSV: OR, 2.81; 95% CrI, 0.16-89.53). The 30-day (DCDs vs BSV: OR, 0.38; 95% CrI, 0.00-110.46), 1-year (DCDs vs BSV: OR, 0.96; 95% CrI, 0.24-3.29), 2-year (DCDs vs BSV: OR, 1.60; 95% CrI, 0.64-4.95), and 5-year all-cause mortality rates (DCDs vs BSV: OR, 2.05; 95% CrI, 0.92-4.39) showed no significant differences between DCDs and BSV, although there was a noticeable tendency toward significant results of a higher 5-year mortality rate.

Conclusions: There is no significant difference between DCDs and BSV in short-term efficacy or short- and long-term mortality. Despite traditional BSV remaining the gold standard, DCDs provide a reasonable alternative therapy. In addition, the DCDs have a lower short-term morbidity associated with the procedure at the cost of the possible risk of higher long-term mortality. Clinical trials with more validity are required for a direct comparison between BSV and DCDs.

Keywords: Bypass surgery; Drug-coated balloon; Drug-eluting stent; Femoropopliteal arterial occlusive disease; Indirect comparison.

Publication types

  • Meta-Analysis
  • Systematic Review

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Bayes Theorem
  • Coated Materials, Biocompatible*
  • Drug-Eluting Stents*
  • Endovascular Procedures / adverse effects
  • Endovascular Procedures / instrumentation*
  • Endovascular Procedures / mortality
  • Female
  • Femoral Artery / diagnostic imaging
  • Femoral Artery / physiopathology
  • Femoral Artery / surgery*
  • Humans
  • Limb Salvage
  • Male
  • Middle Aged
  • Network Meta-Analysis
  • Peripheral Arterial Disease / diagnostic imaging
  • Peripheral Arterial Disease / mortality
  • Peripheral Arterial Disease / physiopathology
  • Peripheral Arterial Disease / therapy*
  • Popliteal Artery / diagnostic imaging
  • Popliteal Artery / physiopathology
  • Popliteal Artery / surgery*
  • Postoperative Complications / etiology
  • Randomized Controlled Trials as Topic
  • Risk Assessment
  • Risk Factors
  • Saphenous Vein / transplantation*
  • Time Factors
  • Treatment Outcome


  • Coated Materials, Biocompatible