Konstantinos Katsanos, Interventional Radiology, Guy’s and St Thomas’ Hospitals, London, reported on data from a mixed treatment comparison network meta-analysis of randomised controlled trials comparing bare nitinol stents, covered nitinol stents, paclitaxel-eluting stents, sirolimus-eluting stents and paclitaxel-coated balloons with each other and with plain balloon angioplasty in the femoropopliteal artery.
Katsanos told Interventional News: “Sixteen randomised controlled trials (n=16 with >2,500 patients) were analysed with a Bayesian probabilistic approach. Immediate technical success proved to be better with use of covered nitinol stents (probability being best >80%), while paclitaxel-eluting stents and paclitaxel-coated balloons offered the best long-term results with the least future events of vascular restenosis and repeat target lesion revascularisation (cumulative probability best >85%).”
The investigators searched medical literature databases such as PubMed, Enbase, AMED, Scopus, Central, and online material in January 2013 for eligible randomised, controlled trial data.
The researchers then fitted a Bayesian generalised random effects binomial model for comparison of different treatment options (WinBUGS). Outcomemeasures included immediate technical success, vascular restenosis (evaluable with Duplex on the 50%–70% threshold) and target lesion revascularisation analysed on an intention-to-treat principle. Pair-wise risk ratios of absolute treatment effects were calculated as the median and accompanying 95% credibility intervals of the posterior distribution and probabilities of each treatment being the best were reported at CIRSE 2013 (14–18 September, Barcelona, Spain).