Reproducibility of quantitative optical coherence tomography for stent analysis

Aims: To assess the inter- and intra- observer reproducibility for strut count, strut apposition and strut tissue coverage measurements with optical coherence tomography (OCT). Methods and results: Ten drug-eluting stents (244 frames, 1712 struts) imaged with OCT nine months after implantation were analysed by two independent analysts. One of the analysts repeated the analysis of five stents (120 frames, 795 struts) one week later. Offline analysis was performed with the proprietary LightLab Imaging software. The number of struts was counted and lumen and stent area contours were traced. Tissue coverage thickness was measured at 360 degrees of vessel circumference and in front of every individual strut. The number of malapposed struts was determined. There was good agreement for strut number count (Kendall's Tau-b 0.90 for inter- and 0.94 for intra- observer variability). The relative difference for lumen area, stent area and tissue coverage measurements was around 1%. There was complete inter- and intra- observer agreement for malapposed struts classification (4 out of 1708 struts, Kappa=1). Conclusions: In a Corelab setting, the inter- and intra- observer reproducibility for strut count, strut apposition and strut tissue coverage measurements with OCT is excellent. This emphasises the value of OCT as a tool for the clinical long-term assessment of stents.