Variability in quantitative and qualitative analysis of intravascular ultrasound and frequency domain optical coherence tomography. Catheterization and cardiovascular interventions Abnousi, F., Waseda, K., Kume, T., Otake, H., Kawarada, O., Yong, C. M., Fitzgerald, P. J., Honda, Y., Yeung, A. C., Fearon, W. F. 2013; 82 (3): E192-9


BACKGROUND: Frequency-domain optical coherence tomography (FD-OCT) is an intravascular imaging technique now available in the United States. However, the importance of level of training required for analysis using intravascular ultrasound (IVUS) and FD-OCT is unclear. The aim of this study was to evaluate inter- and intra-observer variability between expert and beginner analysts interpreting IVUS and FD-OCT images. METHODS AND RESULTS: Two independent expert analysts and two independent beginner analysts evaluated a total of 226?±?2 stent cross-sections with IVUS and 232?±?2 stent cross-sections with FD-OCT in 14 patients after stenting. Inter- and intra-observer variability for determining stent volume index (VI), as well as identifying incomplete stent apposition and dissection were assessed. The inter- and intra-observer variability of stent VI was minimal for both beginner and expert analysts regardless of imaging technology (random variability: 0.38 vs. 0.05 mm(3) /mm for IVUS, 0.26 vs. 0.08 mm(3) /mm for FD-OCT). Although qualitative IVUS analysis at the patient level revealed no significant difference between beginners and experts, this was not the case for FD-OCT. The number of overall qualitative findings noted by beginner and expert analysts were more variable (overestimated or underestimated) with FD-OCT. CONCLUSION: Despite varying levels of training, the increased resolution of FD-OCT compared to IVUS provides better detection and less variability in quantitative image analysis. On the contrary, this increased resolution not only increases the rate but also the variability of detection of qualitative image analysis, especially for beginner analysts. © 2013 Wiley Periodicals, Inc.

View details for DOI 10.1002/ccd.24871

View details for PubMedID 23412754