Discrepancy between frequency domain optical coherence tomography and intravascular ultrasound in human coronary arteries and in a phantom in vitro coronary model

Abstract

BackgroundThis purpose of this study is to evaluate, concomitantly with quantitative coronary angiography (QCA), the potential discrepancy between frequency domain optical coherence tomography (FD-OCT) and intravascular ultrasound (IVUS) measurements in a phantom coronary model and in human coronary arteries within and outside stented segments. MethodsFD-OCT and IVUS images sequentially obtained from a phantom coronary model and 57 stented human coronary arteries were compared between each other and with QCA. ResultsLumen area (LA) by IVUS was 10.1% larger (6.43±0.09mm2) while by FD-OCT was similar (5.78±0.09mm2) to actual phantom LA (5.72mm2); IVUS vs. FD-OCT stent area (SA) was 4.2% larger. In human coronary artery, diameter by QCA was smaller than by IVUS and OCT in reference (by 10.5% and 3.5%, both p<0.001) and stented (3.6%, p<0.001; and 1.7%, p=0.012) segments. IVUS vs. FD-OCT distal reference LA was significantly larger (6.19±2.18mm2 vs. 5.49±2.49mm2, p<0.001, respectively), and SA was numerically larger (7.42±2.28mm2 vs. 7.22±2.48mm2, p=0.059) with larger discrepancy in reference (11.3%) than stented (2.7%) segments. IVUS vs. FD-OCT correlation for diameter was significantly higher for stented than reference segments (R2=0.8670 vs. 0.7351, p=0.047), while numerically higher for area (R2=0.8663 vs. 0.7806, p=0.157). ConclusionsIn phantom model and human coronary arteries, IVUS vs. FD-OCT measurements were larger, particularly in non-stented than stented segments, and diameter was smaller by QCA vs. IVUS or FD-OCT. Despite undefined clinical significance, said discrepancy warrants consideration.