Optical coherence tomography enables more accurate detection of functionally significant intermediate non-left main coronary artery stenoses than intravascular ultrasound: A meta-analysis of 6919 patients and 7537 lesions

Objective Fractional flow reserve (FFR) is regarded as the gold standard for the physiological assessment of intermediate coronary artery stenoses. However, FFR does not allow assessment of plaque morphology and lesion geometry. Intracoronary imaging techniques such as intravascular ultrasound (IVUS) and optical coherence tomography (OCT) can help treatment planning by optimising stent implantation, which can improve patient outcomes. The aim of this meta-analysis is to compare the efficacy of IVUS and OCT-derived metrics in detecting flow limiting stenoses in non-left main stem lesions. Methods A systematic review of PubMed, Medline, and Cochrane databases was performed and identified studies examining the diagnostic accuracy of IVUS and OCT in detecting significant stenoses when compared to FFR. Results A total of 33 (7537 lesions) studies (24 IVUS, 7 OCT and 2 IVUS & OCT studies) were included in the meta-analysis. Pooled analysis showed that IVUS- and OCT-derived minimum lumen area (MLA) had a similar sensitivity in predicting haemodynamically significant lesions (IVUS-MLA: 0.747 vs OCT-MLA 0.732, p = 0.519). However, OCT-MLA had a higher specificity (0.763 vs 0.665, p < 0.001) and diagnostic accuracy in detecting flow-limiting stenoses than IVUS-MLA (AUC 0.810 vs 0.754, p = 0.045). Sub-analysis of the studies with the clinically significant FFR cut-off value of 0.80 yielded similar results demonstrating that OCT-MLA has a better accuracy than IVUS-MLA in detecting haemodynamically significant stenoses (AUC 0.809 vs 0.750, p = 0.034). Conclusions OCT with its superior image resolution appears to be the preferable intravascular imaging modality for the detection of haemodynamically significant stenoses in non-left main stem lesions.