Accuracy of intravascular ultrasound and optical coherence tomography in identifying functionally significant coronary stenosis according to vessel diameter: A meta-analysis of 2,581 patients and 2,807 lesions

Am Heart J. 2015 May;169(5):663-73. doi: 10.1016/j.ahj.2015.01.013. Epub 2015 Feb 21.


Introduction: Accuracy of intracoronary imaging to discriminate functionally significant coronary stenosis according to vessel diameter remains to be defined.

Methods: PubMed, Scopus, and Google Scholar were systematically searched for studies assessing diagnostic accuracy (area under the receiver operating characteristic curve [AUC], the primary end point) and sensitivity and specificity (the secondary end points) of minimal luminal area (MLA) or of minimal luminal diameter (MLD) derived from intravascular ultrasound (IVUS) or optical coherence tomography (OCT) to detect functionally significant stenosis as determined with fractional flow reserve (FFR).

Results: Fifteen studies were included, 2 with 110 patients analyzing only left main (LM), 5 with 224 patients and 306 lesions using OCT, and 9 with 1532 patients and 1681 lesions with IVUS. Median MLA for the OCT studies was 1.96 mm(2) (1.85-1.98 mm(2)), 2.9 mm(2) (2.7-3.1 mm(2)) for MLA of all lesions assessed with IVUS, 2.8 mm(2) (2.7-2.9 mm(2)) for lesions with an angiographic diameter >3 mm, 2.4 mm(2) (2.4-2.5 mm(2)) for lesions <3 mm, and 5.4 mm(2) (5.1-5.6 mm(2)) for LM lesions. For OCT-MLA, AUC was 0.80 (0.74-0.86), with a sensitivity of 0.81 (0.74-0.87) and specificity of 0.77 (0.71-0.83), whereas OCT-MLD had an AUC of 0.85 (0.79-0.91), sensitivity of 0.74 (0.69-0.78), and specificity of 0.70 (0.68-0.73). For IVUS-MLA, AUC was 0.78 (0.75-0.81) for all lesions, 0.78 (0.73-0.84) for vessels with a diameter >3 mm, and 0.79 (0.70-0.89) for those with a diameter <3 mm. Left main AUC was 0.97 (0.93-1).

Conclusion: Intravascular ultrasound and OCT had modest diagnostic accuracy for identification hemodynamically significant lesions, also with specific cutoff for different diameters. Invasive imaging for assessment of LM severity demonstrated excellent correlation with FFR. What is already known about this subject? Fractional flow reserve represents the criterion standard to evaluate the prognostic value of coronary stenosis, whereas its relationship with IVUS and OCT remains to be assessed. What does this study add? Despite improvement, IVUS and OCT do not predict functional stenosis, even with dedicated cutoff, apart from LM disease. How might this impact on clinical practice? The recent guidelines of myocardial revascularization have stressed the crucial role of FFR before performing percutaneous coronary intervention on LM, whereas intravascular imaging is often exploited to drive revascularization. The present analysis stresses the point that LM percutaneous coronary intervention may be driven only by intravascular imaging, given the high accuracy for significant ischemic lesions, whereas for other vessels, these 2 techniques mirror 2 different aspects.

Publication types

  • Meta-Analysis

MeSH terms

  • Adult
  • Area Under Curve
  • Coronary Stenosis / diagnosis*
  • Coronary Stenosis / diagnostic imaging
  • Coronary Vessels / diagnostic imaging
  • Coronary Vessels / pathology*
  • Female
  • Humans
  • Male
  • Middle Aged
  • ROC Curve
  • Sensitivity and Specificity
  • Tomography, Optical Coherence*
  • Ultrasonography, Interventional*