Abstract 10115: The Diagnostic Value of Global and Territorial Longitudinal Strain at Rest for the Detection of Coronary Artery Disease in Patients Without Type 2 Diabetes Mellitus

Abstract

Background: Global longitudinal strain (GLS) at rest aids the detection of coronary artery disease (CAD). However, myocardial strains are affected by both ischemia and diabetes mellitus (DM), and previous studies that evaluated the performance of GLS for detecting ischemia always included a certain DM patients in the study population. Thus, we sought to investigate in patients with no DM the power of GLS for detecting three-vessel CAD, and whether territorial longitudinal strain (TLS) could help identifying individual coronary artery stenosis Methods and results: We retrospectively studied 211 consecutive patients with suspected CAD and normal left ventricular (LV) ejection fraction. The patients with DM were excluded. All patients underwent echocardiography and subsequently coronary angiography. LV global and segmental peak systolic longitudinal strain (PSLS) parameters were quantified by two-dimensional speckle tracking echocardiography (2D STE). Territorial PSLSs were calculated based on the perfusion territories of the 3-epicardial coronary arteries in a 17-segment LV model. Critical CAD was defined as luminal diameter stenosis ≥ 70% in ≥ 1 epicardial coronary artery. Totally 145 patients had critical CAD on coronary angiography. Significant differences were observed in all strain parameters between patients with and without CAD. The AUC for GLS in the detection of three-vessel CAD was 0.875 at a cutoff value of -19.05% with sensitivity 78.1% and specificity 72.7%, which increased to 0.926 after excluding apical segments (cutoff value -18.66%; sensitivity 84.4% and specificity 81.8%). The TLS values were significantly lower in regions supplied by stenotic compared with non-stenotic coronary arteries. It has better power to identify LCX and LAD stenosis than RCA stenosis. An area under the curve (AUC) for the TLS to identify critical LCX, LAD and RCA stenosis, in order of diagnostic accuracy, is 0.818 for LCX, 0.764 for LAD and lastly 0.723 for RCA. Conclusions: In patients with no DM and suspected CAD, GLS is an excellent predictor of three-vessel CAD with high accuracy. A higher cut point than that reported before was obtained and should be used. TLS could identify which coronary artery is stenotic with fair sensitivity and specificity.