The feasibility, reliability, and incremental value of two-dimensional speckle-tracking for the detection of significant coronary stenosis after treadmill stress echocardiography

BACKGROUND: Two-dimensional speckle-tracking echocardiography (STE) may help detect coronary artery disease (CAD) when combined with dobutamine stress echocardiography. However, few studies have explored STE with exercise stress echocardiography (ESE). We aimed to evaluate the feasibility, reliability, and incremental value of STE combined with treadmill ESE compared to treadmill ESE alone to detect CAD. METHODS: We conducted a case-control study of all consecutive patients with abnormal ESE in 2018-2020 who subsequently underwent coronary angiography within a six-month interval. We 1:1 propensity score-matched these patients to those with a normal ESE. Two blinded operators generated a 17-segment bull's-eye map of longitudinal strain (LS). We utilized the mean differences between stress and baseline LS values in segments 13-17, segment 17, and segments 15-16 to create receiver operator curves for the overall examination, the left anterior descending artery (LAD), and the non-LAD territories, respectively. RESULTS: We excluded 61 STEs from 201 (30.3%) eligible ESEs; 47 (23.4%) because of suboptimal image quality and 14 (7.0%) because of excessive heart rate variability precluding the calculation of a bull's-eye map. After matching, a total of 102 patients were included (51 patients in each group). In the group with abnormal ESE patients (mean age 66.4 years, 39.2% female), 64.7% had significant CAD (> 70% stenosis) at coronary angiogram. In the group with normal ESE patients (mean age 65.1 years, 35.3% female), 3.9% were diagnosed with a new significant coronary stenosis within one year. The intra-class correlation for global LS was 0.87 at rest and 0.92 at stress, and 0.84 at rest, and 0.89 at stress for the apical segments. The diagnostic accuracy of combining ESE and STE was superior to visual assessment alone for the overall examination (area under the curve (AUC) = 0.89 vs. 0.84, p = 0.025), the non-LAD territory (AUC = 0.83 vs. 0.70, p = 0.006), but not the LAD territory (AUC = 0.79 vs. 0.73, p = 0.11). CONCLUSIONS: Two-dimensional speckle-tracking combined with treadmill ESE is relatively feasible, reliable, and may provide incremental diagnostic value for the detection and localization of significant CAD.