Echocardiographic prediction of functional coronary stenosis: global longitudinal strain as a key determinant of quantitative flow ratio.

OBJECTIVE

To evaluate the independent and combined value of global longitudinal strain (GLS) and global work efficiency (GWE) for predicting quantitative flow ratio (QFR) in patients with ischemic heart disease (IHD).

METHODS

In a prospective study of 222 patients with stable IHD, all participants underwent coronary angiography with QFR analysis and comprehensive 2D speckle-tracking echocardiography. LASSO regression for variable selection, multivariable logistic regression for association analysis, restricted cubic splines for non-linearity assessment, and SHAP analysis for model interpretation were employed.

RESULTS

Patients with normal QFR (≥ 0.80) showed significantly better GLS (-17.84 ± 2.34% vs -14.47 ± 2.65%, P < 0.01) and GWE (94.00[91.00-95.00]% vs 91.00[88.00-93.00]%, P < 0.01) compared to those with abnormal QFR. LASSO regression identified both GLS (OR = 2.45, 95%CI 1.82-3.30) and GWE (OR = 1.31, 95%CI 1.15-1.50) as remaining independently associated with normal QFR after full adjustment. Both relationships demonstrated significant non-linearity (P < 0.05). GLS alone showed excellent diagnostic accuracy (AUC = 0.862), while adding GWE provided only marginal improvement (AUC = 0.872, P = 0.15). SHAP analysis confirmed GLS as the most important predictor, with good model generalizability (AUC = 0.838) upon validation.

CONCLUSION

GLS is a powerful independent predictor of coronary hemodynamics assessed by QFR in IHD patients. Although GWE shows an independent association, its incremental value beyond GLS is limited, establishing GLS as the primary echocardiographic parameter for non-invasive stratification of functionally significant coronary artery disease.