ScholarWorks@성균관대학교

초록

Background: Doppler-based diastolic dysfunction assessment requires optimal apical views and specialized expertise, and up to 30% of evaluations yield indeterminate results. We investigated whether a parasternal long-axis (PLAX)–only M-mode scoring system derived from mitral valve anterior leaflet (MVAL) motion could provide clinically useful risk stratification for moderate-to-severe left ventricular diastolic dysfunction (LVDD). Methods: This retrospective study analyzed echocardiographic data from 253 patients. Novel M-mode parameters (EPSS, APSS, EPOL, APOL, and their ratios) were compared with conventional Doppler indices across LVDD severity grades (normal, grade 1, grade 2, grade 3). A logistic regression-based scoring system combining five M-mode parameters was developed and internally evaluated using five-fold cross-validation with bootstrap confidence intervals. Results: The EPOL/APOL ratio showed moderate correlations with the E/A ratio in both normal-to-mild (r = 0.353, p < 0.001) and moderate-to-severe (r = 0.397, p = 0.001) LVDD groups. The logistic regression–derived scoring system achieved an AUC of 0.754 (95% CI: 0.672–0.825), with 55.7% sensitivity and 85.9% specificity at the optimal cutoff of 63.5. Decision curve analysis demonstrated positive net benefit over default strategies across the clinically plausible threshold probability range of 0.15–0.35. Conclusion: A PLAX-only M-mode scoring system provides acceptable discrimination and positive net clinical benefit for risk stratification of moderate-to-severe LVDD. External validation in independent populations is required before clinical implementation.

키워드