Looking beyond standard assessments: exploring the diagnostic potential of the MG/AVA ratio in LFLG AS

Introduction

Low-flow, low-gradient (LFLG) aortic stenosis (AS) is characterized by discordant grading on resting echocardiography. Recent evidence showed that mean gradient (MG)/aortic valve area (AVA) ratio can be useful in LFLG AS to confirm AS severity.

Purpose

Evaluate the diagnostic accuracy and prognostic significance of the MG/AVA ratio in our patient population with LFLG AS.

Methods

Multicenter, retrospective study. We included consecutive patients (pts) admitted for LFLG AS between 2021 and 2023. The severity of AS was determined using DSE or AVC. The optimal threshold of the MG/AVA ratio for diagnosing severe aortic stenosis was determined using Youden’s index. Prognosis included all-cause mortality and heart failure (HF) hospitalizations.

Results

A total of 91 pts were included. Baseline characteristics are summarized in Table 1. Based on echocardiography data, 65% of pts were classified as classical LFLG-AS (cLFLG-AS) and 35% as paradoxical LFLG-AS (pLFLG-AS). DSE or AVC data were available in 55% of pts, with 46% classified as having true severe AS and 54% as non-severe.

Pts with higher MG/AVA ratio (47.4 vs. 31.3, p=0.005) were more likely to have true severe AS. ROC analysis demonstrated that the ratio had moderate diagnostic accuracy (AUC 0.769). The optimal threshold for diagnosing true severe AS was 34.5 (specificity 75%, sensitivity 74%). Regarding prognosis, the optimal threshold ratio did not significantly predict all-cause mortality (p=0.466) or HF hospitalizations (p=0.278).

In a subgroup analysis of pts with cLFLG-AS, the median MG/AVA ratio was 33.9 (IQR 22.9). Pts with higher ratios (44.8 vs. 30.9, p=0.036) were more likely to have true severe AS. ROC analysis indicated moderate diagnostic accuracy (AUC 0.735), with an optimal threshold of 29.4 (specificity 67%, sensitivity 78%). Similar to the overall cohort, the ratio was not associated with mortality (p=0.712) or HF hospitalization (p=0.311).

Regarding the more specific thresholds, a ratio of 50.3 (specificity 92%, sensitivity 37%) for LFLG AS and a ratio of 47.2 (specificity 89%, sensitivity 32%) for classical LFLG AS do not demonstrate any significant prognostic impact.

Conclusion

The MG/AVA ratio demonstrated moderate diagnostic accuracy in differentiating true severe AS in pts with LFLG AS. However, this parameter was not significantly associated with prognosis regarding all-cause mortality or HF hospitalizations.For image description, please refer to the figure legend and surrounding text.