We read with great interest the prospective cohort study by Gimena-Rodríguez et al. [1] evaluating the Venous Excess Ultrasound (VExUS) protocol in patients with acute pulmonary embolism (PE). We commend the authors for exploring noninvasive hemodynamic tools in this population. However, we would like to raise important conceptual and methodological considerations regarding the interpretation of venous congestion in this setting.

First, defining VExUS as a marker of systemic congestion or associating it with volume overload in PE represents a pathophysiological pitfall. As emphasized by Guinot [2] and Rola et al. [3], VExUS is not a direct measure of intravascular volume. Instead, it serves as an alarm signal for cardiac-driven congestion, reflecting the complex interplay between venous return and right ventricular (RV) function. In acute PE, the abrupt increase in pulmonary vascular resistance leads to extreme pressure overload. The abnormal hepatic and portal venous flows reported by the authors are retrograde manifestations of this afterload mismatch and ensuing RV failure, rather than isolated hypervolemia.

Second, the study evaluates the consequence, venous excess, without assessing the underlying mechanism: right ventricular–pulmonary arterial (RV-PA) coupling. The authors did not report echocardiographic metrics of RV-PA coupling, such as the ratio of tricuspid annular plane systolic excursion to pulmonary artery systolic pressure (TAPSE/PASP). As validated by Tello et al. [4] against invasive pressure-volume loops, TAPSE/PASP is a robust noninvasive surrogate for the end-systolic to arterial elastance (Ees/Ea) ratio. Furthermore, Yuriditsky et al. [5] recently demonstrated that, in acute PE, a TAPSE/PASP ratio<0.34mm/mm Hg strongly discriminates RV-PA uncoupling and is a powerful independent predictor of reduced cardiac index and normotensive shock. Without integrating RV-PA coupling indices, the isolated interpretation of VExUS in acute PE remains physiologically incomplete.

Finally, methodological limitations restrict the generalizability of the authors’ findings. The high-risk cohort was underpowered, comprising only 6 patients (7.5%). Crucially, in 5 of the most severe cases, ultrasonography was performed after reperfusion therapies, with a median delay of 11h. Assessing the VExUS score hours after relief of pulmonary obstruction alters the baseline RV-PA coupling phenotype. Furthermore, the lack of hard clinical outcomes, such as acute kidney injury or mortality, and the violation of the proportional odds assumption in the primary statistical model limit the prognostic applicability of the scale.