Abstract Objectives Septic shock remains a major cause of ICU mortality, frequently associated with multi-organ dysfunction (J Crit Care 43:122-127, 2018). Cardiovascular involvement, including myocardial dysfunction, is common and significantly increases mortality rates to 70–90% (Br J Anaesth 119(4):583–594, 2017, J Intensive Care 4:22, 2016). While LV systolic dysfunction is well documented, its prognostic value in septic shock remains debated (Br J Anaesth 104(1):3–11, 2010, J Crit Care 27(3):319-e9, 2012), with inconsistent links between ejection fraction (EF) and outcomes (J Crit Care 27(3):319-e9, 2012, Ann Intern Med 100(4):483–90, 1984, Crit Care Med 36(6):1701–6, 2008). Recent studies suggest that three-dimensional (3D) echocardiography and speckle-tracking echocardiography (STE) may offer improved assessment over conventional two-dimensional (2D) echocardiography. This study aimed to compare the prognostic value of ventricular function assessed by 3D versus 2D echocardiography in sepsis and septic shock patients. Methods Fifty patients with sepsis or septic shock, as defined by the Surviving Sepsis Campaign, underwent echocardiographic assessment within 24 h of ICU admission. Left ventricular function was evaluated using 2D echocardiography with Simpson’s method, speckle-tracking echocardiography (2D-GLS), and 3D echocardiography (3D-EF and 3D-GLS). Results Patients were categorized into survival (36%) and non-survival (64%) groups. Cox regression analysis identified 2D-GLS as an independent predictor of mortality. GLS proved the strongest prognostic marker, outperforming EF in both 2D and 3D analyses. The Youden indices for 2D-GLS, 3D-GLS, 2D-EF, and 3D-EF were 0.87, 0.68, 0.75, and 0.59, respectively. Additionally, 2D-derived GLS and EF demonstrated better prognostic performance than their 3D equivalents. Conclusion 2D-GLS was independently associated with higher mortality risk in sepsis and septic shock and provided superior prognostic value compared to 3D-GLS. GLS outperformed EF as the strongest mortality predictor in both 2D and 3D modalities.
Mohamed et al. (Fri,) studied this question.