Current landscape for outcome prediction in patients with cardiogenic shock

Cardiogenic shock (CS) is a critical state of primary cardiac dysfunction characterized by tissue hypoperfusion and cellular hypoxia, with in-hospital mortality rates persistently between 30% and 50%. This review synthesizes contemporary evidence to redefine risk stratification through a dynamic, multi-modal lens, moving beyond static snapshots to prioritize longitudinal assessment aimed at anticipating clinical instability, therapeutic response, or physiological recovery. The evolution of CS staging is evaluated, specifically the Society for Cardiovascular Angiography and Interventions (SCAI) five-stage continuum, which tracks patient progression from "At risk" (Stage A) to "Extremis" (Stage E) and identifies critical windows for preemptive stabilization in Stage B. Essential hemodynamic "signposts" are examined, including cardiac index, right atrial pressure over pulmonary capillary wedge pressure ratio and pulmonary artery pulsatility index, which help identify high-risk phenotypes like the right-dominant CS. We highlight the role of echocardiography, lung ultrasonography and Venous Excess Ultrasound (VExUS) score as real-time biomarkers for monitoring forward flow and central and systemic congestion. Ultimately, accurate prognostication requires a multi-modal risk portrait integrating shock severity, biological phenotype, and patient-specific modifiers to identify non-responders early and allow for the timely treatment escalation, or proper de-escalation of intensive therapies as clinical improvement occurs. In this context, predictive approaches are primarily intended to anticipate imminent clinical deterioration or recovery, whereas prognostic models aim to estimate outcome probabilities, particularly mortality risk, once cardiogenic shock is established.