Renal replacement therapy (OR 4.30; 95% CI 2.07-8.90) and age ≥75 years (OR 1.68; 95% CI 1.17-2.38) independently predicted in-hospital mortality in patients with cardiogenic shock.
Cohort (n=762)
No
Do age ≥75 years and the use of renal replacement therapy independently predict in-hospital mortality and lower LVEF in patients with cardiogenic shock?
In patients with cardiogenic shock, age ≥75 years and the need for renal replacement therapy are independent predictors of in-hospital mortality, defining distinct high-risk clinical phenotypes.
Effect estimate: OR 4.30 (95% CI 2.07-8.90)
p-value: p=<0.001
Abstract Introduction Cardiogenic shock (CS) remains associated with high mortality despite therapeutic advances. Age and severe renal dysfunction requiring renal replacement therapy (RRT) may influence clinical presentation, management, and prognosis. Evaluating their individual and combined effects may improve risk stratification. Objectives To assess the impact of age and the use of RRT on left ventricular ejection fraction (LVEF) and in-hospital mortality in CS patients, and to characterize clinical and therapeutic differences between phenotypes. Methods We retrospectively analyzed 762 consecutive patients diagnosed with CS and admitted between 2011 and 2024 to a cardiovascular intensive care unit. Patients were stratified using a 2×2 factorial design according to age (75 or ≥75 years) and use of RRT, generating four phenotypes: (1) 75 without RRT, (2) 75 with RRT, (3) ≥75 without RRT, and (4) ≥75 with RRT. Demographic, analytical, and therapeutic variables were compared using ANOVA and Chi-square tests. Two-way ANOVA and logistic regression assessed main and interaction effects on LVEF and mortality. Results Mean age was 70 ± 11 years and overall mortality was 32%. LVEF and creatinine differed significantly among phenotypes (p = 0.0047 and p 0.001). LVEF was lowest in patients requiring RRT (28.6 ± 14.3% and 28.9 ± 15.8% in groups 2 and 4) and higher in ≥75 without RRT (34.7 ± 14.5% in group 3). RRT was associated with higher creatinine levels (5.9 ± 1.4 and 3.54 ± 3.0 vs 1.35 ± 1.03 and 1.49 ± 0.75 mg/dL in groups 1 and 3), increased use of vasopressin (p 0.001), invasive ventilation (p 0.001), and intra-aortic balloon pump (p = 0.001). STEMI/NSTEMI prevalence differed (p = 0.015), being higher in ≥75 without RRT (68%). Mortality increased in patients requiring RRT (19.8% and 29.4% in groups 1 and 3 vs 51.5% and 66.7% in groups 2 and 4; p 0.001). In the logistic model, age ≥75 years (OR 1.68, 95% CI 1.17–2.38, p = 0.004) and RRT (OR 4.30, 95% CI 2.07–8.90, p 0.001) independently predicted mortality, without significant interaction (OR 1.09, 95% CI 0.45–2.63, p = 0.84). In the two-way ANOVA, age (p = 0.0037) and RRT (p = 0.041) were associated with lower LVEF, without interaction (p = 0.43). Conclusions Age and RRT define distinct clinical phenotypes in CS with specific biochemical, therapeutic, and prognostic characteristics. RRT identifies patients with severe cardiorenal dysfunction and higher support requirements, with markedly increased mortality. Age adds adverse prognostic impact but does not modify the effect of RRT. Integrating both variables into shock phenotyping may improve risk assessment and individualized management.Table1:baseline clinical characteristics figure1:factors associated mortality
Guia et al. (2026) conducted a cohort in Cardiogenic shock (n=762). Renal replacement therapy (RRT) and age ≥75 years vs. No RRT and age <75 years was evaluated on In-hospital mortality (OR 4.30, 95% CI 2.07-8.90, p=<0.001). Renal replacement therapy (OR 4.30; 95% CI 2.07-8.90) and age ≥75 years (OR 1.68; 95% CI 1.17-2.38) independently predicted in-hospital mortality in patients with cardiogenic shock.