Non-linear relationship between serum phosphate and 30-day mortality in critically ill patients with cardiogenic shock: a cohort study from MIMIC-IV

ObjectiveCardiogenic shock (CS) carries high mortality despite management advances. Elevated phosphate links to poor outcomes in critical illness, but its role in CS remains unclear. This study explored the association between serum phosphate and 30-day mortality in CS patients. MethodsA retrospective cohort study using the MIMIC-IV database stratified 2308 CS patients by serum phosphate tertiles (T1: <3.9 mg/dL, T2: 3.9-5.3 mg/dL, T3: [&ge;]5.3 mg/dL) measured [&le;]24h post-ICU admission. Cox regression analyzed associations between phosphate levels and 30-day (primary outcome) and in-hospital mortality (secondary outcome). Kaplan-Meier curves and restricted cubic spline analysis were conducted to evaluate survival disparities and non-linear relationships. ResultsThe 30-day and in-hospital mortality rates were 39.3% and 35.8%, respectively. Phosphate elevation independently correlated with heightened mortality after multivariable adjustment (adjusted hazard ratios [HR] 1.08, 95% CI 1.03-1.13, p = 0.001 for both 30-day and in-hospital mortality). Mortality risk escalated across phosphate tertiles (T2 vs. T1: HR 1.43, 95% CI 1.16-1.75, p = 0.001; T3 vs. T1: HR 1.70, 95% CI 1.34-2.16, p < 0.001). A non-linear inflection point occurred at 5.52 mg/dL: each 1 mg/dL increase below this threshold raised 30-day mortality risk by 17% (adjusted HR 1.17, 95% CI 1.04-1.32, p = 0.0087), with risk plateauing above it. Kaplan-Meier curves confirmed significant survival differences (p < 0.001). ConclusionElevated serum phosphate levels are independently associated with increased both 30-day and in-hospital mortality in CS patients, with a nonlinear inflection point at 5.5 mg/dL. Serum phosphate may serve as a potential biomarker for risk assessment and therapeutic guidance in CS. Prospective validation is warranted.