Beyond the Baseline Creatinine: A Comparison of Dynamic versus Static Renal Function Metrics and their Association with Mortality in Cardiogenic Shock

BACKGROUND Acute kidney injury (AKI) is strongly associated with death in cardiogenic shock (CS), yet AKI classification at CS onset is challenged by non–steady-state physiology and frequent absence of baseline creatinine. We compared the prognostic performance of dynamic and static creatinine-based renal function metrics measurable early after CS onset. METHODS We analyzed CS admissions from 2015–2022 at a single referral center. Five renal metrics were evaluated: serum creatinine, estimated glomerular filtration rate (eGFR), percent change in creatinine from an assumed baseline eGFR of 75 mL/min/1.73 m², 24-hour percent change in creatinine, and 24-hour kinetic eGFR. Each metric was tested for independent association with inpatient mortality and incremental contribution to model discrimination and calibration beyond baseline covariates. RESULTS Among 976 patients, 73.6% had non-AMI CS. At CS onset, 59.2% had eGFR <60 mL/min/1.73m²; these patients were older (62 vs 54 years), more frequently had non-AMI CS (76.1% vs. 69.8%), and had higher inpatient mortality (29.6% vs. 21.6%). All renal metrics were independently associated with mortality. Kinetic eGFR demonstrated the strongest standardized association (OR per SD 1.41, 95% CI 1.19–1.67; p<0.001) and greatest improvement in model fit (LR χ²=16.23; p<0.001). Cubic spline analysis identified a nonlinear relationship (p<0.001), with mortality increasing below kinetic eGFR<60 and plateauing near 30 mL/min/1.73m². Incorporating the splined kinetic eGFR improved model discrimination (AUC 0.716→0.741). CONCLUSIONS The kinetic eGFR is a more physiologically concordant measure of renal function in early CS that outperforms static measures of renal function even when pre-hospital creatinine is unknown.