CALL-K score: predicting the need for renal replacement therapy in cardiogenic shock

Aims The clinical predictors and outcomes of patients with cardiogenic shock (CS) requiring renal replacement therapy (RRT) have not been studied previously. This study assesses the impact of RRT on mortality in patients with CS and aims to identify clinical factors that contribute to the need of RRT. Methods and results Consecutive patients presenting with CS were included from a prospective registry of cardiac intensive care unit admissions at a single institution between 2014 and 2020. Of the 1030 patients admitted with CS, 123 (11.9%) received RRT. RRT was associated with higher 1-year mortality [adjusted hazard ratio = 1.62, 95% confidence interval (CI) 1.02-2.14], and a higher in-hospital incidence of sepsis [risk ratio = 2.76, P < 0.001], and pneumonia (risk ratio = 2.9, P = 0.001). Those who received RRT were less likely to receive guideline-directed medical treatment at time of discharge, undergo heart transplantation (2.4% vs. 11.5%, P = 0.002) or receive a durable left ventricular assist device (0.0% vs. 11.6%, P < 0.001). Five variables at admission best predicted the need for RRT (age, lactate, haemoglobin, use of pre-admission loop diuretics, and admission estimated glomerular filtration rate) and were used to generate the CALL-K 9-point risk score, with better discrimination than creatinine alone (P = 0.008). The score was internally validated (area under the curve = 0.815, 95% CI 0.739-0.835) with good calibration (Hosmer-Lemeshow P = 0.827). Conclusions RRT is associated with worse outcomes, including a lower likelihood to receive advanced heart failure therapies in patients with CS. A risk score comprising five variables routinely collected at admission can accurately estimate the risk of needing RRT.