Predicting acute kidney injury in adults after cardiac surgery with a novel clinical metric: The cardiac diuretic-responsiveness index.

OBJECTIVE

Cardiac surgery-associated acute kidney injury (CSA-AKI) is a common complication after surgery, yet few risk models are used in practice. We evaluated whether a cardiac diuretic responsiveness index (CDRI) can accurately predict CSA-AKI.

METHODS

Time-series hemodynamic, intake/output volumes, medication, and laboratory data were extracted from electronic health records for 2016 patients who underwent cardiac surgery (2017-2022) and linked to the institutional Society of Thoracic Surgeons Baylor St Luke's Medical Center database. Patients with serum creatinine >4 mg/dL, dialysis before surgery, or those without diuretics administered were excluded. CDRI was defined as urine output in the first hour after initial diuretic administration adjusted for patient weight and diuretic dose (cc/kg/hr/mg furosemide). CSA-AKI was defined by Kidney Disease Improving Global Outcomes creatinine criteria. Segmental regression identified the optimal CDRI threshold.

RESULTS

Median time to first dose was 26.4 hours. CSA-AKI occurred in 16.2% (n = 326) of patients who were primarily stage 1 (82.8%, n = 270). Patients who developed CSA-AKI had significantly lower CDRI values (0.0611 vs 0.111 cc/kg/hr/mg furosemide; P < .0001). A CDRI threshold of 0.075 predicted CSA-AKI with a C-statistic 0.86), outperforming traditional oliguria criteria (<0.5 mL/kg/h, C-stat 0.8).

CONCLUSIONS

CDRI outperforms oliguria in predicting CSA-AKI, enabling real-time stratification, and may inform timely intervention to prevent AKI progression.