BACKGROUND

Cardiac surgery-associated acute kidney injury (CSA-AKI) is prevalent and increasingly reported. Its diagnosis traditionally follows the Kidney Disease: Improving Global Outcomes (KDIGO) AKI criteria. However, little evidence supports its appropriateness for cardiac surgery patients, particularly regarding the subtle serum creatinine change (ΔSCr) that defines mild AKI.

OBJECTIVES

The purpose of the study was to investigate the ΔSCr threshold specific to CSA-AKI and compare its impact on CSA-AKI incidence and prognosis with the KDIGO AKI criteria threshold.

METHODS

A 10-year coronary surgery cohort with serial perioperative SCr measurements was retrospectively analyzed. The relationship and prognostically significant threshold of 48-hour absolute ΔSCr with 30-day mortality were explored using multivariate restricted cubic spline analysis and receiver-operating characteristic curve analysis. AKI incidence and prognostic value were compared between adopting the KDIGO or new thresholds.

RESULTS

Among 37,706 patients, 20,290 (53.8%) developed KDIGO-defined AKI. For stage-1 AKI (18,835, 49.9%), the majority (75.2%) were solely attributed to the KDIGO absolute criterion (48-hour ΔSCr ≥0.3 mg/dL). The 48-hour ΔSCr threshold associated with an adjusted odds ratio ≥1.00 for 30-day mortality was 0.549 mg/dL. A similar threshold (0.553 mg/dL) was also identified based on the Youden index cutoff. Applying the 0.55 mg/dL threshold to define stage 1 CSA-AKI, the overall and stage-1 CSA-AKI incidence decreased to 21.0% and 17.2%, with 27.7% of the stage-1 CSA-AKI solely attributed to the new criterion. The prognostic value for AKI defined by this new threshold was significantly higher than the KDIGO criteria.

CONCLUSIONS

A cardiac surgery-specific ΔSCr threshold in defining AKI was notably higher compared with the current general AKI definition.