Background: 

Cardiac surgery-associated acute kidney injury (CSA-AKI) is a common complication and is associated with the poorest outcomes. Therefore, early prediction of CSA-AKI remains a major issue. Severity scores such as the STS score could estimate the risk of AKI preoperatively. The main objective of this study was to evaluate the risk factors of on-pump CSA-AKI and to assess the performance of the STS score in order to predict CSA-AKI.

Patients: 

We identified 252 patients with on-pump cardiac surgery, and the STS score was defined retrospectively.

Results: 

AKI occurred in 14.6% (n = 37/252) of patients and renal replacement therapy was required in 21.6% of AKI (n = 8/37). CSA-AKI was associated with 35.1% in-hospital mortality (vs. 1.4%) and nearly doubled length of stay (14.5 vs. 8.0 d). The risk of CSA-AKI was mainly determined by preoperative morbidities such as chronic kidney disease, peripheral vascular disease, and severe congestive heart failure. Long cardio-pulmonary bypass time was also a determinant. CSA-AKI + patients exhibited higher STS renal risk (5.6% vs. 2.0%; p < 0.0001), resulting in a good discrimination between AKI + and AKI − patients (area under curve [AUC] 0.80). Interestingly, a basal renal function ≤55 ml/min/1.73m² was as good as the STS score to predict CSA-AKI (AUC 0.75; P 0.26).

Conclusions: 

On-pump CSA-AKI was observed in nearly 15% of cases and was associated with poorer outcomes. Interestingly, the risk of CSA-AKI could be estimated preoperatively, thanks to the basal renal function, which exhibited an equal performance to the STS score.