Objective  To investigate the predictive value of serum lactic acid level at admission of Intensive Care Unit (ICU) for the occurrence of acute kidney injury (AKI) 12 hours after admission by a secondary analysis of the prospective FINNAKI study.

  Methods  A total of 409 sepsis patients without AKI at admission of ICU were selected as study objects from FINNAKI cohort data, and were divided into low level lactic acid group, medium level lactic acid group and high level lactic acid group by blood lactic acid quantile method. The incidence of AKI and the application of renal replacement therapy (RRT) were compared 12 h later in the three groups. Logistic regression analysis was used to investigate the predictive values of serum lactic acid level on the occurrence of AKI and the application of RRT. Receiver operating characteristic (ROC) curve was drawn and the area under the curve (AUC) was calculated.

  Results  The incidence rates of AKI were 21.21% in the low level lactic acid group, 33.58%in the medium level lactic acid group and 49.65% in the high level lactic acid group respectively (P < 0.001). After adjusting independent variables, multivariate regression analysis showed that risk of AKI increased by 23%for every 1 mmol/L increase of blood lactate in septic patients(P=0.001). The risk of AKI in the high level lactate group was 2.31 times higher than that in the low lactate group(P=0.010). ROC curve showed that the AUC serum lactic acid level at admission in predicting AKI in septic patients was 0.66 (95%CI, 0.60 to 0.72). The AUC for lactate predicting the use of RRT in septic AKI patients was 0.70 (95%CI, 0.60 to 0.80).

  Conclusion  Lactate at ICU admission can predict AKI and the use of RRT in septic patients.