Optimum methodology for estimating baseline serum creatinine for the acute kidney injury classification.

Aim This study aimed to investigate how varied methods of determining baseline serum creatinine ( SCr) would affect acute kidney injury ( AKI) diagnosis and prediction of 60 day mortality in critically ill patients following an episode of AKI. Methods This is a single-centre retrospective study conducted at a tertiary referral hospital. All adult intensive care unit ( ICU) patients between January and December 2011, who had at least one SCr values measured between 7 days and 180 days before hospital admission and during ICU stay, were analyzed. The baseline SCr was calculated using either the most recent ( SCr_{most recent}) or the minimum (SCr_min) value of SCr measurement over the specified assessment period before hospital admission. AKI was defined based on KDIGO SCr definition. The primary outcome was 60 day mortality after ICU admission. Results A total of 4020 patients were included in the analysis. AKI was detected in 1204 (30.0%) using the SCr_min and 945 (23.5%) using the SCr_{most recent} ( P < 0.001). Compared with patients without AKI regardless of baseline SCr methodology, the 60 day mortality risk of patients who developed AKI using the SCr_min and SCr_{most recent} was significantly increased (odds ratio ( OR) = 3.74; 95% confidence interval ( CI) 2.98-4.70). Similarly, the risk of 60 day mortality in patients who met AKI criteria using the SCr_min but not the SCr_{most recent} was significant higher than in patients without AKI ( OR = 2.04; 95% CI 1.36-3.00). Conclusion Using the minimum value of preadmission SCr as a baseline kidney function not only can detect more AKI cases, but also provides the better predictive ability for 60 day mortality. [ABSTRACT FROM AUTHOR]