[Value of creatinine clearance rate estimated based on serum cystatin C in patients with acute kidney injury].

Objective: To investigate diagnostic value of creatinine clearance rate (CCr) based on serum cystatin C (SCys C) in acute kidney injury (AKI), and whether it could predict the need for renal replacement therapy (RRT).

Methods: The patients enrolled with the length of intensive care unit (ICU) stay over 3 days were collected from August 2010 to May 2011. According to the diagnosis of AKI during the ICU stay, patients were divided into the AKI group (n=21) and non-AKI group (n=30). After patients were admitted, the level of SCys C and creatinine (SCr) were measured so as to count CCr based on SCys C (SCys C-CCr) or on SCr (SCr-CCr) respectively, meanwhile urine volume and acute physiology and chronic health evaluation II (APACHE II) score were monitored. The value of CCr counted by SCys C and SCr on predict AKI and the correlations between RRT were compared.

Results: SCr-CCr and SCys C-CCr in AKI group both were significantly lower than non-AKI group all the way through on admission, and 2 days and 1 day before AKI diagnosed and the day AKI diagnosed. The level of SCys C-CCr on 2 days prior to AKI diagnosed was significantly lower than the day admitted (70.6±8.4 ml×min(-1)×1.73 m(-2) vs. 114.8±15.8 ml×min(-1)×1.73 m(-2), P<0.01), whereas the level of SCr-CCr were not significantly changed (76.4±19.3 ml×min(-1)×1.73 m(-2) vs. 78.7±22.1 ml×min(-1)×1.73 m(-2), P>0.05). Receptor operative curve (ROC) analysis indicated that SCys C-CCr could predict AKI earlier than SCr-CCr, as the area under curve (AUC) of SCys C-CCr and SCr-CCr on 2 days prior to AKI diagnosed were 0.859 and 0.664, respectively, and the sensitivity were 90.5% and 47.6%, the specificity were 76.2% and 81.0%. In AKI group 6 patients were treated with RRT, the AKI patients receiving RRT had significantly higher APACHE II score on admission (29.6±4.5 vs. 17.0±5.6, P<0.05) and less urine volume within 24 hours (740±465 ml vs. 1780±1230 ml, P<0.05) than patients not received RRT, however, SCys C-CCr has no significant difference between the sub-group (50.4±11.2 ml×min(-1)×1.73 m(-2) vs. 53.0±8.4 ml×min(-1)×1.73 m(-2), P>0.05). SCys C-CCr did not predict the need of RRT on the day to diagnose AKI (AUC=0.65).

Conclusions: The sensitivity of SCys C-CCr were high, but its specificity not. The SCys C-CCr may be helpful for excluding diagnose of AKI in high risk patients. However, it could not predict the need for renal replacement therapy on the day AKI diagnosed.