111 - Serum Cystatin C is an Early Predictor of Acute Renal Injury in Infants after Cardiac Surgery with Cardiopulmonary Bypass

Poster Number: 111
Publication Number: 111.125

Maher Abadeer, Golisano Children's Hospital at The University of Rochester Medical Center, Rochester, NY, United States; Susan D. Martin, Golisano Children's Hospital at The University of Rochester Medical Center, Rochester, NY, United States; Michael F. Swartz, University of Rochester School of Medicine and Dentistry, Rochester, NY, United States; Alison L. Kent, University of Rochester School of Medicine and Dentistry, Rochester, NY, United States; George J. Schwartz, University of Rochester School of Medicine and Dentistry, Rochester, NY, United States; Patrick D. Brophy, Golisano Children's Hospital at The University of Rochester Medical Center, Pittsford, NY, United States; George M. Alfieris, University of Rochester School of Medicine and Dentistry, Rochester, NY, United States; Jill M. Cholette, Golisano Children's Hospital at The University of Rochester Medical Center, Rochester, NY, United States

Background: Acute kidney injury (AKI) following infant cardiac surgery is common, with increased morbidity and mortality.  Changes in urine output and/or serum creatinine (SCr) are unreliable in the detection of AKI as defined by the Kidney Disease Improving Global Outcomes (KDIGO) guidelines, and lag behind onset of renal injury.  Serum Cystatin C (SCC) has been used as a biomarker of AKI in adults and older children. If SCC predicts AKI in infants following cardiopulmonary bypass (CPB) is unknown.

Objective: The primary aim was to determine the diagnostic accuracy of SCC as an early biomarker of AKI following infant cardiac surgery requiring CPB.  The secondary aim was to compare inpatient morbidity in infants with AKI and those without.

Design/Methods: 43 infants ( < 1 year of age) were prospectively enrolled when presenting for cardiac surgery requiring CPB.  SCC and SCr were measured at baseline (pre-operatively), and at 12, 24, and 48 hours following CPB initiation. The difference in post-operative SCC and SCr values from baseline (∆SCC and ∆SCr, respectively) were calculated (e.g. ∆SCC12hr = SCC12hr - SCCbaseline). AKI in the first 3 post-operative days was defined using KDIGO guidelines, as an increase in SCr ≥ 50% and/or a urine output < 0.5mL/kg/hr for ≥ 6-12 consecutive hours. Subjects were divided in 2 groups:  AKI and non-AKI.  Receiver Operating Characteristic (ROC) curves were utilized to predict AKI based on ∆SCC and ∆SCr, and the area under the curve (AUC) was quantified.

Results: AKI occurred in 17 (39.5%) subjects.  Baseline SCC and SCr measurements were similar between groups, as were age, gender, weight, and CPB time.  Post-operative SCC values decreased from baseline, suggesting volume overload.  ∆SCC measurements were significantly different in those with AKI at 12 hours (0.2 ± 0.3 vs 0.4 ± 0.2; p = 0.01) and remained different at 24 hours (0.1 ± 0.3 vs 0.3 ± 0.2; p = 0.03) and 48 hours (0.2 ± 0.3 vs 0.3 ± 0.2; p = 0.01).  ∆SCr measurements were significantly greater in the AKI group at 24 hours (0.1 ± 0.1 vs 0.0 ± 0.1; p < 0.01) and 48 hours (0.1 ± 0.1 vs. 0.0 ± 0.1; p < 0.01) but not at 12 hours.  ROC curves demonstrated that both the ∆SCC at 12 hours (AUC = 0.72; p = 0.02) and the ∆SCr at 24 hours (AUC= 0.78; p < 0.01) predicted AKI.  Conclusion(s): The 12 hour ∆SCC predicts AKI in infants after cardiac surgery, and correlated with a rise in ∆SCr the following day.  Earlier detection of AKI would allow earlier post-operative intervention to mitigate renal injury in infants.   
Table 1Demographics and post-operative secondary outcomes in the AKI and non-AKI groups. Urine output was significantly lower in the AKI group for post-operative days 0-2.
Figure 1∆SCr measurements were significantly greater in the AKI group at 24 and 48 hours, but not at 12 hours. ∆SCC measurements were significantly different in those with AKI at 12 hours and remained different at 24 and 48 hours.