Abstract
Objective: The capability of urinary TIMP-2 (tissue inhibitor of metalloproteinase) and IGFBP7 (insulin-like growth factor binding protein)—NephroCheck Test (NC) = ([TIMP-2] x [IGFBP7]) / 1000)—to predict renal recovery from acute kidney injury (AKI) has been poorly studied. The aim of this study was to assess the performance of measurements of ([TIMP-2] x [IGFBP7]) / 1000) over 24 hours to differentiate transient from persistent AKI.Methods: Of 460 consecutive adult patients admitted to the ICU, 101 were prospectively studied: 56 men, 62 (52–71) years old. A fresh urine sample was collected at H0, H4, H12 and H24 to determine ([TIMP-2] x [IGFBP7]) / 1000) levels. Areas under the curves of Delta NC H4-Ho and H12-H4 and serum creatinine (sCr) for detection of AKI recovery were compared.Results: Forty-one (40.6%) patient were diagnosed with AKI: 27 transient and 14 persistent AKI. At admission (H0), AKI patients had a significantly higher NC score than patients without AKI (0.43 [0.07–2.06] vs 0.15 [0.07–0.35], p = 0.027). In AKI groups, transient AKI have a higher NC, at H0 and H4, than persistent AKI (0.87 [0.09–2.82] vs 0.13 [0.05–0.66] p = 0.035 and 0.13 [0.07–0.61] vs 0.05 [0.02–0.13] p = 0.013). Thereafter, NC level decreased in both AKI groups with a Delta NC score H4-H0 and H12-H4 significantly more important in transient AKI. Roc curves showed however that delta NC scores did not discriminate between transient and persistent AKI.Conclusion: In our population, absolute urinary levels of NC score were higher at early hours after ICU admission (H0 and H4) in transient AKI as compared to persistent AKI patients. NC variations (Delta NC scores) over the first 12 hours may indicate the AKI’s evolving nature with a more significant decrease in case of transient AKI but were not able to differentiate transient from persistent AKI.