Prospective study for prognostic value of serum cystatin C in acute renal failure

AIM: To study the level of serum cystatin C (Cys C) and its prognostic value in acute renal failure (ARF). METHODS: 215 critically ill patients were studied prospectively, and each blood sample was determined daily. Serum creatinine (Scr) was measured by enzymic method. Cys C was detected by particle-enhanced turbidimetric immunoassay (PETIA) and glomerular filtration rate (GFR) was estimated by Cockroft-Gault equation. ARF was diagnosed according to ADQI (RIFLE) criteria. RESULTS: 41 patients developed ARF to some degree, among which 7, 13 and 21 patients fulfilled R-, I-, and FLE-criteria, respectively. 21 were death and 20 survival in total 41 ARF patients. 174 patients without ARF served as controls. In patients with ARF, Cys C was significantly increased than that in patients who did not develop ARF (P<0.01). In different RIFLE criteria of ARF, serum Cys C was linearly correlated with serum creatinine (P<0.01), also ［Cys C］-1 with estimated GFR (P<0.01). When ARF was diagnosed according to serum creatinine increased ≥150%, serum Cys C was demonstrated a highly diagnostic value to detect ARF as indicated by area under the curve (AUC) of the ROC curve of 0.983 (95% confidence interval, 0.960-1.006) (P<0.01). Logistic regression showed serum Cys C and Scr were not independently risk factors for the mortality of ARF (P>0.05). Also ROC analysis demonstrated serum Cys C and Scr had no predicting value of the fatality and survival in ARF (P>0.05). CONCLUSION: Serum Cys C is significantly increased in ARF and correlated well with the severity of ARF. Serum Cys C can be one of the detectable markers of ARF, but it is independent of the mortality and does not predict the prognosis of these patients.