Objective To investigate the predictive value of preprocedural cystatin C level for contrast-in-duced acute kidney injury (CI-AKI) and poor long-term outcome after cardiac catheterization. Methods One thou-sand one hundred and fifty-four patients underwent cardiac catheterization were enrolled in Guangdong general hos-pital. The level of serum cystatin C was determined at 24 hours pre-operation. A 2-year follow up was performed for each patient. Preprocedural cystatin C level was compared between patients with or without CI-AKI. The cystatin C quartiles were compared between patients with incidence of CI-AKI and patients with adverse in-hospital outcomes. Analyses of the receiver operating characteristic curves (ROC) were performed to evaluate the predictive value and cutoff level of cystatin C level for CI-AKI. The log-rank test and Cox regression analyses were also performed to in-vestigate the correlation between cystatin C level and poor long-term outcomes. Results CI-AKI occurred in 42 patients (3.6%). The cystatin C level was significantly higher in the CI-AKI group than that in the non-CI-AKI gu-oup (1.76 ± 1.05 vs 1.20 ± 0.50 mg/L, P=0.001). Patients with higher cystatin C level also had higher risk of CI-AKI and adverse in-hospital outcomes. ROC and Youden index showed that 1.3 mg/L cystatin C of was a fair dis-criminator for CI-AKI, but not significantly different from the Mehran CI-AKI score (AUC, 0.75 vs 0.76, P =0.874). After adjusting for other known CI-AKI risk factors, cystatin C level over 1.3 mg/L remained significantly associated with CI-AKI. During the long-term follow-up , the patients with cystatin C level over 1.3 mg/L were at a higher risk of all-cause mortality and MACEs (P < 0.001). Concusions A preprocedural cystatin C level over 1.3 mg/L was a good predictor of CI-AKI and poor long-term outcomes after cardiac catheterization.

Objective To evaluate the prognostic values of common definition compared to traditional definition of contrast-induced nephropathy (CIN) in patients with normal serum creatinine (SCr). Methods Patients undergoing percutaneous coronary angiology or intervention with normal baseline SCr were enrolled prospectively. Those who were diagnosed as CIN according to common definition were divided into two groups based on the peak increase from baseline in the SCr concentration within 48 ~ 72 hours after the procedure: ≥ 44.2 μmol/L (CIN44.2 group, in common with traditional definition), ≥25% of baseline to < 44.2 μmol/L (CIN25%-44.2 group, interval between the two definitions). Hospital stay and long-term outcomes were compared among CIN44.2, CIN25%-44.2, and non-CIN groups. Results Of all 3,044 patients enrolled, 302 (9.9%) patients developed CIN according to common definition including CIN44.2 occurred in 56 (1.8%) patients and CIN25%-44.2 in 246 (8.1%) patients. Patients in CIN44.2 group indicated significant longer hospital stay and long-term outcomes compared with non-CIN group (P < 0.05). However, patients in CIN25%-44.2 group had similar in-hospital mortality and long-term cumulative risk of major clinical adverse events (MACE) and death with non-CIN group (all, P = 1.00). Multivariate Cox proportional hazard analyses also demonstrated that CIN25%-44.2 did not associate with long-term MACE (HR 1.16, P = 0.645) and death (HR 0.98, P = 0.964) after adjusting for potential confounding factors. Conclusions For patients with normal baseline SCr, common definition based on traditional definition of CIN is unreasonable and overestimates the incidence of CIN, whose extension of traditional denifition proves no significant clinical value.