We aimed to explore the prognostic value of serum cystatin C (CysC) levels on kidney disease outcome in type 2 diabetes mellitus (T2DM) patients with chronic kidney disease (CKD).

The clinical data and pathological examination results of 113 T2DM patients with CKD, who were hospitalized in the First Affiliated Hospital of Nanjing Medical University from January 2011 to July 2020, were retrospectively analyzed in this study. Clinicopathological features and renal outcomes were compared between patients with CysC>1.54 mg/L (n=57) and CysC≤1.54 mg/L (n=56) at the time of renal biopsy. Cox regression analysis was used to analyze the risk factors of poor renal prognosis. The relationship between serum CysC level and renal prognosis was analyzed by smoothing curve fitting and threshold effect. Kaplan-Meier survival curve was used to compare and analyze the difference of renal survival rate. Further, the receiver operator characteristic curve was used to evaluate the predictive value of serum CysC combined with renal tubular marker blood and urinary neutrophil gelatinase-associated lipocalin (NGAL) on renal prognosis in all enrolled patients and those with different kidney disease stages. Besides, the ability of serum CysC level to predict renal prognosis within 3 years was evaluated by time-dependent area under the curve (AUC).

Compared with patients with serum CysC levels≤1.54 mg/L, patients with CysC>1.54 mg/L had more deteriorated renal function, decreased levels of hemoglobin and serum 25(OH) vitamin D, but more severe interstitial inflammation, higher glomerular sclerosis ratio and severe vascular lesion (all P<0.05). During 36.77 (29.34, 44.20) months follow-up, the composite renal outcomes were noted in 37.2% patients. Kaplan-Meier survival curve showed that the cumulative survival rates of patients without renal end points was significantly lower in CysC level>1.54 mg/L group than in CysC≤1.54 mg/L group (χ²=5.752, P=0.016). Adjusted multivariate Cox analysis showed that serum CysC level (HR=7.850, 95%CI 1.248-49.382, P<0.05) was an independent risk factor for renal prognosis. Smoothing curve fitting analysis showed that there was a linear relationship between serum CysC level and relative risk of renal endpoint event (β=2.25, 95%CI 1.06-4.81, P=0.036). The time-dependent receiver operator characteristic curve showed that the AUC of serum CysC in predicting the poor renal prognosis of T2DM patients within 3 years after renal biopsy were 0.714, 0.625 and 0.631, respectively. The AUC of serum CysC combined with blood and urinary NGAL was 0.694 (sensitivity 55.56%, specificity 77.78%). In the population with eGFR less than 60 ml·min^-1·1.73m^-2 (n=51), the AUC was 0.817 (sensitivity 66.67%, specificity 85.00%).

Higher serum CysC level is associated with deteriorated renal function, more severe renal pathological lesions and increased risk of worse renal prognosis in T2DM patients. Serum CysC level presents better predictive value for the renal prognosis of T2DM patients within 1 year after renal biopsy. Combined with renal tubular marker blood and urinary NGAL, serum CysC level might serve as a potential tool for identifying cases with high-risk of unsatisfactory renal prognosis, especially in those with eGFR less than 60 ml·min^-1·1.73m^-2.

Diabetes Mellitus, type 2; Chronic kidney disease; Cystatin C; Prognosis