Analytical and Biological Variability of Urinary Epidermal Growth Factor-to-Creatinine Ratio in Patients with Chronic Kidney Disease and in Healthy Volunteers.

Background: Urinary levels of epidermal growth factor to creatinine ratio are considered as an early predictor of interstitial kidney fibrosis but so far no data are available on their biological variation (BV) and derived parameters. The aim of this study is to determine the BV of epidermal growth factor to creatinine ratio in patients with chronic kidney disease and in healthy volunteers.

Methods: This cross-sectional study included 150 patients with chronic kidney disease (CKD) and 150 healthy volunteers (HV). In both groups, spot second-morning urine samples were collected once a week for 4 consecutive weeks. Measurements of EGF were done by ELISA and expressed as EGF/creatinine ratio. The components of BV, individuality index (II), and reference change values (RCV) were calculated.

Results: The analytical coefficient of variation (CVa) of EGF/creatinine ratio was 3.8% in patients with CKD and 3.9% in HV. The within-patient variation coefficient (CVw) was 11.2% in CKD and 12.1% in HV. The between-patient coefficient of variation (CVg) was 34% in CKD and 22% in HV. In both groups, CVa met the optimum analytical quality specifications for imprecision, since it was lower than 25% of CVw. There were no significant differences between CKD and HV in the analytical or in the within-patient variances of the EGF/creatinine ratio. The between-patient EGF/creatinine ratio variance was significantly higher in patients with CKD than in HV (F: 48.3, p: 0.000). The EGF/creatinine ratio showed an individuality index (II) of 0.3 in CKD and 0.5 in HV. The reference change value (RCV) was 29.2% in CKD and 31.6% in HV.

Conclusions: The CVa associated with the measurement techniques used in our study meets the optimal criteria of analytical imprecision. The urine EGF/creatinine ratio shows a high index of individuality both in patients with CKD and in HV, so the comparison of an isolated value with a reference interval is of little use. In the monitoring of repeated levels in the same individual or patient, the changes can only be considered significant when they are greater than 30% in relation to the previous values.