Background: Despite the increased prenatal diagnosis of congenital abnormalities of the kidney and urinary tract (CAKUT), no reliable renal marker for glomerular filtration rate (GFR) has been validated yet in neonates. Cystatin C (CysC) is specific to the neonate and is proposed as a sensitive marker for this population. The aims of the study were first to define a reference interval in our center of CysC at birth in normal term babies and assess CysC as a marker of GFR in a group of term neonates prenatally diagnosed with CAKUT compared to controls.
Methods: One hundred normal term neonates (control group) and 33 neonates with kidney malformation (KM) had the CysC levels in their cord blood measured. A reference interval for CysC in controls was calculated using non-parametric methods. CysC from controls was compared first to the whole group of neonates with KM, then with KM group divided in infants (n = 20) with unilateral kidney malformation (UKM) and those (n = 13) with bilateral kidney malformation (BKM). A multivariable analysis was performed to assess the difference in CysC between the groups with adjustment on other factors. The ability of CysC to discriminate neonates with BKM from the controls was assessed by a non-parametric receiver-operated characteristics (ROC) curve.
Results: In the control group, the CysC reference interval was [1.54-2.64] mg/L with a median (M) CysC of 2.02 IQR [1.86-2.23]. In the neonates with KM, M CysC was 1.98 IQR [1.79-2.34]; 1.88 IQR [1.76-2.01] in the UKM group and 2.52 IQR [2.16-2.71] in BKM group. Using a multivariate regression analyses, CysC was significantly increased (P < 0.001) in BKM compared to controls with an increment of CysC of 24.5%, and independent from gender, weight and size. The ROC curve analyses, comparing BKM versus controls with a chosen cut-off for CysC of 2.34, showed a sensitivity of 69% and a specificity of 86%.
Conclusions: Comparing CysC with a reference interval of CysC validated in our center, we showed a significant increase of CysC in neonates presenting BKM compared to controls and those with UKM.