Gestational and age-specific cystatin C reference intervals for newborns in China

Backgroud : To establish reference of serum cystatin C (CysC) in Chinese newborn infants. Methods: Serum CysC levels were measured in 1919 blood samples from 1044 newborns during their first 28 days of life. CysC levels were analyzed for association between subgroups dichotomized by postnatal age (PA) and gestational age (GA). Reference intervals of serum CysC were determined according to the PA and GA. The association between serum CysC level and other biochemical parameters as well as perinatal factors were also analyzed. Results: In this study, the mean GA was 35.75±2.90 weeks and birth weight was 2613.77±696.72 g. Reference ranges of serum CysC were determined and a general decreasing trend of CysC levels was observed as GA increased. CysC levels differed significantly among the PA and GA groups (P0.001). Serum CysC levels were relatively stable throughout GA, except being impacted by white blood cell count within postnatal 24 hours. Moreover, their levels always correlated positively with serum creatinine concentrations (P<0.001). Conclusion: The reference level of serum CysC should be determined according to PA and GA. In contrast to creatinine, serum CysC is a dependable index for assessing renal function in neonate, for there are rare factors influencing it.


Background
Newborns are a particularly vulnerable population, especially when born prematurely. Monitoring of renal function in neonates is important for determining the safe and accurate dosing of medication and monitoring the effect of various 4 pathologies on the kidney [1].
As we all know, renal function is typically measured by assessment of the glomerular filtration rate (GFR), and GFR is measured by reference procedures that use exogenous markers, including inulin, 51 Cr-EDTA and 99m Tc-DTPA, and so on.
However, direct assessment of neonatal renal function by GFR is challenging due to the impracticability of frequent blood sampling, urine collection, and constant infusion of exogenous markers [2]. At present, serum creatinine (Cr) is the most commonly measured proxy for renal function. However, variability of muscle mass, age, gender, diet, glomerular and tubular secretion precludes Cr to be a valuable tool for GFR assessment in children [3]. Moreover, Cr is influenced by maternal renal function for at least 72 hours after birth due to placental transfer, and the degree of immaturity [4].
Cystatin C (CysC), a member of family II of cystatins, is a nonglycosylated 13 kDa protein inhibitor of cysteine proteases. CysC was suggested as a marker of GFR, since it is produced at a constant rate by all nucleated body cells, freely filtered in the renal glomeruli and almost completely reabsorbed and catabolized in the proximal tubules [5,6]. In contrast with Cr, CysC does not cross the placenta [7, 8], {Plebani, 1997, Determination of blood cystatin C in pregnant women during labor and in their newborns} and it is independence of muscle mass and has a better diagnostic sensitivity [9][10][11]. Therefore, CysC has been evaluated as a promising alternative to Cr. However, there is not an acknowledged CysC reference standard in neonates because of few studies on CysC reference. We carry out the present study for attempting to identify reference values of CysC.

5
This observational study was conducted at 2 neonatal intensive care unit (NICU), the Union Hospital, Tongji Medical College (Wuhan, China) and Yuhuangding Hospital of Qingdao University (Yantai, China). Approval for the study was obtained from the Ethics Committee of Huazhong University of Science and Technology (NO.2016-S251). The study was carried out on consecutive babies, born to women followed antenatal at the two above hospitals, between August 2015 and July 2019. In order to obtain a reference standard representative for "healthy" newborn infants, the following exclusion criteria were applied: (1) any congenital anomaly, (2)

Results
This study included 1044 newborns and 1919 blood samples. The mean GA was 35.75±2.90 weeks. And the mean BW was 2613.77±696.72g. Demographic characteristics of the study group were shown in Table 1.
Reference intervals of serum CysC levels are presented in Table 2 Table 3 and Table4)

Discussion
Neonates, and particularly preterm newborns, are at increased risk for acute kidney injury (AKI) due to immature kidney function. It was reported 6.5% newborn infants admitted to the NICU developed acute kidney injury [14]. CysC has been shown to 7 be useful in diagnosing AKI and predicting its outcomes in children [15]. To determine the reference levels of serum CysC for newborn population will allow physicians to diagnose the disease accurately and timely. Different countries or regions usually establish corresponding RIs for their local populations. Here, we determined the reference levels of serum CysC for newborns in China. Most clinical studies have shown that neonatal serum CysC levels are mainly affected by PA [16][17][18][19]. However, those studies were limited as they enrolled small numbers of subjects. Our study had a larger sample size, and there were total of 1919 samples from 1044 patients. In our results, when the mean serum CysC levels of five postnatal subgroups in each GA group were compared, significant differences were found in all of the PA groups (P<0.001). And we found a trend toward decreasing serum CysC levels, as PA increases during the first months. (Show in Table 2) Previous researches have shown that CysC is superior to serum Cr for the estimation of the GFR in children [15,20,21]. Therefore, we further assessed the correlation of CysC or Cr with other biochemical parameters and perinatal factors. In the first days of life, neonatal serum Cr levels reflect maternal renal function and are increased in premature infants. It was reported serum Cr values reach stable levels by 1 to 2 weeks of age in term neonates, and inversely correlated with BW and GA in young infants [22,23]. Consistent with these researches, we found BW and GA inversely correlated with Cr on postnatal age 3 in this study (Table 4).. However, they were correlated positively within postnatal 24 hours, which further showed it is unreliable to assess renal function using Cr for newborn especially in the first days of life (Table 3)..
The available literatures have shown that CysC is independent of birth weight and gestational age [3,24,25,19]. In this study, with larger samples, we further confirmed CysC is independent of BW and GA in newborn, at least within 3 days after birth (P>0.01).
There are 2 assay methods for clinical laboratory use to measure serum Cr: the Jaffe and enzymatic methods, and serum Cr were measured by the Jaffe method in this study. The Jaffe method is known to affect seum Cr level as a result of non-specific protein interference [23]. Bilirubin is physiologically elevated in the newborn period.
It was reported bilirubin can affect serum Cr measurement by the Jaffe method [26].
In this study, the results showed that there was a statistically significant correlation between bilirubin levels and Cr level (P <0.001).
However, in our research, CysC was independent of bilirubin levels, which was similar to Treiber's research [12]. In this study, the correlation between HgB levels and serum CysC or Cr levels on the day 0 and day 3 were analyzed. We found no statistically significant correlation between serum CysC and HgB levels, which was similar to Armangil's report [27].
Initial studies suggest that CysC does not vary from non-renal factors, however further investigation has shown some biological variability [28]. Such as thyroid diseases, glucocorticoids treatment, and inflammation may affect CysC concentrations independent of kidney function [29-32]. Evangelopoulos AA ect al.
have a research in 490 adults who underwent a typical health examination. The results of this study demonstrated that peripheral monocyte blood count and WBC, were independently related with CysC concentrations [33]. In this study, we also found there was a statistically significant correlation between WBC and CysC levels within 24 hours after birth. These findings with ours suggest these biological variability factors must be considered when evaluating CysC levels in clinical practice. As most research demonstrated statistically significant correlation also 9 were found between serum CysC with both the Cr and BUN levels in this study (P<0.001) [13].

Conclusion
In summary, we present reference intervals of the serum CysC in newborn according to the GA and PA. In contrast to Cr, only few non-renal factors influence serum CysC, which suggests CysC is more stable to assess the kidney function of neonatal.
Several limitations exist in the present study. First, it was difficult to enroll "healthy" subjects, especially for preterm need treatment. Second, the number of subjects was still small for calculating completely reliable reference ranges, as shown in the subject numbers of GA lower than 28 weeks groups. Third, larger samples from multicenter study will help improve the precision of reliability of the data.

Ethics approval and consent to participate
Approval for the study was obtained from the Ethics Committee of Huazhong University of Science and Technology (NO.2016-S251). The ethics committee approves the exemption of subject's informed consent, because this study is completely harmless to subjects.

Availability of data and materials
Not applicable

Competing interests
The authors declare that they have no conflict of interest.