This prospective cohort study was done at Vidant medical center Greenville, North Carolina after Institutional Review Board approval. We included neonates (24 to 31 weeks' gestational age) within 48 h of birth. We excluded neonates with confirmed sepsis, renal failure and prenatal congenital renal abnormalities. We estimated GFR at 48 hours of life, and on day 14 and day 28 of life. Cystatin C GFR was estimated by Filler's equation [log GFR=1.962+ [1.123xlog (1/cystatin C)], and creatinine based CGFR by schwartz equation [GFR=k × height/cr, k=0.33] [5]. Serum Creatinine was IDMS traceable and Cystatin C was measured by nephelometric immunoassay method using international reference materials. Both methods were compared by Bland Altman plots. Post-conceptional age was plotted against serum cystatin C- and creatinine-based eGFR.

Serum creatinine and Cystatin C was measured in 37 samples taken from 14 infants mostly (12/14) delivered by Caesarian-section. Out of 14 patients, 11 had received gentamicin for suspected sepsis in first 48 hours, and all had respiratory distress syndrome. Four patients had required dopamine for hypotension, and one patient had received indomethacin for hemodynamically- significant patient ductus arteriosus. Mean serum creatinine levels decreased steadily (P=0.007), and mean eGFR using creatinine increased (P=0.001) during first month (Table I, Web Fig. 1a). Mean Serum cystatin C decreased from 1.39 mg/dL to 1.30 mg/dL (P=0.42). GFR estimated by cystatin C was significantly higher, and did not show much variability over time until the end of our observations (Web Fig. 1b,1c). The poor fit was seen in the Bland Altman plot with the mean difference (Filler - Schwarz) around 25 with nearly all differences greater than 25 when the mean of the two measure was 50 or less and nearly all differences less than 25 when the mean was greater than 50 (Web Fig. 1d).

TABLE I	Serum Creatinine And Cystatin C Value And Corresponding Glomerular  Filtration  Rate

In our study, serum creatinine and serum cystatin C levels decreased and eGFR increased with advanced postnatal age, possibly due to renal maturation [6]. This study shows that in preterm infants below 31 weeks, both methods for eGFR are poor fit to each other. They are especially different between 24-27 weeks of post conceptional age. This may, at least in part, be explained due to placental transfer of creatinine.

The limitations of the study are small sample size, influence of typical neonatal drugs and no gold standard measurement. As shown in (Table I), infants had different degrees of severity of illness that could have an impact on serum creatinine and CysC levels. Currently, there is no international calibration for CysC.

In conclusion, this study shows that Cystatin C and serum creatinine estimate GFR differently in preterm infants below 31 weeks.

1. Perrone RD, Madias NE, Levey AS. Serum creatinine as an index of renal function: New insights into old concepts. Clin Chem. 1992;38:1933-53.

2. Newman DJ. Cystatin C. Ann Clin Biochem. 2002;39:89-104.

3. Dharnidharka VR, Kwon C, Stevens G. Serum cystatin C is superior to serum creatinine as a marker of kidney function: A meta-analysis. Am J Kidney Dis. 2002;40:221-6.

4. Filler G, Grimmer J, Huang SH, Bariciak E. Cystatin C for the assessment of GFR in neonates with congenital renal anomalies. Nephrol Dial Transplant. 2012;27:3382-4.

5. Filler G, Huang SH, Yasin A. The usefulness of cystatin C and related formulae in pediatrics. Clin Chem Lab Med. 2012;50:2081-91.