To analyze relationships between results of neonatal screening of congenital hypothyroidism (CH) and phenylketonuria (PKU) and the time of blood collection.

From January 1, 2012 to June 30, 2014, a total of 111 098 newborn infants who were born in Department of Obstetrics in 19 hospitals of Haidian District, Beijing were recruited as research subjects. By retrospective analysis method, according to hospital levels the newborns were born, they were divided into group of third-level hospitals (n=48 472), and group of second-level hospitals (n=62 626). The gestational age, time of blood collection of newborns between two groups were compared statistically by independent-samples t test.The constituent ratios of gender and modes of delivery, detection rates of suspicious CH newborns with serum thyroid stimulating hormone (TSH) ≥ 10 μU/mL, and suspicious PKU newborns with serum phenylketonuria (Phe) ≥ 2 mg/dL, detection rates of suspicious CH newborns with serum TSH = 10 μU/mL and suspicious PKU newborns with serum Phe = 2 mg/dL between two groups were compared statistically by chi-square test. This study protocol was consistent with the requirements of World Medical Association Declaration of Helsinki revised in 2013.

①There were no significant differences between two groups in the constituent ratios of gender and modes of delivery (P>0.05). ②Time of blood collection for neonatal screening of CH and PKU in group of second-level hospitals was (123.5±13.5) h after birth, which was earlier than that of group of third-level hospitals (138.6±20.0) h after birth, and the difference was statistically significant (t=442.345, P<0.001). ③There were no significant differences between two groups in the detection rates of suspicious CH newborns with serum TSH ≥ 10 μU/mL and suspicious PKU newborns with serum Phe ≥ 2 mg/dL (P>0.05). ④The detection rate of suspicious CH newborns with serum TSH = 10 μU/mL in group of second-level hospitals was 0.251% (157/62 626), which was obviously higher than that of group of third-level hospitals 0.140% (69/48 472), and the difference was statistically significant (χ²=15.799, P<0.001). But there was no significant difference between two groups in the detection rates of and suspicious PKU newborns with serum Phe = 2 mg/dL (P>0.05). ⑤ The diagnosis rate of newborns with CH or with PKU was 0.057% (63/111 098) and 0.014% (15/111 098), respectively.

Time of blood collection for neonatal screening of CH and PKU in second-level hospitals is earlier than that of third-level hospitals. Detection rate of suspicious CH newborns with serum TSH = 10 μU/mL in second-level hospitals is much higher than that of third-level hospitals. Because this study is just a retrospective study, the relationships between detection rate of suspicious CH and PKU newborns, and time of blood collection needs to be further explored by prospective randomized controlled trials, so as to reduce the detection rate of suspected CH and PKU newborns, and improve the diagnosis rate of suspected CH and PKU newborns.