Abstract:

Objective To investigate the feasibility of establishing the cut-off value of glucose-6-phosphate dehydrogenase（G6PD） screening for newborns in Shanghai by indirect method. Methods The G6PD determination data of 134 747 newborns were collected,and the Kolmogorov-Smirnov test was used to analyze the datum distribution. The skewed distribution data were converted into an approximately normal distribution using Box-Cox. The interquartile range（Turkey） method was used to eliminate outliers and establish a reference interval,with the lower limit of the reference interval [single-side 5th percentile value（P₅）] as the screening cut-off value. The established cut-off value with the lower limit of the reference interval declared by the manufacturer and the cut-off value established by a small sample in the clinical verification phase of the Neonatal Screening Laboratory of Shanghai Children's Hospital were compared,the relative deviation was calculated,and it was compared with the reference change value（RCV）. Results Among 134 747 newborns,233 cases of G6PD deficiency were diagnosed,and the incidence of G6PD deficiency was 1∶579. The indirect method was used to establish the G6PD screening cut-off value in Shanghai,which was 317.8 U/L for male infants and 325.7 U/L for female infants. The relative deviation of this cut-off value from the lower limit of the reference interval declared by the manufacturer and the cut-off value established by the Neonatal Screening Laboratory of Shanghai Children's Hospital based on a small sample was higher than RCV. The G6PD levels of 233 cases of G6PD deficiency were lower than the established screening cut-off value. Based on this screening cut-off value,the coincidence rate of both negative and positive for 70 external quality assessment samples reached 100%. Conclusions Using indirect method to establish the reference interval and the cut-off value of G6PD can improve the screening efficiency of neonatal G6PD deficiency.

Key words: Glucose-6-phosphate dehydrogenase, Neonatal screening, Indirect method, Cut-off value