Full-term infants with a gestational age of >37 weeks, birth weight ≥2,500 g, and who underwent quantitative detection of G6PD enzyme were included in this study. The age of admission ranged from 1 to 28 days. Jaundiced neonates were identified as those whose maximum total serum bilirubin (TSB) reached or exceeded the 95th percentile of the hour-specific TSB nomogram established by the Chinese Multicenter Study Coordination Group for Neonatal Hyperbilirubinemia in 2015 (6). Demographic and clinical data of the participants were collected from their medical records. The diagnostic standard for severe hyperbilirubinemia was TSB ≥ 342 μmol/L (6).

This study obtained approval from the Ethics Committee of the People's Hospital of Yangjiang in 2022 (No.20220063). The patients' data was analyzed anonymously, and blood samples were used after clinical diagnosis; therefore, the Ethics Committee of the People's Hospital of Yangjiang granted a waiver for written consent.

General information was gathered, including the infant's sex, gestational age, birth weight, age and weight at admission, delivery mode, feeding pattern, as well as treatment measures received. Hemolytic disease of the newborn (ABO and Rh incompatibility), glucose-6-phosphate dehydrogenase (G6PD) deficiency, infections (sepsis, pneumonia, and urinary tract infections), extravascular hemorrhage (intracranial hematoma, scalp hematoma, gastrointestinal, and other bleedings), congenital hypothyroidism (diagnosed through neonatal screening), and breast milk jaundice were among the major clinical etiologies examined. Breast milk jaundice was diagnosed by exclusion, in infants who were exclusively breastfed and older than 14 days, with no other identifiable pathogenic factors for jaundice; As a result, breast milk jaundice was not considered as a contributing factor in the combined analysis. If a patient was found to have two or more contributing factors for jaundice, they were documented in both the combined group and the respective single factor groups. Information about acute bilirubin encephalopathy (typical clinical symptoms of the nervous system and/or MRI imaging) and criteria for blood exchange [as recommended by the American Academy of Pediatrics in 2022 (7)] was also documented.

ABO incompatibility hemolysis was tested by three serological antibody tests using commercial three-cell panel (LIBO biotechnology, China, Co, Ltd.) by gel technique. Three tests included red blood cells direct antiglobulin test (direct Coombs test), free antibody test (free) and antibody release test. Serological diagnostic criteria for ABO incompatibility hemolysis were as follows: (1) confirmed cases were neonates with two positive results of the three tests or with a positive result on the antibody elution test; (2) suspected cases were only positive for either direct Coombs test or serum free antibody test. Antibody elution test was the final confirmed diagnosis for neonatal hemolysis disease (8).

The confirmation of G6PD enzyme deficiency was made by measuring the production rate of NADPH, using a detection kit (Beijing Antu Bioengineering Co., Ltd., China) following the manufacturer's protocols. Infants with a production rate of NADPH lower than 2,500 U/L were categorized as G6PD-deficient (9).

To determine the frequency of G6PD deficiency in Yangjiang, routine body check-ups were conducted on male adults at our hospital, and they were also tested for G6PD deficiency using the same detection kit mentioned above. The deficiency was defined as G6PD enzyme activity lower than 1,300 U/L.

The data was analyzed by SPSS 23.0 (IBM SPSS 23.0) and shown as mean ± standard deviation. Differences in continuous variables between the two groups were analyzed by the Mann–Whitney nonparametric test. The significance of differences in the categorical variables was determined by the Chi-square test or Fisher's exact test. P < 0.05 was considered statistically significant.

From June 2018 to July 2022, a quantitative assay of G6PD enzyme was conducted on 1,828 in-hospital infants. Of these, 1,119 were term infants, with 435 presenting with jaundice and 684 presenting normal bilirubin levels. G6PD deficiency accounted for 9.66% (42/435) hyperbilirubinemia cases, while multiple etiologies, including infection (pneumonia, sepsis, and upper respiratory tract infection), hemorrhage (delivery injury, scalp hematoma, gastrointestinal bleeding), ABO incompatibility hemolysis, and breast-milk jaundice were responsible for 22.3% (97/435), 16.09% (70/435), 8.05% (35/435), 3.45% (15/435), and 2.53% (11/435) of cases, respectively. One case (0.23%, 1/435) was attributed to congenital hypothyroidism, and 2.07% (9/435) of cases were caused by other factors such as thalassemia, mothers with diabetes, and hereditary spherocytosis. The etiology was unknown in 35.63% (155/435) of cases (Figure 1).

Among the 435 cases of jaundice, severe hyperbilirubinemia (TSB ≥342 μmol/L) was observed in 50 cases, 8 cases of which presented with acute bilirubin encephalopathy (2 males and 6 females), and 7 cases required exchange transfusions. The primary causes of acute bilirubin encephalopathy were G6PD deficiency (3 cases) and hemorrhage (3 cases), followed by infection (2 cases of sepsis and 1 case of pneumonia), ABO incompatibility hemolysis (2 cases), and congenital hypothyroidism (1 case) (Table 1). One female infant with acute bilirubin encephalopathy was admitted at the age of 17 days, who had congenital hypothyroidism combined with G6PD deficiency and had not undergone neonatal screening at birth (Table 1).

Of the 435 cases with jaundice, 19.54% (60 males, 25 females) were found to be G6PD-deficient, while only 10.23% (47 males, 23 females) of the 684 cases with normal bilirubin levels were G6PD-deficient. The prevalence of G6PD deficiency was significantly higher in infants with jaundice compared to their counterparts (P < 0.001) (Table 2). The prevalence of G6PD-deficient in severe hyperbilirubinemia (TSB ≥ 342 µmol/L) and mild-medium jaundice (TSB < 342 µmol/L) was 30% (15 out of 50, 11 males, 4 females) and 18.18% (70 out of 385, 49 males, 11 females), respectively. A statistically significant difference was observed between the two groups (P = 0.041) (Table 3). In comparison, G6PD deficiency was present in 8.75% (42/480) of male adults during routine body check-ups.

Of the 435 infants with jaundice, one case was excluded due to the lack of hemoglobin data. The remaining cases were grouped according to jaundice causes—normal, G6PD-deficient, ABO incompatibility hemolysis, and ABO incompatibility hemolysis combined with G6PD-deficiency. The difference in total bilirubin levels among these four groups was compared (Table 4), and the peak bilirubin values were not statistically significant (P = 0.95). Among infants with jaundice, the hemoglobin levels of 80 G6PD-deficient infants were 146.85 ± 24.88 g/L, which was significantly lower than that of infants with normal G6PD (156.30 ± 22.07 g/L) (P = 0.001) (Table 5). Meanwhile, the hemoglobin levels of 27 infants with ABO incompatibility hemolysis were at 134.33 ± 24.18 g/L, slightly lower than that of infants with G6PD deficiency (146.85 ± 24.88 g/L) (P = 0.014).

A total of 65 blood samples from jaundiced infants with G6PD deficiency were analyzed for G6PD genotypes using reverse dot hybridization (10). Five types of gene mutations and one polymorphism were detected in infants with jaundice, namely c.95A > G, c.392G > T, c.1024C > T, c.1376G > T, c.1388G > A, and c.1311C > T (polymorphism). Furthermore, three kinds of compound heterozygous mutations were identified, namely c.871G > A/c.1311C > T, c.392G > T/c.1388G > A, and c.1376G > T/c.1311C > T (Table 6). Specifically, c.871G > A was consistently linked to 1311C > T in all nine cases (Table 6). Moreover, one G6PD-deficient infant developed acute bilirubin encephalopathy with compound heterozygous mutations of c.392G > T and c.1388G > A.