This retrospective study was conducted involving clinical data of 946 consecutive cases from 6892 pregnant women with late antenatal screening of GBS tests from January to December 2020, and the files of GBS-positive neonates from January 2019 to December 2020. Two swabs were collected from the anus and vagina of each pregnant woman, respectively, at 35–38 weeks gestation. The swabs were collected from the nasopharynx or ears of the newborns with clinical symptoms of infection after birth from GBS-positive mothers. All information was obtained from the laboratory information system of YuHuangding Hospital (Yantai, Shandong, China).

Inclusion and exclusion criteria, the 946 consecutive cases were during the period from all 6892 pregnant women with GBS tests and delivered in our hospital later. At the same time excluded the cases with other risk factors which can cause adverse pregnancy: myoma uteri, uterine malformations, multiple pregnancies, preeclampsia, hypertension, diabetes mellitus, fetal malformations, fetal growth restriction, intra-uterine fetal demise, and other genital tract pathogen infections. And then, to devided the 946 cases into three groups, in which 500 GBS-negative cases and 446 GBS-positive cases, shown in Tables 1,2,3. The first group was the premature delivery group at 37 weeks gestation. The second group was the premature rupture of membranes (PROM) group and included pregnant women of all gestational ages with premature rupture of membranes, not limited to preterm women 37 weeks. The third group was the chorioamnionitis group. Inclusion and exclusion criteria of pregnant women are shown in Fig. 1.

Fig. 1.

Flow chart of the inclusion criteria of the pregnant women.

There were 15,967 newborns whom were born in our hospital from January 2019 to December 2020, of which 38 cases were GBS-positive. The other 10 GBS-positive cases were transferred from other hospitals after the onset of GBS infection. The inclusion criteria for the newborns are shown in Fig. 2. All data from the 38 GBS-positive cases and the 10 GBS-positive cases transferred from other hospitals were compared in Table 4. According to the clinical symptoms of GBS-positive children, they were divided into three groups: GBS-colonization group; GBS-infection group; and death group. The deaths in the mode of delivery groups were merged into the corresponding infection groups in Table 4. With respect to the GBS-colonization site in the mother group, the death group was also merged into the corresponding infection groups. The result of pairwise comparison between groups by the Kruskal Wallis H-test is shown in Table 5.

Fig. 2.

Analysis of infants. *When calculating the infection rate and mortality rate of newborns, the 10 cases born in the other hospitals were not included in the 15,967 cases born in our hospital.

The GBS-specific CAMP gene was detected by polymerase chain amplification and fluorescent labeling of the GBS reagent (Taipu Biological Co. LTD, Fuzhou, Fujian, China), and the results were analyzed using an ABI 7500 Real-time PCR System (Thermo Fisher Scientific Inc., Waltham, MA, USA). After sampling, 1 mL of cleaning solution was added to elute the cotton swabs and shaken for 2 min to make the suspension. All suspensions were put into 1.5 mL centrifuge tubes and the supernatants were discarded after centrifuging at 13,000 rpm for 5 min. The above process was repeated twice, then 50 µL of cleaning solution was added to make a suspension. To draw 50 µL of the suspension washed from the sample swabs, a 50 µL positive control and 50 µL negative control was dispensed into 1.5 mL centrifuge tubes. Ten microliters of internal reference were added to each tube, then dry-bathed at 95 °C for 2 min and ice-bathed at –20 °C for 2–5 min. After centrifugation at 13,000 rpm for 1 min, 5 µL of the supernatant was used for PCR amplification with 44.3 µL of PCR reaction solution, 0.5 µL of Taq DNA polymerase, and 0.2 µL of uracil N-glycosylase (UNG). The PCR reaction solution contained buffer, MgCl${}_2$, primers, probes, and dNTP. The cycle parameters of this experiment were 37 °C for 2 min, 94 °C for 2 min, 10 cycles at 94 °C for 20 sec and 55 °C for 45 sec, and 30 cycles at 94 °C for 20 sec and 55 °C for 45 sec.

Data are presented as a percentage, median and interquartile range for different variables. A Chi-square test was utilized to assess the categorical data and the Kruskal-Wallis test was used for measurement data with SPSS 24.0 (IBM Corp., Armonk, NY, USA). p-value $\le$ 0.05 was considered statistically significant.

The GBS-colonization rate of pregnant women in our study was 6.75%, which was close to the average level of the GBS-positive rate in mainland China (3.7–14.52%) [8]. The GBS-positive rates in the vagina, anus, and rectovaginal were 0.51% (35/6892), 3.38% (233/6892), and 2.86% (197/6892), respectively; thus, the GBS-positive rate in the anus was higher than the vagina because the GBS primary reservoir is the gastrointestinal tract [3]. The positive rate of GBS was different among the different age groups. It showed that the prevalence of GBS in younger women was higher than in older women. These results are similar to previous studies [12, 13, 14].

Between GBS-positive group and GBS-negative group, there was no difference in gestational weeks, PROM, preterm delivery, and chorioamnionitis groups. Contrary to previous studies [11, 15, 16], in our current research there was no significant difference between the two groups regarding adverse pregnancy outcomes, which is consistent with the findings by Goel et al. [13] and Ngonzi et al. [17]. Accordingly, in the Tano et al. [18] study, there was no pathologic evidence to support a connection between GBS-infection and chorioamnionitis, which is consistent with our study.

We excluded all cases with diabetes, hypertension, and preeclampsia, which may cause premature labor or PROM [19, 20, 21]. The cases with Ureaplasma urealyticum, Chlamydia trachomatis, Candida, Neisseria gonorrhoeae, and Gardnerella vaginalis, which have a pathological role in PROM and chorioamnionitis [12, 22] were also eliminated. Rocchetti et al. [12] confirmed that candidiasis and cytolytic vaginosis also increase GBS-colonization. Therefore, the seco-infection cases were deleted because it was difficult to determine the actual pathogenic factor leading to the maternal infection.

Additionally, either diabetes or hypertension are risk factors for premature labor and PROM [23], and also increase the likelihood of GBS-colonization [10, 15]. These confounding factors not only increase the rate of GBS-colonization, but also induce adverse pregnancy outcomes. Therefore, it is necessary to exclude the confounding factors and to compare a single factor for GBS to confirm the role of GBS in adverse pregnancy outcomes. Moreover, these confounding factors may be the reason for the differences in results between studies.

Most of the GBS-infected neonates had pneumonia, followed by pyemia, bacteriaemia, meningitis, and even death. The emergence of symptoms of infection and death were mostly concentrated within the first 24 hours. There were significant differences in the three indices (gestational age, neonatal birth weight, and Apgar scores), reflecting the health status of newborns among the three groups. Infants with poor basic conditions are concentrated in the infection and death groups.

In the weight group, there was only a difference between the colonization and death groups. The two dead cases in our study were premature babies with low weight, and similar cases were also reported by Todorova-Christova et al. [24], that low weight at birth or prematurity was confirmed as a substantial risk factor of GBS-infection. In the Apgar score group, there were differences among all three groups. Therefore, we believe that the occurrence and progression of GBS-infections were related to the basic physical health status of newborns, which is consistent with Mousavi et al. [10], which reported that GBS can give rise to life-threatening infections in some vulnerable hosts, especially infants with chronic diseases.

A difference also existed in the mode of delivery so that the colonization and GBS-infection rates in newborns that were delivered spontaneously were higher than newborns delivered by cesarean section, which is consistent with the results of Verani et al. [3] and Joachim et al. [25]

In the GBS-colonization site group, most of the rectovaginal-positive mothers caused vertical transmission (37.5%, 18.75%, and 2.08%). Three GBS-positive newborns were born from GBS-negative mothers, which confirmed the intermittent and transient nature of GBS-colonization [11, 26]. Of the 48 GBS-positive newborns, the mothers of 15 did not have GBS screening or had false-negative results, thus accounting for 57.7% (15/26) of infected children and the infection rate was higher than the newborns from the maternal screening group, which confirmed that prenatal screening of GBS is beneficial to reduce GBS-infections in neonates.

Among the 30 mothers who had GBS screening, there were 10 newborn infections, 1 newborn dead, and 19 newborns colonized. 1 dead newborn was a premature infant at 33 weeks + 5 days gestation with a birth weight of 1700 g, and an Apgar score of 8 at 1 min. Although the mother received IAP, the prophylactic antibiotics failed to effectively prevent fetal infection and death. The mother of another dead newborn was not screened for GBS; the newborn was born at 25 weeks + 2 days gestational age with a birth weight of 800 g and Apgar score of 3 at 1 min. The two deceased children had poor basic conditions and were difficult to survive in the case of combined GBS-infection. All of the above data are consistent with most studies that concluded that premature infants in poor health with GBS-infections usually have a poor prognosis [10, 24].