- Email: [email protected]
Effects of Intrapartum Antibiotic Prophylaxis on Neonatal Acquisition of Group B Streptococci
THE JOURNAL OF PEDIATRICS • www.jpeds.com
ORIGINAL ARTICLES
Effects of Intrapartum Antibiotic Prophylaxis on Neonatal Acquisition of Group B Streptococci Meiwa Toyofuku, MD1,2, Miyuki Morozumi, PhD2, Mariko Hida, MD, PhD1, Yoshitake Satoh, MD, PhD3, Hiroshi Sakata, MD, PhD4, Hiroyuki Shiro, MD, PhD1, Kimiko Ubukata, PhD2, Mitsuru Murata, MD, PhD5, and Satoshi Iwata, MD, PhD2 Objectives To assess the incidence of colonization with group B streptococci (GBS) among neonates as influenced by maternal GBS carriage and intrapartum antibiotic prophylaxis (IAP). Study design Between October 2014 and May 2015, nasopharyngeal and rectal swab samples were collected from 730 neonates at 1 week and 1 month after birth. GBS and capsular serotype were identified by real-time polymerase chain reaction and by culture. IAP at delivery was determined retrospectively from hospital records. Results Sixty-four neonates (8.8%) were GBS-positive by real-time polymerase chain reaction and culture. Among neonates born to mothers who were GBS carriers (n = 107), 94.4% (101/107) had maternal IAP; 19.6% nonetheless were GBS-positive, compared with 6.5% of neonates born to noncarrier mothers (P < .01). Among neonates born to mothers receiving IAP, more were positive only at 1 month of age than at both 1 week and 1 month. The frequency of GBS in neonates born to mothers receiving IAP was significantly lower than that in neonates born to mothers not receiving IAP (P < .05). Capsular serotypes V (25%) and III (23.4%) were common, followed by Ib (15.6%), Ia (14.1%), II (7.8%), IV (6.3%), nontypeable (4.7%), and VI and VIII (each 1.6%). Conclusions Delayed colonization with GBS occurs in infants born to GBS carrier mothers receiving IAP. GBS should be considered in all infants at 1 month after birth with signs of infection. (J Pediatr 2017;190:169-73).
G
roup B streptococcus (GBS) is a leading cause of severe infections in early infancy. These infections include earlyonset disease (EOD), occurring within 6 days and late-onset disease (LOD), occurring from 7 to 89 days. Among pregnant women, the GBS carrier rate was estimated to be near 13% (range, 12%-22%).1 In approximately one-half of carriers, GBS is transmitted to the infant, with invasive GBS disease developing in 1%-2%. To prevent EOD due to GBS, a guideline was introduced by the US Centers for Disease Control and Prevention and the American College of Obstetricians and Gynecologists in 1996. Universal screening for maternal GBS colonization and intrapartum antibiotic prophylaxis (IAP) strategies implemented according to the guideline have reduced the incidence of EOD but had no impact on LOD.2,3 In Japan, a revised guideline for obstetric practice published in 2014 recommended universal GBS screening by conventional culture for all pregnant women at 33-37 weeks of gestation.4 IAP during labor or after premature rupture of membranes was recommended for women with any of the following characteristics: GBS culture positivity, GBS infection during infancy in previous offspring, or incidental detection of GBS in a urine culture during the current pregnancy. If GBS status in the current pregnancy is unknown, IAP should be given to mothers with symptoms at delivery that might be attributable to GBS. In this study, we investigated prevalence of GBS colonization in neonates examined at 1 week and 1 month after birth, influences of IAP and GBS carriage in the mother, and capsular serotype and sequence type (ST) of GBS isolates from neonates.
Methods We conducted a prospective longitudinal cohort study at 4 hospitals, which were located in Hokkaido, Gunma, and Kanagawa prefectures and in Tokyo. The study was performed from October 2014 to May 2015. Nasopharyngeal and rectal swabs
EOD GBS IAP LOD MLST PCR ST
Early-onset disease Group B streptococcus Intrapartum antibiotic prophylaxis Late-onset disease Multilocus sequence typing Polymerase chain reaction Sequence type
From the 1Division of Pediatrics, Yokohama Rosai Hospital, Yokohama; 2Department of Infectious Diseases, Keio University School of Medicine, Tokyo; 3Division of Pediatrics, Ota Memorial Hospital, Ota; 4Division of Pediatrics, Asahikawa Kosei Hospital, Asahikawa; and 5 Department of Laboratory Medicine, Keio University School of Medicine, Tokyo, Japan M.T. received a grant from the Japanese Society for Pediatric Infectious Diseases. The other authors declare no conflicts of interest. 0022-3476/$ - see front matter. © 2017 Elsevier Inc. All rights reserved. https://doi.org10.1016/j.jpeds.2017.07.039
169
THE JOURNAL OF PEDIATRICS • www.jpeds.com were collected from all eligible infants at 1 week and 1 month after birth and transported immediately to our laboratory. All samples were tested to identify GBS by amplification of the dltS gene, which encodes a histidine kinase, and also to identify capsular types Ia, Ib, and III using real-time polymerase chain reaction (PCR) as developed by Morozumi et al.5 Correlation between results of real-time PCR and bacterial culture also was examined. As shown in Figure 1 (available at www.jpeds.com), samples were collected via cotton-tipped swabs (BBL CultureSwab Plus, Copan, Italy). Immediately after receipt, swab samples were suspended in 0.5 mL of Todd-Hewitt broth (Becton, Dickinson, Franklin Lakes, New Jersey) and centrifuged at 2000g for 5 minutes at 4°C to collect bacterial cells. After the supernatant was discarded, 55 µL of pellet was stirred and used for direct DNA extraction (50 µL) and for culture (5 µL). Each sample of 50 µL was placed in 45 µL of a lysis solution that included SimplePrep reagent for DNA (Takara Bio, Shiga, Japan) and 2 U of mutanolysin (Sigma Aldrich, St. Louis, Missouri). The lytic reaction was carried out for 10 minutes at 37°C, followed by 3 minutes at 95°C. The lysate was added to each of the tubes containing PCR mixtures for the following gene identifications: the dltS gene, which encodes a histidine kinase specific to GBS; and capsular types Ia, Ib, and III. Total volume (30 µL) of the PCR mixture included 20 pmol of each primer, 25 pmol of each probe, 2× Multiplex Powermix (Bio-Rad, Hercules, California), and DNase- and RNase-free distilled water. DNA amplification was carried out for 45 cycles as follows: 95°C for 10 seconds, 50°C for 30 seconds, and 72°C for 20 seconds. The remaining 5 µL of each sample was inoculated on sheep blood agar (Nippon Becton, Dickinson, Tokyo, Japan), which then was incubated at 37°C for 20 hours in an atmosphere containing 5% CO2. Colonies of b-hemolytic streptococci grown on the blood agar plates underwent identification of capsular type (Ia, Ib, II, III, IV, V, VI, VII, and VIII) and the dltS gene by real-time PCR. For multilocus sequence typing (MLST) analysis, primer sets corresponding to 7 housekeeping genes (adhP, atr, glnA, glcK, pheS, sdhA, and tkt) used for MLST were constructed with reference to the MLST Web site (http://pubmlst.org/sagalactiae/). MLST was applied to the sequences for these 7 genes accord-
Volume 190
•
November 2017
ing to previously described methods,6 with alleles and ST assignments determined using the Streptococcus agalactiae MLST database. Statistical Analyses Statistical analyses were performed with Ekuseru-Toukei 2012 software for statistics (Social Survey Research information, Tokyo, Japan). The Mann-Whitney U test, the c2 test, Fisher exact test, the Wilcoxon rank sum test, or univariate logistic regression analysis was used as appropriate. A P value < .05 was considered to indicate statistical significance. Field workers explained the purpose of the study to mothers of eligible infants, and each mother signed an informed consent form just before discharge or at 1 week after delivery. The study was approved by the Yokohama Rosai Hospital Ethics Committee (approval number 26-51).
Results Figure 2 is a flowchart showing transmission of GBS among 730 neonates according to GBS status of their mothers as determined by prenatal screening. GBS screening via bacterial cultures was performed at 33-37 weeks of gestation for 710 mothers (97.3%) receiving care at institutions participating in the study, and 107 mothers (14.7%) proved to be GBS carriers. Twenty mothers (2.7%) did not undergo screening. GBS-positive neonates numbered 64, including 46 positive by both real-time PCR and culture and 18 positive only by real-time PCR. Twenty-one neonates born to the 107 GBS carriers were positive for GBS (19.6%); among neonates born to the 603 noncarriers, 39 (6.5%) were positive for GBS. Frequency of GBS positivity in neonates differed significantly between carrier and noncarrier mothers (P < .001). Four neonates born to mothers without GBS screening were positive for GBS (20%). No infants developed invasive GBS infection during the study period. Relationships between time points of GBS positivity in neonates and maternal IAP are shown in Table I. GBS-positive neonates were divided into 3 groups: GBS persistence (n = 16, 25%), indicating GBS positivity at both 1 week and 1 month after birth; GBS clearance (n = 7, 10.9%),
Subjects Mother & infant (n = 730)
Maternal GBS (+) 14.7% (n = 107)
Infant GBS (+) 19.6% (n = 21)
Infant GBS (-) 80.4% (n = 86)
Maternal GBS (-) 82.6% (n = 603)
Infant GBS (+) 6.5% (n = 39)
Infant GBS (-) 93.5% (n = 564)
Maternal GBS unknown 2.7% (n = 20)
Infant GBS (+) 20.0% (n = 4)
Infant GBS (-) 80.0% (n = 16)
Figure 2. Flowchart showing presence or absence of GBS carriage in mothers and neonates. 170
Toyofuku et al
ORIGINAL ARTICLES
November 2017
Table I. Association between GBS positivity in neonates and maternal GBS colonization Mothers GBS(+) (n = 107) Neonates Persistence group, n (%)‡ Clearance group, n (%)‡ Delayed positivity group, n (%)‡ Total
GBS(−) (n = 603)
Screening (−) (n = 20)
n (%)
IAP* Yes (n = 101)
IAP No (n = 6)
IAP* Yes (n = 196)
IAP No (n = 407)
IAP* Yes (n = 17)
IAP† No (n = 3)
16 (25.0) 7 (10.9) 41 (64.1) 64
2 (10.5) 2 (10.5) 15 (78.9) 19
1 0 1 2
1 0 6 7
10 (31.3) 5 (15.6) 17 (53.1) 32
1 0 2 3
1 0 0 1
*Among mothers with IAP, cefazolin was given to 189; ampicillin to 107; and other agents to 15. †One woman was not given IAP despite preterm delivery. Two women were not given IAP, in accordance with the guideline. ‡Persistence group, positive for GBS at 1 wk and 1 mo after birth; clearance group, positive only at 1 wk after birth; Delayed positivity group, negative for GBS at 1 wk, but positive at 1 mo after birth.
indicating GBS positivity at 1 week but not at 1 month after birth; or delayed GBS positivity (n = 41, 64.1%), indicating GBS negativity at 1 week but positivity at 1 month after birth. Of the 107 GBS carrier mothers, 101 (94.4%) received IAP, among whom ampicillin was administered to 84 (83.2%), cefazolin to 14 (13.9%), and other agents to 3 (3%). Although IAP was given, 19 neonates (18.8%) born to GBS carrier mothers were positive for GBS after birth. Notably, most of these neonates, numbering 15 (78.9%), belonged to the delayedpositivity group. Of 603 pregnant women found to be GBS noncarriers, 196 (32.5%) received intravenous antibiotics during labor; among their infants, 6 of the 7 GBS-positive neonates were in the delayed-positivity group. Among GBS-positive neonates whose mothers received no antibiotic agents, 10 (31.3%) had GBS persistence, 5 (15.6%) had clearance, and 17 (53.1%) had delayed positivity. Generally, delayed positivity was significantly more prevalent in neonates born to mothers who received IAP (P = .04) than in neonates born to mothers not receiving IAP. In the persistence group, a significant difference was not evident between infants of mothers with and without IAP (P = .08). Table II shows a correlation between threshold cycle values derived from real-time PCR assays in neonates and implementation of IAP. Neonates born to mothers receiving IAP showed a significantly lower GBS bacterial load than neonates born to mothers not receiving IAP (P = .04).
Table II. Threshold cycle values derived from realtime PCR in neonates according to IAP Mothers (n = 710) Threshold cycle and number of bacterial cells <24 cycles (>5 × 104 cells/sample) 25-28 cycles (103 –5×104 cells/sample) >29 cycles (<103 cells/sample) Total
IAP Yes (n = 297)
IAP No (n = 413)
Total
2 (18.2)
9 (81.8)
11
8 (44.4)
10 (55.6)
18
19 (54.3)
16 (45.7)
35
29
35
64
P = .04. P values were calculated with the Wilcoxon rank sum test.
Maternal and neonatal characteristics are analyzed for relationships to neonatal GBS positivity at 1 month after birth in Table III (available at www.jpeds.com). Only maternal GBS carrier status showed a significant difference between GBS-positive neonates (n = 57) and GBS-negative neonates after birth (n = 673) at 1 month (P < .001). Univariate logistic regression analysis of these characteristics identified maternal GBS colonization as a risk factor associated with neonatal GBS positivity (OR 3.6; 95% CI 2.0-6.6). Capsular serotypes for all GBS isolates were identified by real-time PCR. As shown in Table IV, capsular serotypes V (n = 16, 25%), III (n = 15, 23.4%), Ib (n = 10, 15.6%), and Ia (n = 9, 14.1%) were predominant, followed by II (n = 5), IV (n = 4), nontypeable (n = 3), VI and VIII (n = 1, respectively). In addition, the most common serotype in the persistence group (n = 16), clearance group (n = 7) and delayedpositivity group (n = 41) was Ib (31.3%), III (57.1%), and V (29.3%), respectively. Forty-five GBS strains isolated from 64 GBS-positive samples by bacterial culture in neonates (70.3%) were analyzed for capsular serotype and MLST. Although 10 STs were found, ST1 (28.9%), ST335 (24.4%), and ST10 (15.6%) accounted for 68.9% of all 45 isolates. Most isolates with capsular serotype III were ST335, which is derived from ST19. ST17, linked with high virulence, was found in only 1 strain.
Discussion In this prospective cohort study of neonatal GBS colonization following the 2014 introduction of the current guidelines for obstetric practice in Japan, combining real-time PCR with culture methods, we found an overall prevalence of GBS colonization within 1 month after birth of 8.8%. Notably, the positivity rate in neonates born to GBS carrier mothers, almost all of whom received IAP, was significantly greater (19.6%) than positivity in neonates whose mothers were not carriers (6.5%). Positivity rates in our Japanese infants were considerably lower than those among infants in Denmark (11.3%),7 Gambia (24.8%),8 and Italy (21.9%).9 We previously estimated GBS carriers among pregnant women to be 15.7% by real-time PCR and 12.6% by bacterial culture.5 We believe that high compliance with the obstetric guideline in Japan most likely was
Effects of Intrapartum Antibiotic Prophylaxis on Neonatal Acquisition of Group B Streptococci
171
THE JOURNAL OF PEDIATRICS • www.jpeds.com
Volume 190
Table IV. Relationships between capsular type and ST Capsular type STs* ST1 ST10 ST17 ST19 ST23 ST26 ST130 ST144 ST196 ST335† Unknown‡ Total (%)
Total, n (%) 13 7 1 1 4 1 1 2 4 11 19
(20.3) (10.9) (1.6) (1.6) (6.3) (1.6) (1.6) (3.1) (6.3) (17.2) (29.7) 64
Ia
Ib
II
III
IV
4
V
VI
8
1
VIII
NT
1 1 (1.6)
3 3 (4.7)
7 1 1 3
1 1 1
2 4 4 9 (14.1)
3 10 (15.6)
1 5 (7.8)
11 1 15 (23.4)
4 (6.3)
6 16 (25.0)
1 (1.6)
*Only GBS isolates by culture were tested. †ST335 was derived from ST19. ‡Capsular type in 16 cases was identified by real-time PCR.
responsible for lower neonatal GBS positivity than in other countries. Neonatal GBS colonization of the pharynx, intestine, or umbilical cord is well known to precede development of invasive GBS infection.10,11 Reported incidence of GBS disease was 0.57 per 1000 live births in Europe and 0.67 in the Americas.1 According to a previous report,12 incidence of invasive GBS disease in Japan was much lower, 0.18 per 1000 live births, which may be attributable to a low rate of GBS maternal transmission to infants. Interestingly, delayed positivity was significantly more frequent than persistent positivity, especially in neonates born to mothers receiving IAP. The density of GBS colonization in neonates with delayed positivity was estimated to be relatively low by real-time PCR. In contrast, persistent positivity was more frequent among neonates born to mothers without IAP compared with those born to mothers with IAP. We suspect that after IAP was initiated, density of GBS colonization of mothers decreased temporarily, but the organism was not eradicated. These small numbers of GBS, transmitted from mother to newborn, may then have multiplied gradually in the respiratory or intestinal tract after escaping detection at 1 week after birth. Accordingly, IAP strategies could reduce incidence of EOD but lack impact on incidence of LOD. Pregnant women carrying GBS therefore remain an important risk factor for neonatal invasive GBS infection. Neonatal GBS infection also could occur following transmission from the mother by a route unrelated to delivery, or by horizontal transfer from another person such as a family member after discharge from the hospital. In our study, the predominant capsular serotypes of GBS carried by neonates were types V and III. Serotype III is most common among isolates from infants with invasive GBS.12-14 Relative capsular serotype prevalence is similar to that previously reported worldwide.15 Among GBS isolates from Japanese maternal carriers, the predominant capsular serotype was type Ib, followed by types V and III5; other countries have differed in maternal carrier serotype prevalence.8,9 In Japan, se172
rotype V has shown little involvement in invasive GBS disease during early infancy, whereas in the US serotype V is increasing in prevalence both in pregnancy and in invasive disease during early infancy.16 Importantly, serotype patterns can shift. ST17 GBS, which has high virulence and belongs to capsular serotype III, accounted for 50% of invasive GBS isolates in Japan,13 similar to reports from the Netherlands.3 In our study, only one ST17 GBS was identified. Our study has several limitations. First, US Centers for Disease Control and Prevention guidelines indicate that IAP administration beginning at least 4 hours before delivery has been found highly effective.2 Regrettably, detailed information about when IAP was administered relative to delivery was unavailable for our study. Second, bacterial culture of maternal samples was carried out without use of a selective medium for GBS isolation at each institution. Therefore, false-negative cases may have been included among “noncarrier” mothers. Application of real-time PCR or conventional PCR with high sensitivity and specificity is expected to enhance accuracy of maternal testing. Finally, we could not identify capsular serotypes in GBS-colonized mothers, leaving open the possibility that instances of GBS colonization of infants may not all have resulted from vertical transmission at birth. Uniform testing that combines real-time PCR with bacterial culture for both pregnant women and neonates will be necessary to further address this question. However, such testing cannot distinguish transmission at delivery from subsequent transmission involving ordinary maternal contact. The ability of IAP to prevent LOD due to GBS may be limited even with high-precision testing for both pregnant women and neonates. Further studies should determine whether surveillance microbiologic assessment at 1 month after birth and augmented hygiene can reduce risk of LOD. GBS vaccination of pregnant women represents an alternative way to protect infants from invasive GBS disease. This strategy is under development, with the leading candidate for maternal vaccination being a trivalent GBS vaccine (CRM197-conjugated capsular polysaccharides of GBS serotype Ia, Ib, and III).17 ■ Toyofuku et al
ORIGINAL ARTICLES
November 2017 We thank Atsuo Sato, MD, Masaki Nakayama, MD, Miyuu Matsui, MD, Tomoko Furutani, MD, and Kumiko Tago, MD for collecting data. Submitted for publication Apr 22, 2017; last revision received Jun 22, 2017; accepted Jul 19, 2017
References 1. Kobayashi M, Vekemans J, Baker CJ, Ratner AJ, Le Doare K, Schrag SJ. Group B Streptococcus vaccine development: present status and future considerations, with emphasis on perspectives for low and middle income countries. F1000Res 2016;5:2355. 2. Verani JR, McGee L, Schrag SJ. Prevention of perinatal group B streptococcal disease—revised guidelines from CDC, 2010. MMWR Recomm Rep 2010;59:1-36. 3. Van Dyke MK, Phares CR, Lynfield R, Thomas AR, Arnola KE, Craig AS, et al. Evaluation of universal antenatal screening for Group B Streptococcus. N Engl J Med 2009;360:2626-36. 4. Minakami H, Maeda T, Fujii T, Hamada H, Iitsuka Y, Itakura A, et al. Guidelines for obstetrical practice in Japan: Japan Society of Obstetrics and Gynecology (JSOG) and Japan Association of Obstetricians and Gynecologists (JAOG) 2014 edition. J Obstet Gynaecol Res 2014;40:146999. 5. Morozumi M, Chiba N, Igarashi Y, Mitsuhashi N, Wajima T, Iwata S, et al. Direct identification of Streptococcus agalactiae and capsular type by realtime PCR in vaginal swabs from pregnant women. J Infect Chemother 2015;21:34-8. 6. Jones N, Bohnsack JF, Takahashi S, Oliver KA, Chan MS, Kunst F, et al. Multilocus sequence typing system for group B streptococcus. J Clin Microbiol 2003;41:2530-6. 7. Hansen SM, Uldbjerg N, Kilian M, Sorensen UB. Dynamics of Streptococcus agalactiae colonization in women during and after pregnancy and in their infants. J Clin Microbiol 2004;42:83-9.
8. Le Doare K, Jarju S, Darboe S, Warburton F, Gorringe A, Heath PT, et al. Risk factors for Group B Streptococcus colonisation and disease in Gambian women and their infants. J Infect 2016;72:283-94. 9. Berardi A, Rossi C, Creti R, Gherardi G, Venturelli C, Rumpianesi F, et al. Group B Streptococcal colonization in 160 mother-baby pairs: a prospective cohort study. J Pediatr 2013;163:1099-104, e1. 10. Baker CJ, Barrett FF. Transmission of group B streptococci among parturient women and their neonates. J Pediatr 1973;83:91925. 11. Carl MA, Ndao IM, Springman AC, Manning SD, Johnson JR, Johnston BD, et al. Sepsis from the gut: the enteric habitat of bacteria that cause late-onset neonatal bloodstream infections. Clin Infect Dis 2014;58:12118. 12. Matsubara K, Hoshina K, Suzuki Y. Early-onset and late-onset group B streptococcal disease in Japan: a nationwide surveillance study, 20042010. Int J Infect Dis 2013;17:e379-84. 13. Morozumi M, Wajima T, Kuwata Y, Chiba N, Sunaoshi K, Sugita K, et al. Associations between capsular serotype, multilocus sequence type, and macrolide resistance in Streptococcus agalactiae isolates from Japanese infants with invasive infections. Epidemiol Infect 2014;142:8129. 14. Chang B, Wada A, Hosoya M, Oishi T, Ishiwada N, Oda M. Characteristics of group B streptococcus isolated from infants with invasive infections: a population-based study in Japan. Jpn J Infect Dis 2014;35660. 15. Edmond KM, Kortsalioudaki C, Scott S, Schrag SJ, Zaidi AK, Cousens S, et al. Group B streptococcal disease in infants aged younger than 3 months: systematic review and meta-analysis. Lancet 2012;379:547-56. 16. Puopolo KM, Madoff LC. Type IV neonatal early-onset Group B Streptococcal disease in a United States hospital. J Clin Microbiol 2007;45:13602. 17. Heath PT. Status of vaccine research and development of vaccines for GBS. Vaccine 2017;34:2876-9.
Effects of Intrapartum Antibiotic Prophylaxis on Neonatal Acquisition of Group B Streptococci
173
THE JOURNAL OF PEDIATRICS • www.jpeds.com
Volume 190
Figure 1. Protocol of real-time PCR and culture for identification of GBS in samples collected from neonates.
Table III. Maternal and neonatal characteristics with respect to neonatal GBS positivity at 1 month after birth Characteristics
GBS-positive GBS-negative P n = 57 (%) n = 673 (%) value
Neonates Boys, n Born at full term, n (%) Birth weight, mean ± SD, g
30 (52.6) 51 (89.5) 2852 ± 565
367 (54.5) 614 (91.2) 2898 ± 454
.782 .654 .638
Mothers Vaginal delivery, n (%) Maternal age, mean ± SD, y Maternal GBS colonization, n (%) One or more prior live births, n (%) Only breastfeeding, n (%) Maternal mastitis, n (%)
46 (80.7) 32.7 ± 5.3 19 (33.3) 25 (43.9) 18 (31.6) 1 (1.8)
504 (74.9) 32.5 ± 4.4 88 (13.1) 316 (47.0) 245 (36.4) 5 (0.7)
.328 .518 <.001 .653 .481 .323
173.e1
Toyofuku et al
ORIGINAL ARTICLES
Effects of Intrapartum Antibiotic Prophylaxis on Neonatal Acquisition of Group B Streptococci Meiwa Toyofuku, MD1,2, Miyuki Morozumi, PhD2, Mariko Hida, MD, PhD1, Yoshitake Satoh, MD, PhD3, Hiroshi Sakata, MD, PhD4, Hiroyuki Shiro, MD, PhD1, Kimiko Ubukata, PhD2, Mitsuru Murata, MD, PhD5, and Satoshi Iwata, MD, PhD2 Objectives To assess the incidence of colonization with group B streptococci (GBS) among neonates as influenced by maternal GBS carriage and intrapartum antibiotic prophylaxis (IAP). Study design Between October 2014 and May 2015, nasopharyngeal and rectal swab samples were collected from 730 neonates at 1 week and 1 month after birth. GBS and capsular serotype were identified by real-time polymerase chain reaction and by culture. IAP at delivery was determined retrospectively from hospital records. Results Sixty-four neonates (8.8%) were GBS-positive by real-time polymerase chain reaction and culture. Among neonates born to mothers who were GBS carriers (n = 107), 94.4% (101/107) had maternal IAP; 19.6% nonetheless were GBS-positive, compared with 6.5% of neonates born to noncarrier mothers (P < .01). Among neonates born to mothers receiving IAP, more were positive only at 1 month of age than at both 1 week and 1 month. The frequency of GBS in neonates born to mothers receiving IAP was significantly lower than that in neonates born to mothers not receiving IAP (P < .05). Capsular serotypes V (25%) and III (23.4%) were common, followed by Ib (15.6%), Ia (14.1%), II (7.8%), IV (6.3%), nontypeable (4.7%), and VI and VIII (each 1.6%). Conclusions Delayed colonization with GBS occurs in infants born to GBS carrier mothers receiving IAP. GBS should be considered in all infants at 1 month after birth with signs of infection. (J Pediatr 2017;190:169-73).
G
roup B streptococcus (GBS) is a leading cause of severe infections in early infancy. These infections include earlyonset disease (EOD), occurring within 6 days and late-onset disease (LOD), occurring from 7 to 89 days. Among pregnant women, the GBS carrier rate was estimated to be near 13% (range, 12%-22%).1 In approximately one-half of carriers, GBS is transmitted to the infant, with invasive GBS disease developing in 1%-2%. To prevent EOD due to GBS, a guideline was introduced by the US Centers for Disease Control and Prevention and the American College of Obstetricians and Gynecologists in 1996. Universal screening for maternal GBS colonization and intrapartum antibiotic prophylaxis (IAP) strategies implemented according to the guideline have reduced the incidence of EOD but had no impact on LOD.2,3 In Japan, a revised guideline for obstetric practice published in 2014 recommended universal GBS screening by conventional culture for all pregnant women at 33-37 weeks of gestation.4 IAP during labor or after premature rupture of membranes was recommended for women with any of the following characteristics: GBS culture positivity, GBS infection during infancy in previous offspring, or incidental detection of GBS in a urine culture during the current pregnancy. If GBS status in the current pregnancy is unknown, IAP should be given to mothers with symptoms at delivery that might be attributable to GBS. In this study, we investigated prevalence of GBS colonization in neonates examined at 1 week and 1 month after birth, influences of IAP and GBS carriage in the mother, and capsular serotype and sequence type (ST) of GBS isolates from neonates.
Methods We conducted a prospective longitudinal cohort study at 4 hospitals, which were located in Hokkaido, Gunma, and Kanagawa prefectures and in Tokyo. The study was performed from October 2014 to May 2015. Nasopharyngeal and rectal swabs
EOD GBS IAP LOD MLST PCR ST
Early-onset disease Group B streptococcus Intrapartum antibiotic prophylaxis Late-onset disease Multilocus sequence typing Polymerase chain reaction Sequence type
From the 1Division of Pediatrics, Yokohama Rosai Hospital, Yokohama; 2Department of Infectious Diseases, Keio University School of Medicine, Tokyo; 3Division of Pediatrics, Ota Memorial Hospital, Ota; 4Division of Pediatrics, Asahikawa Kosei Hospital, Asahikawa; and 5 Department of Laboratory Medicine, Keio University School of Medicine, Tokyo, Japan M.T. received a grant from the Japanese Society for Pediatric Infectious Diseases. The other authors declare no conflicts of interest. 0022-3476/$ - see front matter. © 2017 Elsevier Inc. All rights reserved. https://doi.org10.1016/j.jpeds.2017.07.039
169
THE JOURNAL OF PEDIATRICS • www.jpeds.com were collected from all eligible infants at 1 week and 1 month after birth and transported immediately to our laboratory. All samples were tested to identify GBS by amplification of the dltS gene, which encodes a histidine kinase, and also to identify capsular types Ia, Ib, and III using real-time polymerase chain reaction (PCR) as developed by Morozumi et al.5 Correlation between results of real-time PCR and bacterial culture also was examined. As shown in Figure 1 (available at www.jpeds.com), samples were collected via cotton-tipped swabs (BBL CultureSwab Plus, Copan, Italy). Immediately after receipt, swab samples were suspended in 0.5 mL of Todd-Hewitt broth (Becton, Dickinson, Franklin Lakes, New Jersey) and centrifuged at 2000g for 5 minutes at 4°C to collect bacterial cells. After the supernatant was discarded, 55 µL of pellet was stirred and used for direct DNA extraction (50 µL) and for culture (5 µL). Each sample of 50 µL was placed in 45 µL of a lysis solution that included SimplePrep reagent for DNA (Takara Bio, Shiga, Japan) and 2 U of mutanolysin (Sigma Aldrich, St. Louis, Missouri). The lytic reaction was carried out for 10 minutes at 37°C, followed by 3 minutes at 95°C. The lysate was added to each of the tubes containing PCR mixtures for the following gene identifications: the dltS gene, which encodes a histidine kinase specific to GBS; and capsular types Ia, Ib, and III. Total volume (30 µL) of the PCR mixture included 20 pmol of each primer, 25 pmol of each probe, 2× Multiplex Powermix (Bio-Rad, Hercules, California), and DNase- and RNase-free distilled water. DNA amplification was carried out for 45 cycles as follows: 95°C for 10 seconds, 50°C for 30 seconds, and 72°C for 20 seconds. The remaining 5 µL of each sample was inoculated on sheep blood agar (Nippon Becton, Dickinson, Tokyo, Japan), which then was incubated at 37°C for 20 hours in an atmosphere containing 5% CO2. Colonies of b-hemolytic streptococci grown on the blood agar plates underwent identification of capsular type (Ia, Ib, II, III, IV, V, VI, VII, and VIII) and the dltS gene by real-time PCR. For multilocus sequence typing (MLST) analysis, primer sets corresponding to 7 housekeeping genes (adhP, atr, glnA, glcK, pheS, sdhA, and tkt) used for MLST were constructed with reference to the MLST Web site (http://pubmlst.org/sagalactiae/). MLST was applied to the sequences for these 7 genes accord-
Volume 190
•
November 2017
ing to previously described methods,6 with alleles and ST assignments determined using the Streptococcus agalactiae MLST database. Statistical Analyses Statistical analyses were performed with Ekuseru-Toukei 2012 software for statistics (Social Survey Research information, Tokyo, Japan). The Mann-Whitney U test, the c2 test, Fisher exact test, the Wilcoxon rank sum test, or univariate logistic regression analysis was used as appropriate. A P value < .05 was considered to indicate statistical significance. Field workers explained the purpose of the study to mothers of eligible infants, and each mother signed an informed consent form just before discharge or at 1 week after delivery. The study was approved by the Yokohama Rosai Hospital Ethics Committee (approval number 26-51).
Results Figure 2 is a flowchart showing transmission of GBS among 730 neonates according to GBS status of their mothers as determined by prenatal screening. GBS screening via bacterial cultures was performed at 33-37 weeks of gestation for 710 mothers (97.3%) receiving care at institutions participating in the study, and 107 mothers (14.7%) proved to be GBS carriers. Twenty mothers (2.7%) did not undergo screening. GBS-positive neonates numbered 64, including 46 positive by both real-time PCR and culture and 18 positive only by real-time PCR. Twenty-one neonates born to the 107 GBS carriers were positive for GBS (19.6%); among neonates born to the 603 noncarriers, 39 (6.5%) were positive for GBS. Frequency of GBS positivity in neonates differed significantly between carrier and noncarrier mothers (P < .001). Four neonates born to mothers without GBS screening were positive for GBS (20%). No infants developed invasive GBS infection during the study period. Relationships between time points of GBS positivity in neonates and maternal IAP are shown in Table I. GBS-positive neonates were divided into 3 groups: GBS persistence (n = 16, 25%), indicating GBS positivity at both 1 week and 1 month after birth; GBS clearance (n = 7, 10.9%),
Subjects Mother & infant (n = 730)
Maternal GBS (+) 14.7% (n = 107)
Infant GBS (+) 19.6% (n = 21)
Infant GBS (-) 80.4% (n = 86)
Maternal GBS (-) 82.6% (n = 603)
Infant GBS (+) 6.5% (n = 39)
Infant GBS (-) 93.5% (n = 564)
Maternal GBS unknown 2.7% (n = 20)
Infant GBS (+) 20.0% (n = 4)
Infant GBS (-) 80.0% (n = 16)
Figure 2. Flowchart showing presence or absence of GBS carriage in mothers and neonates. 170
Toyofuku et al
ORIGINAL ARTICLES
November 2017
Table I. Association between GBS positivity in neonates and maternal GBS colonization Mothers GBS(+) (n = 107) Neonates Persistence group, n (%)‡ Clearance group, n (%)‡ Delayed positivity group, n (%)‡ Total
GBS(−) (n = 603)
Screening (−) (n = 20)
n (%)
IAP* Yes (n = 101)
IAP No (n = 6)
IAP* Yes (n = 196)
IAP No (n = 407)
IAP* Yes (n = 17)
IAP† No (n = 3)
16 (25.0) 7 (10.9) 41 (64.1) 64
2 (10.5) 2 (10.5) 15 (78.9) 19
1 0 1 2
1 0 6 7
10 (31.3) 5 (15.6) 17 (53.1) 32
1 0 2 3
1 0 0 1
*Among mothers with IAP, cefazolin was given to 189; ampicillin to 107; and other agents to 15. †One woman was not given IAP despite preterm delivery. Two women were not given IAP, in accordance with the guideline. ‡Persistence group, positive for GBS at 1 wk and 1 mo after birth; clearance group, positive only at 1 wk after birth; Delayed positivity group, negative for GBS at 1 wk, but positive at 1 mo after birth.
indicating GBS positivity at 1 week but not at 1 month after birth; or delayed GBS positivity (n = 41, 64.1%), indicating GBS negativity at 1 week but positivity at 1 month after birth. Of the 107 GBS carrier mothers, 101 (94.4%) received IAP, among whom ampicillin was administered to 84 (83.2%), cefazolin to 14 (13.9%), and other agents to 3 (3%). Although IAP was given, 19 neonates (18.8%) born to GBS carrier mothers were positive for GBS after birth. Notably, most of these neonates, numbering 15 (78.9%), belonged to the delayedpositivity group. Of 603 pregnant women found to be GBS noncarriers, 196 (32.5%) received intravenous antibiotics during labor; among their infants, 6 of the 7 GBS-positive neonates were in the delayed-positivity group. Among GBS-positive neonates whose mothers received no antibiotic agents, 10 (31.3%) had GBS persistence, 5 (15.6%) had clearance, and 17 (53.1%) had delayed positivity. Generally, delayed positivity was significantly more prevalent in neonates born to mothers who received IAP (P = .04) than in neonates born to mothers not receiving IAP. In the persistence group, a significant difference was not evident between infants of mothers with and without IAP (P = .08). Table II shows a correlation between threshold cycle values derived from real-time PCR assays in neonates and implementation of IAP. Neonates born to mothers receiving IAP showed a significantly lower GBS bacterial load than neonates born to mothers not receiving IAP (P = .04).
Table II. Threshold cycle values derived from realtime PCR in neonates according to IAP Mothers (n = 710) Threshold cycle and number of bacterial cells <24 cycles (>5 × 104 cells/sample) 25-28 cycles (103 –5×104 cells/sample) >29 cycles (<103 cells/sample) Total
IAP Yes (n = 297)
IAP No (n = 413)
Total
2 (18.2)
9 (81.8)
11
8 (44.4)
10 (55.6)
18
19 (54.3)
16 (45.7)
35
29
35
64
P = .04. P values were calculated with the Wilcoxon rank sum test.
Maternal and neonatal characteristics are analyzed for relationships to neonatal GBS positivity at 1 month after birth in Table III (available at www.jpeds.com). Only maternal GBS carrier status showed a significant difference between GBS-positive neonates (n = 57) and GBS-negative neonates after birth (n = 673) at 1 month (P < .001). Univariate logistic regression analysis of these characteristics identified maternal GBS colonization as a risk factor associated with neonatal GBS positivity (OR 3.6; 95% CI 2.0-6.6). Capsular serotypes for all GBS isolates were identified by real-time PCR. As shown in Table IV, capsular serotypes V (n = 16, 25%), III (n = 15, 23.4%), Ib (n = 10, 15.6%), and Ia (n = 9, 14.1%) were predominant, followed by II (n = 5), IV (n = 4), nontypeable (n = 3), VI and VIII (n = 1, respectively). In addition, the most common serotype in the persistence group (n = 16), clearance group (n = 7) and delayedpositivity group (n = 41) was Ib (31.3%), III (57.1%), and V (29.3%), respectively. Forty-five GBS strains isolated from 64 GBS-positive samples by bacterial culture in neonates (70.3%) were analyzed for capsular serotype and MLST. Although 10 STs were found, ST1 (28.9%), ST335 (24.4%), and ST10 (15.6%) accounted for 68.9% of all 45 isolates. Most isolates with capsular serotype III were ST335, which is derived from ST19. ST17, linked with high virulence, was found in only 1 strain.
Discussion In this prospective cohort study of neonatal GBS colonization following the 2014 introduction of the current guidelines for obstetric practice in Japan, combining real-time PCR with culture methods, we found an overall prevalence of GBS colonization within 1 month after birth of 8.8%. Notably, the positivity rate in neonates born to GBS carrier mothers, almost all of whom received IAP, was significantly greater (19.6%) than positivity in neonates whose mothers were not carriers (6.5%). Positivity rates in our Japanese infants were considerably lower than those among infants in Denmark (11.3%),7 Gambia (24.8%),8 and Italy (21.9%).9 We previously estimated GBS carriers among pregnant women to be 15.7% by real-time PCR and 12.6% by bacterial culture.5 We believe that high compliance with the obstetric guideline in Japan most likely was
Effects of Intrapartum Antibiotic Prophylaxis on Neonatal Acquisition of Group B Streptococci
171
THE JOURNAL OF PEDIATRICS • www.jpeds.com
Volume 190
Table IV. Relationships between capsular type and ST Capsular type STs* ST1 ST10 ST17 ST19 ST23 ST26 ST130 ST144 ST196 ST335† Unknown‡ Total (%)
Total, n (%) 13 7 1 1 4 1 1 2 4 11 19
(20.3) (10.9) (1.6) (1.6) (6.3) (1.6) (1.6) (3.1) (6.3) (17.2) (29.7) 64
Ia
Ib
II
III
IV
4
V
VI
8
1
VIII
NT
1 1 (1.6)
3 3 (4.7)
7 1 1 3
1 1 1
2 4 4 9 (14.1)
3 10 (15.6)
1 5 (7.8)
11 1 15 (23.4)
4 (6.3)
6 16 (25.0)
1 (1.6)
*Only GBS isolates by culture were tested. †ST335 was derived from ST19. ‡Capsular type in 16 cases was identified by real-time PCR.
responsible for lower neonatal GBS positivity than in other countries. Neonatal GBS colonization of the pharynx, intestine, or umbilical cord is well known to precede development of invasive GBS infection.10,11 Reported incidence of GBS disease was 0.57 per 1000 live births in Europe and 0.67 in the Americas.1 According to a previous report,12 incidence of invasive GBS disease in Japan was much lower, 0.18 per 1000 live births, which may be attributable to a low rate of GBS maternal transmission to infants. Interestingly, delayed positivity was significantly more frequent than persistent positivity, especially in neonates born to mothers receiving IAP. The density of GBS colonization in neonates with delayed positivity was estimated to be relatively low by real-time PCR. In contrast, persistent positivity was more frequent among neonates born to mothers without IAP compared with those born to mothers with IAP. We suspect that after IAP was initiated, density of GBS colonization of mothers decreased temporarily, but the organism was not eradicated. These small numbers of GBS, transmitted from mother to newborn, may then have multiplied gradually in the respiratory or intestinal tract after escaping detection at 1 week after birth. Accordingly, IAP strategies could reduce incidence of EOD but lack impact on incidence of LOD. Pregnant women carrying GBS therefore remain an important risk factor for neonatal invasive GBS infection. Neonatal GBS infection also could occur following transmission from the mother by a route unrelated to delivery, or by horizontal transfer from another person such as a family member after discharge from the hospital. In our study, the predominant capsular serotypes of GBS carried by neonates were types V and III. Serotype III is most common among isolates from infants with invasive GBS.12-14 Relative capsular serotype prevalence is similar to that previously reported worldwide.15 Among GBS isolates from Japanese maternal carriers, the predominant capsular serotype was type Ib, followed by types V and III5; other countries have differed in maternal carrier serotype prevalence.8,9 In Japan, se172
rotype V has shown little involvement in invasive GBS disease during early infancy, whereas in the US serotype V is increasing in prevalence both in pregnancy and in invasive disease during early infancy.16 Importantly, serotype patterns can shift. ST17 GBS, which has high virulence and belongs to capsular serotype III, accounted for 50% of invasive GBS isolates in Japan,13 similar to reports from the Netherlands.3 In our study, only one ST17 GBS was identified. Our study has several limitations. First, US Centers for Disease Control and Prevention guidelines indicate that IAP administration beginning at least 4 hours before delivery has been found highly effective.2 Regrettably, detailed information about when IAP was administered relative to delivery was unavailable for our study. Second, bacterial culture of maternal samples was carried out without use of a selective medium for GBS isolation at each institution. Therefore, false-negative cases may have been included among “noncarrier” mothers. Application of real-time PCR or conventional PCR with high sensitivity and specificity is expected to enhance accuracy of maternal testing. Finally, we could not identify capsular serotypes in GBS-colonized mothers, leaving open the possibility that instances of GBS colonization of infants may not all have resulted from vertical transmission at birth. Uniform testing that combines real-time PCR with bacterial culture for both pregnant women and neonates will be necessary to further address this question. However, such testing cannot distinguish transmission at delivery from subsequent transmission involving ordinary maternal contact. The ability of IAP to prevent LOD due to GBS may be limited even with high-precision testing for both pregnant women and neonates. Further studies should determine whether surveillance microbiologic assessment at 1 month after birth and augmented hygiene can reduce risk of LOD. GBS vaccination of pregnant women represents an alternative way to protect infants from invasive GBS disease. This strategy is under development, with the leading candidate for maternal vaccination being a trivalent GBS vaccine (CRM197-conjugated capsular polysaccharides of GBS serotype Ia, Ib, and III).17 ■ Toyofuku et al
ORIGINAL ARTICLES
November 2017 We thank Atsuo Sato, MD, Masaki Nakayama, MD, Miyuu Matsui, MD, Tomoko Furutani, MD, and Kumiko Tago, MD for collecting data. Submitted for publication Apr 22, 2017; last revision received Jun 22, 2017; accepted Jul 19, 2017
References 1. Kobayashi M, Vekemans J, Baker CJ, Ratner AJ, Le Doare K, Schrag SJ. Group B Streptococcus vaccine development: present status and future considerations, with emphasis on perspectives for low and middle income countries. F1000Res 2016;5:2355. 2. Verani JR, McGee L, Schrag SJ. Prevention of perinatal group B streptococcal disease—revised guidelines from CDC, 2010. MMWR Recomm Rep 2010;59:1-36. 3. Van Dyke MK, Phares CR, Lynfield R, Thomas AR, Arnola KE, Craig AS, et al. Evaluation of universal antenatal screening for Group B Streptococcus. N Engl J Med 2009;360:2626-36. 4. Minakami H, Maeda T, Fujii T, Hamada H, Iitsuka Y, Itakura A, et al. Guidelines for obstetrical practice in Japan: Japan Society of Obstetrics and Gynecology (JSOG) and Japan Association of Obstetricians and Gynecologists (JAOG) 2014 edition. J Obstet Gynaecol Res 2014;40:146999. 5. Morozumi M, Chiba N, Igarashi Y, Mitsuhashi N, Wajima T, Iwata S, et al. Direct identification of Streptococcus agalactiae and capsular type by realtime PCR in vaginal swabs from pregnant women. J Infect Chemother 2015;21:34-8. 6. Jones N, Bohnsack JF, Takahashi S, Oliver KA, Chan MS, Kunst F, et al. Multilocus sequence typing system for group B streptococcus. J Clin Microbiol 2003;41:2530-6. 7. Hansen SM, Uldbjerg N, Kilian M, Sorensen UB. Dynamics of Streptococcus agalactiae colonization in women during and after pregnancy and in their infants. J Clin Microbiol 2004;42:83-9.
8. Le Doare K, Jarju S, Darboe S, Warburton F, Gorringe A, Heath PT, et al. Risk factors for Group B Streptococcus colonisation and disease in Gambian women and their infants. J Infect 2016;72:283-94. 9. Berardi A, Rossi C, Creti R, Gherardi G, Venturelli C, Rumpianesi F, et al. Group B Streptococcal colonization in 160 mother-baby pairs: a prospective cohort study. J Pediatr 2013;163:1099-104, e1. 10. Baker CJ, Barrett FF. Transmission of group B streptococci among parturient women and their neonates. J Pediatr 1973;83:91925. 11. Carl MA, Ndao IM, Springman AC, Manning SD, Johnson JR, Johnston BD, et al. Sepsis from the gut: the enteric habitat of bacteria that cause late-onset neonatal bloodstream infections. Clin Infect Dis 2014;58:12118. 12. Matsubara K, Hoshina K, Suzuki Y. Early-onset and late-onset group B streptococcal disease in Japan: a nationwide surveillance study, 20042010. Int J Infect Dis 2013;17:e379-84. 13. Morozumi M, Wajima T, Kuwata Y, Chiba N, Sunaoshi K, Sugita K, et al. Associations between capsular serotype, multilocus sequence type, and macrolide resistance in Streptococcus agalactiae isolates from Japanese infants with invasive infections. Epidemiol Infect 2014;142:8129. 14. Chang B, Wada A, Hosoya M, Oishi T, Ishiwada N, Oda M. Characteristics of group B streptococcus isolated from infants with invasive infections: a population-based study in Japan. Jpn J Infect Dis 2014;35660. 15. Edmond KM, Kortsalioudaki C, Scott S, Schrag SJ, Zaidi AK, Cousens S, et al. Group B streptococcal disease in infants aged younger than 3 months: systematic review and meta-analysis. Lancet 2012;379:547-56. 16. Puopolo KM, Madoff LC. Type IV neonatal early-onset Group B Streptococcal disease in a United States hospital. J Clin Microbiol 2007;45:13602. 17. Heath PT. Status of vaccine research and development of vaccines for GBS. Vaccine 2017;34:2876-9.
Effects of Intrapartum Antibiotic Prophylaxis on Neonatal Acquisition of Group B Streptococci
173
THE JOURNAL OF PEDIATRICS • www.jpeds.com
Volume 190
Figure 1. Protocol of real-time PCR and culture for identification of GBS in samples collected from neonates.
Table III. Maternal and neonatal characteristics with respect to neonatal GBS positivity at 1 month after birth Characteristics
GBS-positive GBS-negative P n = 57 (%) n = 673 (%) value
Neonates Boys, n Born at full term, n (%) Birth weight, mean ± SD, g
30 (52.6) 51 (89.5) 2852 ± 565
367 (54.5) 614 (91.2) 2898 ± 454
.782 .654 .638
Mothers Vaginal delivery, n (%) Maternal age, mean ± SD, y Maternal GBS colonization, n (%) One or more prior live births, n (%) Only breastfeeding, n (%) Maternal mastitis, n (%)
46 (80.7) 32.7 ± 5.3 19 (33.3) 25 (43.9) 18 (31.6) 1 (1.8)
504 (74.9) 32.5 ± 4.4 88 (13.1) 316 (47.0) 245 (36.4) 5 (0.7)
.328 .518 <.001 .653 .481 .323
173.e1
Toyofuku et al