Group B streptococcus (GBS) is the leading cause of neonatal sepsis and meningitis in the United States, with 1,000–1,200 cases occurring each year among neonates younger than 1 week old.1–5 Approximately 25% of pregnant women carry GBS in the vagina or rectum and are at risk of intrapartum transmission to their newborns.6 Intravenous intrapartum antibiotic prophylaxis reduces the risk of invasive GBS disease in the first week of life (early-onset disease).7 Widespread use of intrapartum prophylaxis has reduced the incidence of early-onset GBS disease by approximately 80% since the early 1990s.5,7
Prenatal screening of all pregnant women for GBS colonization has been recommended since 2002,8 and GBS-colonized women should receive intrapartum prophylaxis.1,8 Guidelines from the Centers for Disease Control and Prevention and key partners provide detailed recommendations regarding the timing and technique for prenatal screening, laboratory methods for processing prenatal screening specimens, the use of intrapartum prophylaxis, and the management of newborns at risk for early-onset GBS disease.1,8 A multicenter study in 2003–2004 found that 85% of pregnant women underwent prenatal GBS screening and 85% of women with an indication for intrapartum prophylaxis received it.9 However, effective prevention of GBS disease requires correct implementation of recommended practices at multiple points of care, including physicians' offices, prenatal clinics, microbiology laboratories, labor and delivery wards, and newborn nurseries.
In recent years, the incidence of early-onset GBS disease has remained approximately 0.3 cases per 1,000 live births,2–5 a rate consistent with predictions based on prevention implementation in the late 1990s.10 Novel approaches such as maternal GBS vaccination or rapid, accurate point-of-care intrapartum tests for GBS colonization and susceptibility may be necessary for additional decline. However, until such become available, it may be that further reductions could be achieved through better adherence to current recommendations.
We reviewed early-onset GBS cases among term and preterm neonates from 2008–2009 from a population of more than 456,000 annual live births to identify errors in implementation of the prevention recommendations and to describe how much of the remaining burden of disease would be preventable under optimal prenatal screening and intrapartum prophylaxis.
MATERIALS AND METHODS
We conducted a retrospective case series of invasive early-onset GBS disease, defined by isolation of GBS from a normally sterile site (eg, blood, cerebrospinal fluid) in a liveborn neonate aged younger than 7 days. Patients were identified through the Active Bacterial Core surveillance system, a component of the Emerging Infections Program Network that conducts active, laboratory-based and population-based surveillance in selected counties in 10 states (active Bacterial Core surveillance site locations are in California, Colorado, Connecticut, Georgia, Maryland, Minnesota, New Mexico, New York, Oregon, and Tennessee) and covers approximately 10% of all births in the United States. All early-onset GBS disease cases in 2008–2009 were included.
The protocol was determined by the Centers for Disease Control and Prevention to be a program evaluation and, therefore, informed consent was not required. The local institutional review board at each site also reviewed the protocol and waived the requirement for informed consent.
Trained personnel used standardized forms to abstract information for each case-patient. Labor and delivery records were reviewed to gather relevant prenatal and intrapartum information. For mothers with any prenatal care, data were abstracted from medical records at all sites where care was received during that pregnancy. Variables included the date and time of all GBS screens, urine culture results, and prior history of neonates with GBS disease. For mothers who underwent prenatal GBS screening, we sought to interview the prenatal care provider who collected the mother's specimen using a standardized questionnaire regarding their typical practices for collection of specimens for prenatal GBS screens. We also contacted the laboratory where the screen was processed to request the records regarding that specimen as well as standard operating procedure documents for information on how the specimen was processed and tested for GBS.
Prevention implementation errors were defined as errors in prenatal screening, laboratory methods, communication of relevant results, or the use of intrapartum prophylaxis when indicated (Table 1). Lack of prenatal screening was not considered an error for mothers of case-patients born before 37 weeks of gestation; however, if a screen was performed, we evaluated specimen collection and laboratory methods regardless of gestational age. Information in labor and delivery records was used to determine whether GBS prophylaxis was indicated. Antibiotics were discontinued 24 hours or more before delivery, and those with unknown timing of administration were not considered intrapartum prophylaxis. Antibiotic agents used for intrapartum prophylaxis were judged correct or incorrect based on the 2002 GBS guidelines; the only change under the 2010 guidelines revision was the elimination of erythromycin, which was rarely used and recommended only for a small subset of penicillin-allergic women. For a lack of intrapartum prophylaxis to be considered an implementation error, the mother had to be admitted 1 hour or more before delivery; similarly, for prophylaxis to be considered inadequate duration (shorter than 4 hours), the mother had to be admitted 5 hours or more before delivery. Case-patients born before 37 weeks of gestation were classified as preterm.
To estimate the number of cases that could have been prevented under optimal prevention implementation, we assumed that the sensitivity of prenatal screening was 87% for a combined vaginal–rectal specimen within the recommended timeframe and using correct laboratory methods.11 We assumed that the effectiveness of 4 or more hours of a β-lactam antibiotic was 90% and 45% for durations shorter than 4 hours.12 Although cefazolin was considered an incorrect antibiotic choice for women with no penicillin allergy, for estimating the preventable fraction, it was considered equivalent to penicillin and ampicillin based on pharmacokinetic data.13 Implementation errors that were unlikely to affect the outcome were not factored into the preventable fraction. Thus, we assumed no potential reduction in cases under the following scenarios regardless of whether any errors were identified: mother received optimal intrapartum prophylaxis; mother hospitalized for less than 1 hour before delivery; mother hospitalized 1 or more hours but fewer than 5 hours before delivery and received the appropriate antibiotic for shorter than 4 hours; prophylaxis not indicated but correct prenatal screening and laboratory practices were performed; and prophylaxis not indicated but no prenatal care received. Where key data such as health care provider specimen collection technique were missing, we used a range for the potential reduction from 0% (assuming no implementation errors for missing data) to the maximum possible reduction (assuming all implementation errors for missing data). Analyses were conducted using SAS Enterprise Guide 4.2.
In 2008–2009, a total of 312 early-onset cases were detected (California: 12; Colorado: 16; Connecticut: 16; Georgia: 105; Maryland: 43; Minnesota: 40; New Mexico: 16; New York: nine; Oregon: 13; Tennessee: 41). Medical records were available for 309 (99.0%) cases, including labor and delivery records for 308 (98.7%) and prenatal clinic records for 270 (86.8%) (Fig. 1). Among case-patients with available data, 222 (71.8%) were delivered at term, 85 (27.5%) were preterm, and two (0.6%) had unknown gestational age. Overall, among 303 mothers who received prenatal care (including one with no labor and delivery records), 298 (98.3%) had some prenatal care data available, and 232 (77.9%) of those underwent prenatal GBS screening. Among those screened, prenatal provider interviews were completed for 165 (71.1%) and laboratory information was obtained for 121 (52.2%). Results were available for 231 (99.6%) and were negative for 170 (73.6%) (Fig. 2).
At least one implementation error (Table 1) was noted among 179 (57.9%) cases, including 126 (40.8%) with one error, 43 (13.9%) with two, and 10 (3.2%) with three errors (Table 2). The frequency of any implementation error among term patients was 57.2% (127 of 222) and 61.2% (52 of 85) among preterm patients (P=.53).
Errors in prenatal screening were observed in 80 (36.0%) term and seven (8.2%) preterm patients. The most frequent error was incorrect specimen collection; among 165 cases with available data on the anatomical site the health care provider most commonly swabbed, 55 (33.3%) were sites other than vaginal–rectal (33=vaginal–perianal, eight=vaginal–perineal, eight=vaginal only, three=cervical–vaginal, two=perianal only, one=perineal only). Of 209 mothers who delivered at term and underwent prenatal screening, 201 (96.2%) had data available on the gestational age at screening; 23 (11.4%) of those were screened before 35 weeks of gestation. Prenatal screens were unnecessarily performed on 23 (63.9% of 36) women with either GBS bacteriuria (n=22) or a previous neonate with GBS disease (n=1).
Errors in laboratory methods were documented in 38 cases (12.3% of all cases and 30.8% of cases with available laboratory data, n=121). Key information was frequently unavailable from worksheets and standard operating procedures. Data on prenatal specimen enrichment were available for 55 (45.4%) of those cases with laboratory records; 15 (27.3%) had specimens that did not undergo any enrichment. The actual incubation time for the enrichment broth, or the duration specified in the standard operating procedure, was available for 66 (54.5%) cases with laboratory records; 20 (30.3%) had a duration less than the recommended 18 hours. Among 170 cases with a negative prenatal screening result, laboratory records were obtained for 98 (57.6%); 70 of 98 (71.4%) had sufficient information to assess at least one laboratory practice, and 32 of 70 (45.7%) had evidence of errors.
Communication errors regarding GBS status affected 19 (6.2%) case-patients, including eight in whom GBS bacteriuria was noted during prenatal care but not in the labor and delivery record and 11 in whom a positive prenatal screening result was not conveyed to the labor and delivery record.
Intrapartum prophylaxis was indicated for 136 (44.0%) mothers of case-patients; 52 (38.2%) received no prophylaxis despite being in the hospital for 1 hour or longer before delivery (range 1.2 hours to 48 days), and eight (5.9%) received prophylaxis for less than 4 hours despite being in the hospital for 5 hours or more before delivery (range 8.5 hours to 11 days). Among 51 women with an indication for prophylaxis who received it, 19 (37.2%) received an incorrect antibiotic, including nine who received clindamycin (with no severe penicillin allergy or without antimicrobial susceptibility testing), nine who received cefazolin (with no penicillin allergy), and one who received vancomycin (with no severe penicillin allergy). Intrapartum prophylaxis errors were more commonly observed among preterm patients than among term (54.1% compared with 13.5%, P<.001).
For 15 cases in which optimal intrapartum prophylaxis was administered, we assumed no reduction in the number of observed cases. Among 122 patients in which prophylaxis was indicated but not received or was suboptimal, we estimated a potential decline to 71 to 72 cases (41–42% reduction) (Table 3). Among 167 women for whom indications for prophylaxis were not identified, 155 (92.8%) were screened prenatally and had a negative result; only five (3.2%) had correct prenatal screen timing, specimen collection, and laboratory methods. No potential reduction was assumed for those five cases and for two cases in which the mother received no prenatal care. For the remaining 160 cases in which no indication for prophylaxis was identified, we estimated a potential reduction to 34–130 cases (19–79% decline). Among all 309 cases, 50–51 cases (16%) might have been averted through improved use of intrapartum prophylaxis, and 22–118 (7–38%) might have been averted through improved prenatal screening (including specimen collection and laboratory methods). Overall, with optimal implementation the observed cases might have been reduced by 26–59%. The national early-onset GBS incidence rate of 0.3 cases per 1,000 live births in recent years2,3 might therefore be reduced to 0.12–0.22 cases per 1,000 live births.
Early-onset GBS disease incidence fell dramatically during the 1990s as a result of widespread use of intrapartum antibiotic prophylaxis and declined further following the 2002 recommendation for screening all pregnant women for GBS colonization.14 In recent years, the incidence plateaued, suggesting the limits of current prevention strategies might have been reached. However, this study demonstrates the potential for meaningful further disease reductions.
Improved intrapartum prophylaxis implementation, particularly among preterm deliveries, offers the most direct route to accomplish further reductions in early-onset GBS disease. Although some cases with suboptimal or no intrapartum prophylaxis were not preventable, most mothers of case-patients for whom prophylaxis was indicated were hospitalized long enough before delivery to receive adequate antibiotics. Incorrect choice of antibiotic was common for women with a penicillin allergy, a finding consistent with other studies of adherence to GBS prevention recommendations.9,14–16 Intrapartum prophylaxis errors were most common among preterm patients, a group at particularly high risk for illness and death resulting from GBS14 and where poor adherence to prophylaxis recommendations has been noted.9 Importantly, intrapartum prophylaxis is highly effective at preventing early-onset disease among preterm neonates.12
Frequent errors occurred in the collection of specimens and in the laboratory methods used to process them. However, the link between those errors and the persistent burden of early-onset GBS disease is not as clear. Prenatal screening results may be positive despite suboptimal methods, and women with false-negative results may receive intrapartum antibiotics for other reasons (eg, chorioamnionitis). Furthermore, missing data, particularly for laboratory practices, greatly limited our ability to estimate the potential reduction in patients in whom prophylaxis was not indicated. Nonetheless, improved adherence to recommendations for prenatal screening specimen collection and laboratory testing would likely reduce the proportion of false-negative screens and lead to some decrease in early-onset disease burden.
Yet even with optimal implementation of all GBS prevention activities, some burden of early-onset disease would persist. Therefore, additional approaches to GBS prevention are needed. Molecular assays for GBS have raised the possibility of rapid intrapartum testing as a potential improvement over prenatal screening.15–19 However, suboptimal sensitivity of available assays on nonenriched specimens, test complexity, turnaround time, and inability to detect resistance to clindamycin currently limit their use.1 Among the mothers of case-patients in this study, more than half had no indication for GBS prophylaxis because they either had no prenatal screen or the prenatal screen yielded false-negative results. Such mothers might be detected by a highly sensitive, low-complexity, point-of-care test.
Maternal GBS vaccines have been shown to be safe and immunogenic20 and phase II clinical trials are currently underway.21 An efficacious GBS vaccine could help prevent both those early-onset cases which are unavoidable under current prevention strategies, and those in which lapses in implementation of prenatal screening or use of intrapartum prophylaxis have occurred. An additional benefit of maternal GBS vaccines is the potential to reduce late-onset GBS disease, the incidence of which has been unchanged by intrapartum antibiotic prophylaxis and is now similar to that of early-onset GBS disease in the United States.5,22 Vaccines would also have certain limitations, including lack of protection against serotypes not included in the vaccine and likely reduced effectiveness in preterm neonates, yet these novel tools, in conjunction with current approaches, could potentially reduce invasive neonatal GBS disease even further.
The study is subject to several limitations. Because of incomplete data particularly from laboratory records, it is likely that some implementation errors went undetected. In addition, some of the data sources, including health care providers' self-reported specimen collection techniques and laboratory standard operating procedures, were not specific to the case and may not reflect what occurred with the specimen collected from the case-patient’s mother. Finally, case-only data do not reflect prevention practices at the population level.
Prenatal care providers, labor and delivery staff, and laboratory personnel all have a critical role to play in effective perinatal GBS disease prevention. Resources available to improve prevention include a mobile application to provide patient-specific guidance in accordance with the 2010 GBS prevention guidelines (www.cdc.gov/groupbstrep/resources/multimedia.html) and sample standard operating procedure documents for GBS detection (www.cdc.gov/groupbstrep/lab/sops.html). Other GBS prevention tools such as improved intrapartum tests and maternal vaccines may open pathways to new approaches in the future. In the meantime the results of this study show that the maximum benefit of current GBS prevention strategies has not yet been reached and that provision of intrapartum prophylaxis when indicated should be prioritized.
1. Verani JR, McGee L, Schrag SJ; Division of Bacterial Diseases, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention (CDC). Prevention of perinatal group B streptococcal disease—revised guidelines from CDC, 2010. MMWR Recomm Rep 2010;59:1–36.
6. Schuchat A. Epidemiology of group B streptococcal disease in the United States: shifting paradigms. Clin Microbiol Rev 1998;11:497–513.
7. Schrag SJ, Zywicki S, Farley MM, Reingold AL, Harrison LH, Lefkowitz LB, et al.. Group B streptococcal disease in the era of intrapartum antibiotic prophylaxis. N Engl J Med 2000;342:15–20.
8. Schrag S, Gorwitz R, Fultz-Butts K, Schuchat A. Prevention of perinatal group B streptococcal disease. Revised guidelines from CDC. MMWR Recomm Rep 2002;51:1–22.
9. Van Dyke MK, Phares CR, Lynfield R, Thomas AR, Arnold KE, Craig AS, et al.. Evaluation of universal antenatal screening for group B streptococcus. N Engl J Med 2009;360:2626–36.
10. Schrag SJ, Zell ER, Lynfield R, Roome A, Arnold KE, Craig AS, et al.; Active Bacterial Core Surveillance Team. A population-based comparison of strategies to prevent early-onset group B streptococcal disease in neonates. N Engl J Med 2002;347:233–9.
11. Yancey MK, Schuchat A, Brown LK, Ventura VL, Markenson GR. The accuracy of late antenatal screening cultures in predicting genital group B streptococcal colonization at delivery. Obstet Gynecol 1996;88:811–5.
12. Fairlie T, Zell ER, Schrag S. Effectiveness of intrapartum antibiotic prophylaxis for prevention of early-onset group B streptococcal disease. Obstet Gynecol 2013;121:570–7.
13. Allegaert K, van Mieghem T, Verbesselt R, de Hoon J, Rayyan M, Devlieger R, et al.. Cefazolin pharmacokinetics in maternal plasma and amniotic fluid during pregnancy. Am J Obstet Gynecol 2009;200:170.e1–7.
14. Phares CR, Lynfield R, Farley MM, Mohle-Boetani J, Harrison LH, Petit S, et al.; Active Bacterial Core surveillance/Emerging Infections Program Network. Epidemiology of invasive group B streptococcal disease in the United States, 1999–2005. JAMA 2008;299:2056–65.
15. Daniels JP, Gray J, Pattison HM, Gray R, Hills RK, Khan KS, et al.; GBS Collaborative Group. Intrapartum tests for group B streptococcus: accuracy and acceptability of screening. BJOG 2011;118:257–65.
16. de Tejada BM, Pfister RE, Renzi G, Francois P, Irion O, Boulvain M, et al.. Intrapartum Group B streptococcus detection by rapid polymerase chain reaction assay for the prevention of neonatal sepsis. Clin Microbiol Infect 2011;17:1786–91.
17. Alfa MJ, Sepehri S, De Gagne P, Helawa M, Sandhu G, Harding GK. Real-time PCR assay provides reliable assessment of intrapartum carriage of group B Streptococcus. J Clin Microbiol 2010;48:3095–9.
18. Edwards RK, Novak-Weekley SM, Koty PP, Davis T, Leeds LJ, Jordan JA. Rapid group B streptococci screening using a real-time polymerase chain reaction assay. Obstet Gynecol 2008;111:1335–41.
19. Honest H, Sharma S, Khan KS. Rapid tests for group B Streptococcus colonization in laboring women: a systematic review. Pediatrics 2006;117:1055–66.
20. Edwards MS. Group B streptococcal conjugate vaccine: a timely concept for which the time has come. Hum Vaccines 2008;4:444–8.
21. Madhi SA, Dangor Z, Heath PT, Schrag S, Izu A, Sobanjo-Ter Meulen A, et al.. Considerations for a phase-III trial to evaluate a group B Streptococcus polysaccharide-protein conjugate vaccine in pregnant women for the prevention of early- and late-onset invasive disease in young-infants. Vaccine 2013;31(suppl 4):D52–7.
© 2014 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
22. Jordan HT, Farley MM, Craig A, Mohle-Boetani J, Harrison LH, Petit S, et al.; Active Bacterial Core Surveillance (ABSs)/Emerging Infections Program Network, CDC. Revisiting the need for vaccine prevention of late-onset neonatal group B streptococcal disease: a multistate, population-based analysis. Pediatr Infect Dis J 2008;27:1057–64.