Implementing new GBS guidelines requires coordinated care
- Copyright © 2003 by the American Academy of Pediatrics
Coordinating care by pediatricians, obstetric and nursing staff as well as laboratory services will simplify implementation of group B streptococcus (GBS) guidelines, further reducing GBS disease burden in neonates less than 7 days of age (early-onset) and pregnant women.
To solidify this approach, last August the U.S. Centers for Disease Control and Prevention (CDC) revised, and the Academy later endorsed, guidelines for Prevention of Perinatal Group B Streptococcal (GBS) Disease (MMWR. 2002;51(RR-11):1-22OpenUrlPubMed). On Nov. 29, the American College of Obstetricians and Gynecologists (ACOG) issued recommendations for universal culture screening of pregnant women for GBS as the single strategy for selecting women who should receive intrapartum antibiotic prophylaxis (IAP), supporting the CDC’s newly revised guidelines for the prevention of early-onset GBS disease.
It is anticipated that more GBS cases will be prevented once the new recommendations are implemented. Optimal infant management (e.g., no invasive testing, no empirical therapy, minimally prolonged hospital stay) depends on recommended IAP regimens for four or more hours before delivery. Chemoprophylaxis has been demonstrated to be a successful strategy to reduce GBS perinatal disease. This remains the interim strategy until prevention through a GBS vaccine becomes a reality.
Most women in whom GBS is colonized will be identified before the onset of labor or rupture of membranes, thereby allowing prompt initiation of IAP at hospital admission and greater likelihood that IAP would be administered four or more hours before delivery, a timing that allows for optimal efficacy (N Engl J Med. 2002;347:233-240OpenUrlCrossRefPubMedWeb of Science).
Women who present for delivery prior to screening or for whom GBS screening results are not available will receive IAP only if:
the gestation is less than 37 weeks,
the membranes have been ruptured for 18 or more hours or
fever is 100.4 F or higher during labor.
Women with one or more of these three “risk factors” associated with early-onset GBS disease in the infant will not be candidates for IAP if their screening cultures are negative. As in prior recommendations, women previously delivering an infant with GBS disease or with GBS bacteriuria during the current pregnancy always will receive IAP, obviating the need for culture screening.
Infant management algorithm
Based on new evidence, the new algorithm for the management of newborns born to women who have received IAP to prevent early-onset GBS disease (see figure) has been modified since publication of the 1997 AAP recommendations (Pediatrics. 1997;99:489-496OpenUrl).
Three points to emphasize are:
The new algorithm, like the prior one, is for the newborn infant whose GBS-positive mother received IAP (not the woman who failed to receive IAP). Management of the infant whose mother didn’t receive IAP requires clinical judgment, but physicians should recognize that the risk of GBS sepsis for these neonates, including all birth categories, is 1% to 2%. Judgments will range from observation for 48 hours without laboratory evaluation or empirical therapy (e.g., some healthy-appearing term infants) to evaluation and treatment of all infants, or at least most of those born before 37 weeks’ gestation.
The algorithm is evidence-based to the extent that data are available. It is not an exclusive approach to management; variation that incorporates individual circumstances or institutional preferences may be appropriate.
The algorithm provides guidelines only for those infants born to mothers who received penicillin, ampicillin or cefazolin at the recommended dosing regimens. Data to dictate the management of neonates born to women receiving a potentially less effective regimen (e.g., vancomycin) are not available.
Chorioamnionitis and its risks
Suspected chorioamnionitis is included on the new algorithm because the woman with GBS colonization and clinical chorioamnionitis as well as her infant are at increased risk for GBS bacteremia. (Clinical chorioamnionitis is defined as fever of 100.4 F for more than one hour and two or more clinical or laboratory findings suggesting intra-amniotic fluid infection.) In such a circumstance, prompt evaluation and initiation of empirical antimicrobial therapy for the infant is prudent irrespective of maternal antibiotic regimen or duration before delivery.
The inclusion of “suspected chorioamnionitis” in the algorithm was not intended to provide guidelines for infants born to GBS-negative women or for those born to women with fever likely to be caused by infections other than chorioamnionitis or by non-infectious etiologies.
Diagnosis requires communication
The clinical diagnosis of chorioamnionitis can be difficult, and good communication between the obstetrical and pediatric care providers will assist with appropriate infant management.
For example, antibiotics may be initiated during labor for a single temperature elevation accompanied by tachycardia and if by the time of delivery it is unclear whether chorioamnionitis has caused these signs. In this circumstance, healthy-appearing infants born at greater than or equal to 37 weeks’ gestation could receive routine care and close observation for a change in clinical status. Contrarily, a physician caring for a stable 32-week gestation neonate born to a woman with the same clinical scenario likely would decide to fully evaluate for sepsis and initiate empirical broad-spectrum therapy. The footnotes in the algorithm and the text of the CDC guidelines attempt to provide data to assist with these clinical decisions.
Full diagnostic evaluation and lumbar puncture
There are two clinical scenarios that may merit consideration of an examination of cerebrospinal fluid (CSF) in the “full diagnostic evaluation,” as shown in the figure.
The first is when an infant, despite maternal IAP, has signs of neonatal sepsis. The blood culture may be sterile in up to 15% of neonates with meningitis, and this frequency increases when some IAP has been administered. The clinical management of and prognosis for infants with meningitis differ from those with normal CSF. Thus, “if feasible,” a lumbar puncture (LP) is recommended. The physician must be prudent in the decision to defer the LP, but in those neonates who are critically ill or very immature, this course of action is reasonable.
The decision to continue antimicrobial therapy beyond 48 hours in infants with sterile blood cultures and no CSF examination should prompt examination of the CSF as would be indicated for the infant with bacteremia. Abnormal CSF findings suggesting meningitis would dictate use of an antimicrobial agent(s) that achieves bactericidal concentrations in the CSF, treatment for a minimum of 14 to 21 days and careful long-term assessment for neurologic sequelae.
The second clinical circumstance that would lead to a full diagnostic evaluation including examination of CSF is the infant born to a woman given antibiotics for suspected chorioamnionitis. This is intended for the infant with signs of sepsis and allows the physician discretion with respect to performance of the LP, as discussed previously by including the phrase “if feasible.”
For example, a healthy-appearing term infant might have determination of a complete blood cell count and differential, collection of a blood culture, and initiation of broad-spectrum antibiotics (e.g., ampicillin plus gentamicin); if the clinical course and laboratory evaluation are inconsistent with bacterial infection, therapy could be discontinued at 48 hours without examination of the CSF.
On the other hand, a neonate born at 32 weeks’ gestation without clear signs of sepsis may be evaluated more extensively including examination of CSF and managed more conservatively. While the algorithm attempts to provide guidance in this circumstance of enhanced risk for neonatal sepsis, individual circumstances will dictate the extent of the evaluation and the need to extend empirical antimicrobial therapy.
Ampicillin-resistant E. coli sepsis
The concern that maternal IAP to prevent early-onset GBS sepsis may lead to increases in gram-negative infection, especially ampicillin-resistant Escherichia coli, in neonates has been raised by several authors.
While ongoing surveillance should be encouraged, to date the only documented increase in ampicillin-resistant E. coli sepsis is in the small proportion of neonates who are very low birthweight (VLBW; less than 1,500 g) (N Engl J Med. 2002;347:240-247OpenUrlCrossRefPubMedWeb of Science; Pediatrics. 2002;110:690-695OpenUrl). Even among VLBW infants, there are no data to indicate that maternal IAP for GBS prevention causes the increased likelihood that an E. coli isolate from an infant with sepsis will be ampicillin-resistant.
Studies to date provide no information regarding the prevalence of ampicillin-resistant E. coli in the community where the mother lives or of antimicrobial use in women delivering VLBW infants prior to hospital admission for delivery.
While population-based surveillance to date is reassuring that non-GBS early-onset sepsis is not increasing in the era of IAP (Pediatrics. 2001;108:1094-1098OpenUrl; MMWR. 2002;51[RR-11]:1-22OpenUrlPubMed), obstetric care providers should heed the recommendation not to treat GBS colonized women before the intrapartum period and to use the narrow spectrum drug of choice for IAP, penicillin.
The CDC recommendations are available at www.cdc.gov/groupbstrep.
Dr. Baker is a member of the AAP Committee on Infectious Diseases. Dr. Kanto is a member of the AAP Committee on Fetus and Newborn.