Group B Streptococcal Infection in Newborns

July 26, 2011

In the newborn period, group B streptococcus agalactiae infection is the leading cause of sepsis in the United States. The overall attack rate of early onset GBS infection (EOGBSI) per 1000 deliveries ranges from 0.72 to 5.5.


Incidence, Morbidity and Mortality

In the newborn period, group B streptococcus agalactiae infection is the leading cause of sepsis in the United States.[1] The overall attack rate of early onset GBS infection (EOGBSI) per 1000 deliveries ranges from 0.72 to 5.5.[3] In term infants, the attack rate ranges from 0.5 to 1.5 per 1000 deliveries.[1] In premature infants, the attack rate (in parenthesis) correlates inversely with the birth weight; the estimates are: <1000 g (7.6 [4] - 26.2 [5]), 1001-1500 g (1.7-18.8) , 1501-2000 g (6.7- 11.0), and 2001-2500 g (3.2-4.8).[4]

Early onset GBS infection (less than 7 days of age) and late onset GBS infection (7 days to three months) has estimated mortality rates of 15% and 10%, respectively.[1] Term infants have mortality rates of 2% to 8% and premature infants 25%-30%.[1] Each year, of 3.5 million babies born in the United States[5], approximately 12,000 infants develop morbidity or mortality from GBS infection, 1600 have neurologic sequelae following meningitis, and 1600 die.[1]

Epidemiology and Clinical Manifestations

When both lower vaginal and anorectal sites are sampled and inoculated into selective broth media (Todd-Hewitt broth containing 5% defibrinated sheep blood, 8 mcg/mL gentamicin and 15 mcg/mL nalidixic acid), virtually 100% of the GBS carriers are detected.[6] Culturing the cervix alone, or not using selective broth media may fail to identify as many as 50% of the GBS carriers.[6] Colonization can be chronic (40%), intermittent or transient.[7] Although overal colonization rates do not vary by trimester, only 4% to 7% of women who had negative vaginal and anorectal cultures late in second trimester will have positive GBS culture at delivery [7,8], and approximately 33% of the women who had positive GBS cultures during the second trimester (weeks 15 to 28) have negative cultures at delivery.[9]

Approximately 20% of pregnant women harbor group B streptococci in the lower genital tract.[10] Two studies involving 2040 and 327 pregnant women, employing combined vaginal and rectal cultures using selected broth media and performed in the third trimester, showed colonization rates of 16.3% and 21%, respectively.[11,12] Vertical transmission from GBS colonized mothers results in colonization of 40% to 73% of their newborns; but only 1% to 2% will develop invasive disease (EOGBSI).[13]

In pregnant women, GBS infection may cause urinary tract infection, sepsis, chorioamnionitis, postpartum endometritis, pelvic thrombophlebitis, and endocarditis . In newborns, the major manifestations of GBS infection are congenital pneumonia, sepsis and meningitis.

Risk Factors For GBS Infection

The American Academy of Pediatrics (AAP) [1] and American College of Obstetricians and Gynecologists (ACOG) [14] have identified the following intrapartum risk factors as indications for antibiotic prophylaxis to prevent EOGBSI:

  • Prematurity ( < 37 weeks gestation).
  • Premature rupture of membranes ( <37 weeks gestation).
  • Prolonged rupture of membranes ( >18 hours) at all gestational ages. When membranes have been ruptured for 12 hours, an assessment of the likely duration of labor should be made so that chemoprophylaxis can be initiated if rupture is considered likely to extend 18 hours.
  • Maternal fever ( >37.5 C or >99.5 F).
  • Multiple births.
  • Previous sibling with GBS infection.

In contrast to ACOG, the AAP considers previous delivery of a sibling with GBS infection an indication for intrapartum maternal chemoprophylaxis regardless of maternal GBS status.

Screening and Chemoprophylaxis Strategies

Currently, no universal maternal GBS screening is mandated. As many as seven percent of the GBS negative mothers cultured late in the second trimester (by combined vaginal and rectal cultures in selective broth media) will be positive at delivery.[8] Conversely, approximately 33% of the GBS positive mothers identified late in the second trimester, will be negative at delivery.[9] Thus, intrapartum identification of the carriers would be desirable. It would avoid exposure of women who are GBS positive and their infants to antibiotics if the mothers become negative at delivery. At least as important, intrapartum testing will identify GBS positives women who were negative earlier in pregnancy. Unfortunately, reliable rapid diagnostic tests, which would be ideally suited for this situation, are not available at the present time.

The American Academy of Pediatrics recommends screening of all pregnant women for GBS at 26-28 weeks gestation.[1] The GBS cultures should be obtained from two sites (lower vagina and anorectum) using separate swabs, and placed into a single tube of selective broth media. If carrier mother has symptoms of GBS infection, has GBS bacteriuria or requires cerclage during antepartum period, she should be treated. At the present time, the ACOG and AAP do not recommend chemoprophylaxis of carrier mothers in labor who have no risk factors. Chemoprophylaxis would place many mothers at risk of reaction to penicillin. Additionally, in many cases the carrier state will not be eradicated by penicillin, and approximately 33% of the carrier mothers will not be colonized at the time of delivery in absence of prophylaxis.

If the mother is a carrier, with one or more risk factors at the onset of labor or at the time of rupture of membranes, intrapartum intravenous ampicillin should be administered. Two grams are given initially followed by 1-2 g every 4-6 hours until delivery. Alternatively, penicillin G can be used, at a dose of 5 million Units every 6 hours. If there is a history of penicillin allergy, erythromycin 500 mg every 6 hours or clindamycin 900 mg every 8 hours should be given intravenously. Ideally, chemoprophylaxis should be administered at least 4 hours prior to delivery. This allows sufficient time for the antibiotics to achieve optimal concentrations in placental circulation and amniotic fluid.[15] If the mother has had premature labor, or has had premature rupture of membranes and her GBS status is unknown, the mother should be cultured by one of the rapid cultures if available, or by selective broth cultures. Those who are positive should be treated. Whether those for whom no results are available should be treated should be left to the discretion of the treating obstetrician. Mothers who develop chorioamnionitis should receive therapy rather than chemoprophylaxis. Therapy for chorioamnionitis consists of intravenous ampicillin 2 g every 4 hours and gentamicin 1.3-1.6 mg/kg every 8 hours, or an equivalent regimen. Therapy should be continued until symptoms and signs resolve and the patient is afebrile for at least 24 hours.

Maternal chemoprophylaxis decreases the incidence of EOGBSI when risk factors for GBS infection are present at delivery.[16,2] However, it is estimated that chemoprophylaxis fails to prevent up to 30% of EOGBSI and 10% of the fatalities.[16,17]

Management of Infants of GBS Carrier Mothers

Attempts to eradicate maternal vaginal and rectal colonization with antibiotics during the antepartum period have been unsuccessful.[18] Thus, intrapartum antibiotic therapy of GBS carriers mothers has been promoted to prevent GBS infection of the fetus and newborns. Prophylactic antibiotics given to newborns whose mothers are GBS carriers may be ineffective because some neonates which became infected in utero are already septic at birth.[19] Even if the fetus becomes infected in utero, she or he may have sterile blood cultures (BC) due to maternal chemoprophylaxis and transplacental transfer.

Antepartum chemoprophylaxis of GBS colonized mothers with risk factors for neonatal GBS infection at delivery has reduced the risk of EOGBSI.[16,2] No studies regarding management of infants born to mothers who received antepartum chemoprophylaxis have been published. Thus, the recommendations are empirical.[18]

In newborn infants, we identified the following risk factors for the purposes of this management guideline:

  • Prematurity with gestational age <34 weeks
  • Prolonged rupture of membranes ( >18 hrs)
  • Previous sibling with GBS infection
  • Multiple births if gestational age is <37 wks
  • Intrapartum maternal fever (38.0 C or 100.4 F)

The AAP includes multiple gestations in the list of risk factors without specifying the gestational age. However, we limited this risk factor to only infants of <37 wks gestation as we anticipate that full term infant products of multiple gestation are not at high risk for infection. Even though a lower intrapartum temperature (37.5 C) was defined as a risk factor by ACOG, we elected to chose a higher temperature to avoid unnecessary inclusion of many newborns in the high risk group, as we feel that it is very common to have an intrapartum temperature of 37.5 C in healthy women.

Infants who have one of the listed risk factors should have a complete blood cell count and differential (CBC) and blood culture (BC) drawn, a chest radiograph (CXR) if respiratory symptoms are present, and should receive intravenous ampicillin + gentamicin therapy (150 mg/kg/d ampicillin or 200,00 U/kg/d penicillin G divided TID, and 5 mg/kg/d gentamicin if <=7 days old or gentamicin 7.5 mg/kg/d if >7 days old divided BID).[20] Penicillin G or ampicillin alone may be given when GBS infection has been identified and the sensitivity of GBS organism has been determined.[21] Lumbar puncture (LP) is recommended in all infants who have neurological symptomps or signs of impaired cardiac output and/or infants who have positive blood cultures. When the CBC or CXR are abnormal, it is up to the discretion of the treating physician whether or not to perform an LP. If the LP is positive for GBS infection, a minimal course of 14 days intravenous antibiotic therapy is indicated [150-200 mg/kg/d ampicillin or 250,000-400,000 U/kg/d penicillin G divided TID (if <=7days old) and 300-400 mg/kg/d ampicillin or 400,000 U/kg/d penicillin G divided to four doses (if >7 days old).[20,21]

Asymptomatic newborns without any risk factors whose mothers received chemoprophylaxis do not require any further work up or therapy. At-risk infants whose mothers did not receive chemoprophylaxis, on the other hand, are more problematic since the rapid discharge policies now in effect may not allow sufficient time to observe the infant for development of symptoms. Therefore, in these infants we recommend obtaining a screening CBC. If the CBC is normal, no therapy is indicated. If mildly abnormal, the CBC may be repeated, and if the repeat CBC is normal or normalizing, no therapy is indicated. If the CBC is very abnormal, obtain a blood culture and give antibiotic therapy for at least two days, until the blood culture is known to be negative. If the repeat CBC is markedly abnormal, longer antibiotic therapy may be warranted. Obtaining a CBC in these asymptomatic infants is optional, if medical supervision can be maintained for >24 hrs.

If newborns are symptomatic, a CBC, blood culture, and CXR should be obtained. If the CBC or chest radiograph is abnormal, antibiotic therapy should be continued for five to ten days. If the blood culture is positive, it should be repeated, a lumbar puncture (LP) should be performed, and antibiotic therapy should be continued for ten days. If the LP is positive, antibiotic therapy should be continued for 14-21 days.

A flowchart for managing newborns born to GBS colonized mothers is presented along with this document. The recommendations are empirical, and provide guidelines that will include majority of the clinical situations. However, inevitably, there will be some cases in which further discretion of the treating physician will be needed. These guidelines are intended to guide medical practice and not to replace clinical judgment.

Clinical Signs of Bacterial Sepsis

Signs of bacterial sepsis are not specific. A study of clinical signs of sepsis of 455 newborn infants at four medical centers [22-25] showed the following prevalence:

  • Hyperthermia 55%
  • Jaundice 35%,
  • Tespiratory distress 33%
  • Anorexia 28%
  • Vomiting 25%
  • Cyanosis 24%
  • Apnea 22%
  • Abdominal distension 17%
  • Hypothermia 15%
  • Diarrhea 11%

In addition to these signs, the following clinical signs were observed in infants with neonatal sepsis: lethargy, poor feeding, persistent signs of impaired cardiac output, bloody stools and meconium stained amniotic fluid.[26] Of course, many of these signs can also be present in sick newborns who do not have bacterial infection.

Almost all infants with bacterial infection in the first few hours after birth will have one of the following symptoms: lethargy, apnea, seizures, hypothermia, hyperthermia, impaired cardiac output, respiratory distress, or poor feeding. Additional signs or symptoms such as abdominal distension, vomiting, bloody stool, non-physiologic jaundice and meconium stained amniotic fluid, in the abscence of any of the above signs, may not suggest infection. In these circumstances, discretion should be used by the treating physician regarding whether or not to pursue a sepsis workup.

White Blood Cell Count and Differential

Neutropenia, an increased ratio of immature forms to mature neutrophils (I:T ratio), or an increase in the absolute immature neutrophil count have a better predictive value than increased absolute neutrophils count (ANC) and absolute white blood cell count.

In term infants and premature infants >1500 grams, the following parameters for neutropenia, neutrophilia and absolute immature neutrophil count are considered abnormal (adopted from Manroe [27]):


 Diagnostic values (neutrophils/mm3) [27]Postnatal age (hours)
Immature neutrophils*>11000

* Mature neutrophils = polymorphonuclears (PMNs or "polys"); immature neutrophils = bands + metamyelocytes + myelocytes + promyelocytes.

In very-low-birth-weight infants (<1500 g), the following parameters of neutropenia and neutrophilia are considered abnormal (adopted from Mouzinho [27]):


 Diagnostic values (neutrophils/mm3) [28]Postnatal age (hours)

In term and premature infants the following parameters of I:T ratio are considered abnormal (adopted from Maroe [27]):


 Diagnostic values [27]Post-natal age (hours)
I:T ratio>0.1624

Increased ANC may be found in infants in the following conditions: Maternal fever (healthy newborn), oxytocin therapy (6 hours or more), stressful labor, asphyxia, meconium aspiration syndrome, pneumothorax, seizures, prolonged crying (4 minutes), hypoglycemia, hemolytic disease and surgery. Decreased ANC may be found infants whose mothers have hypertension, or who themselves have periventricular hemorrhage and hemolytic disease.[27,29] The total white cell count is the least helpful in the diagnosis of neonatal infection and was dismissed by some investigators as a diagnostic tool.[30]

The majority of the decisions regarding workup and management of neonatal GBS infection are made within the first 24 hours postnatally. In the first day of life, the following CBC values (reference values at 12 hours of age), are considered abnormal:

  • Neutropenia <7200 in term infants or preterm infants >1500 g, and <2200 in preterm infants <1500 g.
  • I:T ratio > 0.16 (for both term and preterm infants).
  • Immature neutrophils >1400 (for both term and preterm infants).

The sensitivity (probability that the test will be abnormal in a presence of the disease) for neutropenia is 87%, for increased absolute immature neutrophil count is 42%, and for an I:T ratio of >0.3 is 91%.[31] In premature infants less than 32 weeks gestation during the first five days of life, an I:T ratio of <0.2 is associated with no infection in 96%.[32] Neutrophilia of >14400 (term infants) and >12600 (preterm infants) and leukocytosis >31000 (term infants) and >21000 (preterm infants) are much less sensitive parameters than those mentioned above.

Counterimmune Electrophoresis (CIE)

The latex agglutination urine test for GBS antigen by CIE method is an unreliable screening test for GBS sepsis because of the high false positive rate. In most instances positive urine CIE for GBS is due to contamination of bagged specimens from perineal colonization[33], or due to absorption of GBS antigen from gastrointestinal tract[34]. As few as 100 group B streptococcal organisms contaminating a specimen may result in a false positive results.[35] Therefore, urine CIE testing is usually not recommended as a part of GBS infection work-up.



Copyright © 1998

Neonatology on the Web