Group B streptococcus is a commensal organism that is capable, under adverse conditions, of causing invasive infection. It is an important perinatal pathogen causing obstetric complications and neonatal infectious morbidity including early- and late-onset neonatal sepsis.1'5 Approximately 8000 to 12,000 cases of neonatal sepsis result from group B streptococcus annually in the United States. Over the past two decades, a number of regimens have been evaluated for the identification of carriers and prevention of neonatal disease. Although most obstetricians agree that there needs to be a concerted effort directed toward the prevention of neonatal group B streptococcal infection, the lack of an available program that is highly predictive and efficacious has led to considerable controversy. This article reviews the microbiology of group B streptococcus, the epidemiology and consequences of its carriage, optimal methods for identification, and the various regimens available for the prevention of maternal and neonatal sequelae.
What is group B streptococcus (Streptococcus ogolactiae)?
Streptococcus agalactiae is a member of the family Streptococcaceae, which was identified as a cause of puerperal sepsis in the 180Os. Using cell wall carbohydrate antigen structure, Rebecca Lancefield classified streptococci into different serologie groups, some of which have been identified as human pathogens (Table).6 Streptococci are gram-positive facultative anaerobes, ie, capable of growth in both aerobic and anaerobic conditions.
Streptococcus agalactias are spheres or ovoids that grow in pairs or short chains. When grown on blood agar, S agalactiae produce large (>2 mm) colonies with a surrounding zone of complete (beta) hemolysis in 99% of cases. The colonies are yellow to red in color unless grown in glucose-rich media in which carotinoid production is suppressed. Its identity can be confirmed by the ability to colonize hippurate or through the use of antisera against specific group B streptococcal antigens.
Classification of Streptococcaccac Involved In Human Diseases
A number of surface carbohydrate and protein antigens are specific to S agalactiae. Work by researchers7'10 has led to the definition of a number of subtypes (Ia, Ib II, III, IV, and V). A small number of strains are nontypable based on the presence or absence of characteristic carbohydrate surface antigens. R, C, and X surface protein antigens have also been identified and may be present on the bacterial surface in various combinations. These protein antigens are seen with varying frequency depending on the carbohydrate serotype evaluated.11'13 A number of additional serotypes recently have been studied, including serotype VI, strain 7271 (type VII), and JM9 (type VIII).13'14
The term "group B streptococcus" is synonymous with and frequently used in place of "Streptococcus agalactiae" in the obstetric and pediatrie literature. "Group B streptococcus" will be used in this article to avoid confusion.
Where is group B streptococcus found?
In asymptomatic humans, group B streptococcus can be isolated from a number of sites. Group B streptococcus grows more efficiently on mucous membranes rather than dry skin, and is found primarily in the oropharynx, upper respiratory tract, and the gastrointestinal tract. The female genital tract is colonized secondarily from the rectum. In 1981, Islam found similar incidences of positive cultures obtained from the perianal region, perineum, and low vagina (18% to 21%). Recovery of group B streptococcus from the cervix was significantly lower than the vagina (9.1% versus 15,6%) in the Vaginal Infections and Prematurity study group (VIP study) of 7742 women.16 Boyer et al ' found a 22.8% carriage of group B streptococcus in 5586 women. Vaginal and rectal cultures were positive in 14.7% and 18.7% of grávidas, respectively. Thus, cultures are less likely to be positive, as sampling is performed further from the rectum (rectal>distal vagina>proximal vagina >cervical).
What is the best place to culture for group B streptococcus!
Because group B streptococcus is carried primarily in the rectum and the female genital tract is secondarily colonized, combined sampling should be performed from both the distal third of the vaginal mucosa (just inside the hymen) and the anorectal canal. Isolated cervical cultures are not adequately predictive.17
How common is group B streptococcal carriage in pregnancy?
The incidence of maternal colonization is difficult to define because of the wide variation in culture site(s) and microbiologie techniques used by various investigators. Colonization has been identified in 2.8% to 31% of pregnant women but is generally identified in 15% to 28% of gravidas.18
Are there different clinical and demographic factors that alter carriage?
In 1978, Anthony et al19 demonstrated group B streptococcal carriage to be less common among Mexican Americans, women 5*20 years old, and multigravidas (5*4). Similarly, in 1980, Yow et al20 found multiparity to be protective against carriage (13.8% versus 31.4%). Blacks have a higher camer rate than either whites or non-New York Hispanics.16 Additionally, this large multicenter trial revealed a wide variance in carriage based on geographic location, a protective effect of smoking and schooling >12 years, and an increased risk with advancing age and low parity. Group B streptococcal carriage was more common among women having frequent intercourse with multiple partners, suggesting the potential for horizontal transmission. An early study suggested maternal blood type-B to be a risk factor for group B streptococcal carriage.21 However, this was not supported by a subsequent study of 1213 women that unexpectedly found Rhnegative status to be a risk factor for carriage (15.7% versus 9.3%; P<.01).22
Is group B streptococcal carriage constant?
In 1975, Baker et al23 published results of a trial of 183 second-trimester patients and 205 women admitted in labor. Their results suggested vaginal carriage to be present more often at delivery when compared with earlier sampling (22.4% versus 13.1%). Subsequent studies have yielded conflicting results. In 1978, Anthony et al17 obtained group B streptococcal cultures from the posterior pharynx, urethral orifice, and endocervical canal. Carriage was not consistently identified among the study patients. Thirty-six percent of women were classified as chronic carriers while 20% were transient and 15% were intermittent carriers. A similar pattern was identified in 1980 by Yow et al17 and a large study of 5568 grávidas from whom vaginal and rectal group B streptococcus samples were obtained prenatally and at delivery. Antepartum carriers were colonized at a rate 67-2% of the rate at time of delivery. A progressive decline in predictive value was seen with cultures obtained more than 6 weeks prior to delivery. These studies are the basis for our belief that group B streptococcal cultures remote from delivery are inadequately predictive of maternal colonization at delivery.
Will serially repeated cultures improve sensitivity?
Aber et al24 demonstrated repeat vaginal cultures over a 6- to 8-week period to increase the identification of carriers from 17% to 34%. While this finding is encouraging, it is possible that initial cultures were less than optimal due to culturing of just one site and the use of nonselective culture techniques. Regardless of culture technique, there may be a background incidence of false-negative testing that could be reduced by serial testing. However, this practice is not currently advocated.
What techniques are available for the detection of group B streptococcus ?
A number of techniques for the identification of group B stteptococcal carriage have been studied. Culture provides the most reliable results but is time consuming. Recent study has focused on a more rapid diagnosis through the evaluation of Gram's stain, rapid antigen tests, and rapid culture.
Our microbiology laboratory currently cultures group B streptococcus on blood agar plates (5% sheep blood agar in a trypticase soy agar base) with 8 Mg/mL gentamicin and 15 pg/mL nalidixic acid to "select" out other bacteria. Cultures are incubated in 5% carbon dioxide at 37° for 24 to 48 hours. The use of such "selective media" improves identification by 30% to 40%. Detection also can be enhanced by preincubating the specimen in Todd Hewitt broth (with 15 Mg/mL nalidixic acid and 8 pg/mL gentamicin) for 18 to 24 hours at 350C or Lim broth (Todd Hewitt broth containing 1% yeast extract, 10 \igjmL colistin and 15 pg/mL nalidixic acid) prior to subculturing onto blood agar plates.25,26
Gram's stain is inconsistent in the prediction of subsequent culture because of the wide variety of gram-positive cocci found within the oropharynx and vagina (Table). In a study by Feld et al, a positive Gram's stain was identified in 42% of vaginal specimens but just 30% of these had a subsequently positive group B streptococcal culture. However, sensitivity was 100%, and the absence of gram-positive cocci was uniformly associated with a negative culture. Subsequent studies of 98 and 115 women revealed sensitivity of 93% and 38% and specificity of 61% and 69%, respectively.28'29 Most recently, a large multicenter trial evaluated women at 23to 26 weeks gestation, and again in labor. The sensitivity, specificity, positive and negative predictive values of Gram's stain in the prediction of vaginal culture results were 28.2%, 69.4%, 17.2%, and 81.1%, respectively, with similar findings in laboring women (34%, 72%, 18%, and 86%, respectively).30
A number of antigen tests have been developed for the identification of group B streptococcus based on the identification of specific surface antigens. Antigen detection through coagglutination, latex particle agglutination, enzyme immunoassay, and immunofluorescent antibody testing have proven insensitive or impractical in the identification of group B streptococcus (4% to 88%).31,32
What are the maternal sequelae of group B streptococcal carriage?
The annual incidence of invasive group B streptococcal disease is approximately 4 to 6.2/100,000 adults, with blacks carrying a greater risk than whites.33'34 Complications specific to pregnancy, including septic abortion, urinary tract infections, premature rupture of the membranes, preteriti birth, chorioamnionitis, endometritis, and wound infections, have been attributed to invasive group B streptococcal disease. The resulting bacteremia offers the opportunity for sequelae including endocarditis, generalized sepsis, and meningitis.
A number of trials have suggested group B streptococcal bacteriuria to be associated with adverse pregnancy outcome. Wood and Dillon identified group B streptococcal bacteriuria in 14 women (2.5%) in an evaluation of 569 gravidas. Two of 14 group B streptococcal carriers suffered unanticipated fetal demise, while one of the remaining 12 newboms suffered neonatal sepsis and one had a group B streptococcal scalp abscess. Moller et al36 found a similar increase in the incidence of premature rupture of the membranes (35% versus 15%) and preteriti labor (20% versus 8.5%) with group B streptococcal carriage. As was seen by Wood et al, 5 of 68 bacteriuric patients (7.4%) had infants who suffered group B streptococcal sepsis while no group B streptococcal sepsis occurred in the nonbacteriuric group. In a subsequent prospective randomized placebo-controlled study of group B streptococcus bacteriuric grávidas, Thomsen et al37 demonstrated penicillin therapy to reduce the incidences of premature rupture of the membranes (11% versus 53%; P<.001 ) and preteriti labor (5.4% versus 38%; P<.002). In 1988, Daugaard et al38found cervical colonization and group B streptococcal bacteriuria to be more common in women suffering spontaneous miscarriages when compared with uncomplicated gestations. A recent study failed to confirm the association between bacteriuria and adverse pregnancy outcomes.39
The association between prematurity and genital tract group B streptococcal carnage in the absence of bacteriuria is less clear. In a review of seven case-control studies, Romero et al40 found no consistent correlation with just one study demonstrating significant risk. Most recently, in 1994, Yancey et al41 evaluated 823 women admitted for preteriti labor or premature rupture of the membranes within 2 weeks of delivery. Women underwent vaginal sampling and cultures were performed using selective broth media. Multivariate analysis revealed chorioamnionitis to be linked with group B streptococcal carriage. The odds of chorioamnionitis increased with light (odds ratio [ORl =1.9), moderate (OR= 2.6 ) , and heavy (OR=3.2) colonization.
Postpartum endometritis due to group B streptococcus complicates approximately 1/1000 deliveries and frequently will have a striking clinical course. The vast majority of cases appear to occur postoperatively. Patients present with a high-spiking, early-onset postoperative fever (mean of 12 hours), chills, and maternal tachycardia. Up to one third of women are bacteremic (versus 9% of non-group B streptococcal maternal infections). Importantly, the infants of these mothers have a significant risk of sepsis and subsequent mortality. This highlights the importance of early communication between obstetric and neonatal caregivers should maternal infectious complications occur.
What are the neonatal consequences of maternal group B streptococcal carriage?
Neonatal sepsis is the most dramatic and devastating complication of neonatal colonization with group B streptococcus and complicates approximately 2/1000 deliveries. Disease occurring within the first 7 days of delivery is classified as "early-onset" sepsis and accounts for 60% to 80% of neonatal group B streptococcal sepsis. Early-onset neonatal sepsis presents frequently as sepsis, pneumonia, and meningitis. Lateonset neonatal disease, occurring after 7 days of life, is less common with two thirds of infants presenting with meningitis and the remainder suffering generalized infection. Omphalitis, septic arthritis, and osteomyelitis are less common sequelae.
The majority of infants with early-onset disease will exhibit the classic risk factors for group B streptococcal sepsis and become symptomatic within 24 hours of delivery. One third will demonstrate symptoms of sepsis including tachypnea, grunting apnea, cyanosis, and hypotension. Group B streptococcal pneumonia may present a classic picture of lobar pneumonia. Unfortunately, about half of the cases have a radiographie picture similar to respiratory distress syndrome. Neonates with meningitis frequently have generalized symptoms early in the course, although up to 50% will eventually have seizures.
Infants suffering late-onset neonatal sepsis frequently have no specific risk factors, returning to the hospital with lethargy, irritability, and poor feeding. Up to one third will have antecedent respiratory symptoms. Survival after neonatal group B streptococcal sepsis appears to be improving, with recent rates of up to 80% to 90%.42
Has group B streptococcus always been an important pathogen?
The relative frequency of group B streptococcus as a cause of neonatal infection has varied over the past 65 years. Sepsis due to group B streptococcus was uncommon before the 1960s (1.4%-6.5%).4'43'45 Group A streptococcus was the predominant pathogen in the 1930s and 1940s, and subsequently was replaced by Escherichia coU until the 1960s in one tertiary care center.46 Subsequently, the impact of group B streptococcus as a perinatal pathogen increased substantially in the 1960s and 1970s. Group B streptococcus is now responsible for up to one third of neonatal sepsis and is the most common pathogen identified in bacterial cultures of septic infonts.3'44,45'47-49
How is group B streptococcus transmitted to neonates?
Neonatal colonization with group B streptococcus results primarily from vertical transmission during the birth process. The incidence of neonatal colonization of infants born of noncolonized mothers is low, generally between 1% and 3%,z'17'50'52 suggesting either horizontal transmission from care givers and family members, or that these mothers were actually colonized. Horizontal transmission is further supported by the presence of infants who are delivered of non-carriers and who have negative cultures at delivery (2.2%).50
Are all neonates of carrier mothers colonized?
As is the case in adults, neonatal colonization by group B streptococcus is primarily on moist mucosal surfaces including the ear, umbilical cord, pharynx, and rectum. Skin cultures at the time of delivery are likely to identify contamination from the mother rather than true colonization. Mucosal cultures obtained 24 to 48 hours postdelivery will identify those infants truly colonized by viable group B streptococci. Early neonatal rectal cultures are less likely to be positive even in the presence of neonatal group B streptococcal sepsis, suggesting a delay in colonization of this site. In a large study of 23 1 7 neonates, Pass et al2 reviewed cultures of external ear, umbilicus, throat, and anus prior to cleansing of the infant, as well as repeat cultures of the throat and pharynx prior to discharge. Half of the infants had positive cultures from just one of the sites, highlighting the importance of cultuung multiple sites if knowledge of neonatal colonization is desired. Invasive disease in symptomatic neonates can be identified through blood, urine, and cerebrospinal fluid cultures.
Despite extensive contact between the fetus and the birth canal, neonatal colonization occurs in just 43% to 63% of untreated mothers who were identified as carriers in labor.2,17'49,50'52 Neonatal colonization is more frequently seen in the face of heavy maternal colonization (64.5% versus 20%). 17 In an excellent review, Baker and Edwards18 summarized over 20 studies of matemal and neonatal group B streptococcal colonization and disease. Vertical transmission occurred in 29% to 72% of cases although cultures were obtained over a wide range of times subsequent to delivery. The median incidence of vertical transmission was 50%.
What are the chances of neonatal infection given a colonized mother?
Neonatal group B streptococcal sepsis complicates approximately 1.8/1000 deliveries. Some variation is seen in the incidence in different geographic locations and over time (range 0.3 to 3.7/1000).4 In a 6year prospective evaluation of mothers and newborns in Birmingham, Alabama, maternal and neonatal colonization were relatively constant at approximately 20% and 12%, respectively. Early-onset neonatal sepsis occurred at a rate of 1.8/1000, suggesting an attack rate of approximately 1/67 colonized infants (1.5%) and 1/110 colonized mothers (U.9%).1
Are there specific risk factors for earlvOnset neonatal sepsis?
A number of factors have been identified as important associations with early-onset neonatal sepsis. These include low birthweight, prematurity, maternal fever in labor, and prolonged duration of membrane rupture.42'53,54 Twin pregnancy has been suggested as a risk factor; however, this finding has not been supported in recent reports.55
The volume of bacteria colonizing the neonate also appears to be important in determining the risk of neonatal sepsis after colonization. Gerards et al53 demonstrated a relatively high incidence of neonatal sepsis in 145 colonized infants (14.5%). In that study, infants with moderate to heavy colonization had a 44% risk of early-onset sepsis (19/43) while those with light colonization had a 2% risk (2/102).
How can maternal group B streptococcal disease be prevented?
Reduction of maternal risk can be achieved through treatment of women with group B streptococcal bacteriuria. Currently, there is no evidence that routine antepartum treatment of asymptomatic vaginal or rectal carriers will alter pregnancy outcome.
What can be done to prevent neonatal disease?
Neonatal Prophylaxis. In an early observational study in which neonates received routine intramuscular penicillin for gonococcal prophylaxis, Steigman et al5 ' identified no cases of neonatal sepsis due to group B streptococcus over a 25-year period. Similarly, two retrospective studies using nonconcurrent controls showed a reduction of neonatal group B streptococcal sepsis in the penicillin-treated group of infants.58,59 However a large prospective randomized trial of 32,058 newborns by Seigel et al60 identified a reduction in the incidence of neonata! sepsis due to group B streptococcus but no reduction in the overall incidence of sepsis. The likely reason for failure of this treatment is the fact that many infants developing early-onset disease will already be bacteremic at the time of delivery.
Antepartum Prophylaxis. The first publication evaluating antepartum treatment of group B streptococcal carriers at term suggested treatment to adequately reduce matemal colonization in labor.61 However, subsequent studies demonstrated an inconsistent effect of early treatment of group B streptococcal carriers. In a study of maternal antimicrobial therapy remote from term, Gardner et al62 demonstrated persistence of carriage in 73% and colonization of the vagina in 70% of women intrapartum. In a study of neonatal therapy, Paredes et al65 similarly demonstrated persistent carriage in 8/12 infants after complete therapy. More recently, Lewin and Amstey64 randomized third-trimester group B streptococcal carriers to either intramuscular penicillin (mother and father) or expectant management and found a significant reduction in carriage at delivery (18.2% versus 75%). The difficulty in obtaining satisfactory eradication of group B streptococcus may occur due to reinfection from a colonized partner or by self-inocculation from the rectal reservoir. Penicillin therapy is frequently unsuccessful because of the large number of penicillinaseproducing Enterobacteriaceae within the rectum.
Intrapartum Prophylaxis. Intrapartum prophylaxis appears to be the short-term answer to prevention of early-onset group B streptococcal sepsis. A number of retrospective51,65 and prospective51,54,66-68 studies have demonstrated a reduction in vertical transmission of group B streptococcus with intrapartum prophylaxis. While several authors claim a reduction in early-onset neonatal sepsis with such therapy,54 others have failed to demonstrate this outcome.66'68,69 Although promising, these results are not quite as clear-cut as they appear on the surface. First, some studies claiming benefits of intrapartum prophylaxis actually included or allowed postnatal therapy within the study design. Additionally, antimicrobial therapy will reduce the incidence of positive bacterial blood cultures. Thus, intrapartum prophylaxis might significantly decrease the frequency of the diagnosis of neonatal group B streptococcal sepsis without actually reducing the number of sick infants. Although intrapartum prophylaxis seems encouraging, one reviewer questions whether intrapartum prophylaxis is effective in preventing neonatal disease.70
What is the best antibiotic regimen for prophylaxis or treatment?
The vast majority of group B streptococci are susceptible to penicillin. While many physicians currently give ampicillin for antepartum therapy of bacteriuria and intrapartum prophylaxis, this agent has a wider spectrum of antimicrobial coverage including gram- positive, gram-negative, and some anaerobic species. Because of the potential for selecting resistance to ampicillin, it would be more prudent to use penicillin. Adequate antepartum therapy can be achieved with penicillin V 250 to 500 mg orally every 6 hours for 7 days. Because of the potential for recolonization and failed treatment, repeat cultures should be performed. For intrapartum prophylaxis, penicillin should be given as an intravenous dose with a 5 million unit initial bolus followed by 2.5 million units every 4 hours until delivery. Penicillin-allergic women can be treated with either clindamycin (900 mg intravenously every 8 hours) or erythromycin (500 mg intravenously every 6 hours) intrapartum.
Chlorhexidine Irrigation. Intrapartum irrigation with intravaginal chlorhexidine has been suggested as an alternative treatment. Such therapy might reduce the excess neonatal therapy used because of concern that blood cultures may be falsely negative if the neonate has antibiotics on board. Several early studies were promising.72 However, we were unable to demonstrate a high success rate with such therapy. Vaginal irrigation with chlorhexidine did not significantly reduce intrapartum or postpartum febrile morbidity or overall neonatal colonization with group B streptococcus despite a trend toward reducing maternal group B streptococcal carriage.73
What protocols of intrapartum prophylaxis are available?
Recently, a number of strategies for the reduction of neonatal group B streptococcal disease have been proposed. Authors have suggested empirical treatment of high-risk patients in the absence of group B streptococcus cultures or treatment of high-risk patients if culture status is known.74 The American Academy of Pediatrics has suggested routine universal screening for group B streptococcus at 26 to 28 weeks' gestation with intrapartum treatment of high-risk carriers.75 Minkoff and Mead76 suggested performance of group B streptococcus cultures on patients in whom preterm birth is anticipated and subsequent treatment of carriers who deliver preterm. Most recently, the Centers for Disease Control and Prevention published a detailed policy suggesting intrapartum prophylaxis of all patients delivering prior to 37 weeks, all women with group B streptococcal bactertuna, and those with a previously affected infant.77 Women reaching 36 to 37 weeks' gestation would undergo culture. Group B streptococcus carriers would be counseled as to the risks and benefits of expectant management or intrapartum therapy. Intrapartum treatment would be given based on informed consent, regardless of risk status. This approach has the virtue of providing prophylaxis to all patients at risk who present for prenatal care but also will lead to intrapartum antimicrobial therapy of a significant proportion of the entire obstetric population, including those at intermediate risk (term carriers with no additional risk factors for neonatal disease).
Other suggested protocols have included universal intrapartum prophylaxis of all gravidas, and screening and treatment of high-risk groups. Unfortunately, a recent analysis by the Maternal-Fetal Medicine Research Unit of the National Institute of Child Health and Human Development (NICHD) demonstrated the impracticality of conducting a clinical trial to compare the efficacy of the screening and prophylaxis regimens suggested by the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists.78
Several authors have attempted to determine the optimal protocol of intrapartum prophylaxis through cost analysis.79'80 Unfortunately, they were unable to quantitate all costs associated with the various regimens, including the incidence and costs of non-group B streptococcal sepsis, drug-resistant infection, and excess maternal and neonatal treatment and hospitalization resulting from prophylactic therapy. In a survey of current practice among 925 maternal-fetal medicine subspecialists, 60% of physicians would give intrapartum prophylaxis based on maternal group B streptococcal carriage in the absence of other risk factors.81 A companion survey of 982 pediatricians demonstrated the potential for excess treatment with broad-spectrum antibiotic regimens and extended hospitalizaron of newborns delivered to mothers who were group B streptococcal carriers or who received intrapartum antimicrobial therapy.82
Regardless of the impact of intrapartum therapy against early-onset neonatal disease, such therapy will not eradicate late-onset neonatal sepsis due to the potential for postnatal colonization and will not prevent complications of carriage in the nongravid adult. Thus, antimicrobial prophylaxis cannot offer a long term solution to group B streptococcal sepsis. Active immunization offers hope for disease prevention in the general population. Passive immunization may be of benefit in those situations in which antenatal immunity has not been established.
Active Immunization. Active immunity carries the greatest promise for the prevention of group B streptococcal infection in adults and newboms. Unfortunately, no single antigen has been identified that is specific to all serotypes of group B streptococcus. Additionally, the prevalence of antibodies to the various serotypes is variable and infrequent. In 1980, Vogel et al83 reported on the incidence of maternal antibodies to various group B streptococcal serotypes. Antibody was detected against serotypes Ia, Ib, II, and III in 26%, 52%, 82%, and 42% of grávidas with just 5% of women having antibodies to all four serotypes. Thus, to attain immunity, a polyvalent vaccine is necessary.
Vaccine development is confounded by a number of problems. First, there are a number of serotypes against which a vaccine must be developed. Early study showed serotypes I, II, and HI to be equally common among asymptomatic group B streptococcal carriers. Recently, serotype V has been identified in a significant number of infant isolates in Maryland (9%).84 Second, polysaccharide antigens lead to a low incidence of antibody response.8 Third, specific group B streptococcal protein antigens are not consistently present in the various polysaccharide serotypes. Fourth, transplacental passage of antibodies is necessary to generate passive immunity. Although such transfer of antibodies has been demonstrated at term,86 it is inconsistent early in pregnancy when the risk of neonatal infection is greatest.
The immune response to polysaccharide antigens can be enhanced by linking the polysaccharide or a subcomponent oligosaccharide to a carrier protein such as tetanus toxoid.86,87 A polyvalent vaccine using polysaccharide-tetanus toxoid conjugates for several serotypes of group B streptococcus has been demonstrated to confer immunocompetence against the target serotypes in gravid mice and to improve survival of group B streptococcus-exposed newborns.88 More recently, conjugation of type III polysaccharide with the C-protein surface antigen has been demonstrated to generate an immune response against the C-protein as well as enhance the response to the carbohydrate antigen.89
Passive Immunization. An antenatal or postnatal program of passive immunization might offer protection or palliation in these circumstances. In-vitro studies regarding passive immunization with hyperimmune or monoclonal antibodies have been promising in vitro but await adequate in vivo trials on high-risk neonates.90-91
Group B streptococcus is a significant pathogen for both mother and child. Routine urine culture in pregnancy will identify and allow treatment of women with asymptomatic bacteriuria. An optimal protocol for the prevention of neonatal sepsis has not yet been developed. While intrapartum antimicrobial prophylaxis appears to provide the best potential, each of the currently suggested protocols has significant drawbacks. Drawbacks include the potential for missing high-risk carriers, failure to treat a significant proportion of those destined to deliver an affected infant because no risk factors are present, and empirical treatment of a large proportion of the population in order to present significant disease in a few. Until an effective program of immunization becomes available, intrapartum prophylaxis of group B streptococcal carriers appears to offer the best hope of reducing the incidence of neonatal disease.
Caregivers should adopt a uniform practice with regard to screening and prophylaxis. It is essential that any broad-based screening program include an evaluation of efficacy as well as complications including the development of new etiologic agents as causes of neonatal sepsis and the emergence of resistant bacteria. Further, mothers and newboms should be evaluated for drug adverse reactions and the impact of intrapartum prophylaxis on the use of prolonged empirical broad-spectrum antimicrobial therapy on the asymptomatic infant. Additional research is necessary regarding the required duration of therapy for optimal effect of intrapartum prophylaxis, the need for postnatal prophylaxis of the asymptomatic neonate, and the optimal agent for neonatal prophylaxis (penicillin versus broad-spectrum agents) if neonatal therapy is necessary after intrapartum prophylaxis.
1. DlllonHC. Khare S, Gray BM. Group B gtreptococcal carriage and disease: a 6-year prospective study. ) Pedlatr. 1967:110:31-36.
2. Paw MA, Gray BM, Khare S. Dillon HC )r. Prospective studies of group B strepiococcal Infections in infants. J Pediatr. 1979;95:437-443.
3. Freedmam RM, Ingram DL, Grow I. Ehrenkranz RA, Warshaw JB, Baltimore RS. A half century of neonaral sepsis at Yale. 1928 to 1978. Am J Dis Child 1981;35:140-144.
4. Nyhan WL, Fousek MD. Septicemia in the newborn. Pefionici. 1958; 12: 268-2 78.
5. Boyer KM. Maternal screening in prevention of neonatal infections: current statuì and rationale for group 8 «repiocnccal screening. J HOJ(I fn/ect. 1988:11(suppl A):328-3S3.
6. Lancefield RC. A serologieal differentiation of human and other groups of hemolytic streptococci. J Exp Mid. 1933 ;57: 571-595.
7. Lancefield RC, Hare R. The ierological differentiation of pathogenic and nonpathogenic strains of hemolytic streptococci from parturient women. J Exp Med. 1935:61:335-349.
8. Lancefield RC. Two serological types of group B hemolytic streptococci with related, but not identical, type-specific substances. J Exp Med 1938:67:25-40.
9. Perch B, Kjemg E, HenhcKsen J, New serotypes of group B itreprococcl isolated from human sources. J Cfcn Microfftol. 1979; 10. 109-110.
10. Jelinkova J1 Motlova J. Worldwide distribution of two new serotypei of group B streptococci: type IV and provisional type V. J Clin MierohwL 1985:21:361-362.
11. Flores AK, Ferrleri P. Molecular aspects of R-protein antigens produced by clinical isolates of group B streptococcus, J CIm MtcrafcioJ. 1989:27:1050-1054
12. Kvam Al, Efstratiou A, Bevenger L, et al. Distribution of serovariants of group B streptococci in isolates from Norway and England. J Med Mierobioi. 1995;42:246-250.
13. Wagner M, Murai T, Wagner B, Gumher E, Jelinkova J. JM9 strain!, anew type of group B streptococci from Japan. Im J Mea Murobioí Viral Pariatal Infect Dis. 1994:280:480-496.
14. Lammler CH, Schwarz S, Wlhawan IWT, Oil E, Bopp V. Martinez-Tagle A. Comparison of streptococci of serological group B Isolated from healthy carriers and active disease in Chile J Med MicroWoi. 1995:42:161-164.
15. Islam A. Primary carrier sites oí group B streptococci In pregnant women correlai - ed with serotype dlstrlbutioni and maternal pariry. J Clin Pathol. 198 1134: 78-81.
16. Began JA. Klcbanoff MA, Nugcnt RP. The epidemiology of group B streptococcal colonization in pregnancy. Vaginal infect kins and prematurity study group. Obtut Gynecai. 1991 ;77:604-610.
17. Boyer KM. Gadiala CA, Kelly PD, BunJ LI, Cotoff SP. Selective lntrapartum chemoprophylaxis of neonatal group B streptococcal early-onset disease. 11 predktlve value of prenatal culture». J tn/ect Du. 1983:148:802-809.
18. Baker Cj, Edwards MS. Croup B streptococcal Infections. In: Remington JS, Klein JO, eds. Injections Dneoitj of the Fetta ani ???t?tp Infant. Philadelphia. Pa: WB Saundcrs Co; 1990:742-811.
19. Anthony BF, Okada DM, Hobel Cj. Epidemiology of group B streptococcus: longitudinal observations during pregnancy. J Infect Du. 1 978; 137: 524- 5 30.
20. Yow MD. Leeds W, Thompson PK, Mason EO Jr. Clerk DJ, Bencher CW. The natural history of group B streptocoecal coloniiaiion In the pregnant woman and her offspring, I: colonliatkm studies. AmJ Ohswt Gynecai. 1980:1 57:14- 38.
21. Regan JA, Chao S, James LS. Maternal ABO blood group type B: a ridi factor In the development of neonatal group B streptococcal disease. Pediatrici 1978:62:504-509.
22. PSMI Ick M. Mead PB1 Philip AOS. Selective maternal culturlng to identity group B streptococcal infection. AmJ Ubati GjiwooL 1980:138:480-484.
23. Baker C], Barrett FF, Yow MD. The influence of advancing gestation on group B streptococcal colonization in pregnant women. AmJ Obstet QynecoL 1975;122:820-823.
24. Aber RC, Allen N, Howell JT, Wilkenson HW. Fact lam RR. Nosocomial trammission of group B streptococci. Pediatrici. 1976; 58: 346; 353.
25. Phitipson EH, Palermino DA, Robinson A. Enhanced antenatal detection of group B streptococcus colonitatkm. Ornen Gynecai. 1995:85:437-439.
26. Altaic SS, Dryja D. Detection of group B streptococcus. Comparison of H)IId and liquid culture media with and without selective antibiotics. Diagn Microbiol Inject Dis. 1994:18:141-144.
27. FeIdSM, Kanlgan JT. Vaginal gram stain asan immediate detector of group B streptococci in selected obstetric patients. Am } Obstet Gynecai. 1987:1 56:446-448.
28. Holls WM, Thomas J, Troyet V. Cervical Gram stain for rapid detection of colonization with ß-itrcptococcus. Orst« Gynecai 1987:69:354-7.
29. Sandy EA II, Blumenfel ML, lams JD. Gram Main in the rapid determination of maternal colonization with group B beta -streptococcus, Obtui Gvntcof. 1988:71:796-798.
30. Carey JC, Klebanoff MA. Regan JA. for the Vaginal Infections in Prematurity Study Group. Evaluation of the Gram stain at a screening tool for matemal carriage of group B Beta-hemolytic Hrepiococci. Obsta Gynecol. 1990:76:693-697.
31. Yancey MK, Armer T. Clark P, Duff P. Assessment of rapid ideniflcation tests for genital carriage of group B streptococcus. Objtn Gynecai. 1992:80: 1038-1047.
32. Skoll MA. Mercer BM. Batelski V, Gray JP, Ryan G, Sitai BM. Evaluation nf two rapid group B streptococcal antigen tests in labor and delivery pallenti. Ofettfi Gynecol. 1991;77:122-326.
33. Zangwlll KM, Sc hucha t A, Wagner JD. Group B streptococcal disease in the United States, 1990: report from a mulllstatc active surveillance system. MMWR CDC SurvetU Summ. 1992:41:25-32.
34. Parley MM, Harvey RC, Stull T. et al. A population-based assessment of invasive disease due to group B streptococcui in nonpregnant adults. N Engt J Med. 1993:328:1807-1811.
35. Wood EG. Ditlon HCJr. A prospective study of group B streptococcal bacierluria In pregnancy. AmJQbuaGynetoL 1981:140:515.520.
36. Moller M, Thomsen AC, Horch K, Dinesen K, Zdravkovlc M. Rupture of fetal membranes and premature delivery associated with group B streptococci in the urine of pregnant women. Lone«. 1984;2:fj9-70.
37. Thomsen AC, Morup L1 Hansen KB. Antibiotic elimination of group B streptococci in urine prevention of preterm labour. Lancet. 1987:591-593.
38. Daugaard HO, Thomsen AC, Henriques U. Ostergaard A. Group B streptococci in the lower urogenital tract and late abortions. Am } Obuci Gynecot. 19T8; 158:28-31.
39. McKeruie H, Donnei ML, Howie PW, Palei NB, Benvie DT. Risk of preterm delivery in pregnant women with group B streptococcal urinary Infections or urinary antibodies to group B itreptococcal and E coli antigens. Br J Obsui GyntcoL 1994:101:107-113.
40. Romero R, Maior M, Oyanum E, Sinori M, Wu YK. Hobbins JC. Is there an assodation between colonliation with GBS and prematurity? J Refnod Med. 1989;34:797-801.
41. Ysncey MK. Duff P, Clark P. Kurtzer T, Premien BH, Kubilis P. Peripartum infertion associated with vaginal group B sireptococcal coloniiatiun. Obstet Gynecol 1994:84:816-819.
42. Schuchat A, Wenger JD. Epidemiology of group B streptococcal disease. Risk feetot!, prevention strategin, and vaccine development. Epidemioi Rev. 1994; 16: 374-402.
43. Cluck L, Wood HF, Fousek MD. Septicemia of the newborn. Pediotr CIm North Am. 1966;13:1131-1148.
44. Pourcyrous M, Bada HS. Korones SB, Basekki V, Wong SP. Significance of serial C-reactlve protein responses in neonatal infection and other disorder*, Pediatrics. 1993;92:43l-435.
45. VesikariT, IsolauriE,TuppurainenN, et al. Neonatal septicaemia in Finland 198185- Predominance of group B streplococca! infections with very early onset. Acta Pediatcr Scan A 1989; 78:44- 50.
46. Klein JO, Marcy SM. Bacterial sepsis and meningitis. In: Remington JS, Klein JO, eds. Infectious Diseases of the Fata and Newborn infam. Philadelphia, Pa: WB SaundersCo; 1990:7601-7656.
47. Thompion PJ, Greenough A, Hird MF, Philpott-Howatd ], Oamsu HR. Nosocomial bacterial infecrionj in very low birth weight infants. Eia i Pediatr. 1992; 151:451-454.
48. Tewin I, Trollfors B, Thmnger K. Incidence and etiology of neonatal septicaemia and meningitis in western Sweden. Ana PoedJatr Senni 1990-.79: 1023-1030.
49. Philip AGS. The changing bee of neonatal infection: experience at a regional medical center. Pediatt Infea Dis J. 1994; 13: 1098- 1102.
50. Fernen P, Cleary PP, Seeds AK. Epidemiology of group-B streptococcal carriage in pregnant women and newborn infants. J Med MionfsoL 1977:10:103-114.
51. Altardice JC, Basten TF, Seshia MMK. Bowman N, Malaidrewici R. Perinatal group B streptococcal colonization and infection. Am J Obstet Gynecoi 1982:142:617-620.
52. Anthony BF, Oleada DM, Hobel C]. Epidemiology of the group B streptococcus: maternal and nosocomial sources for infant acquisitions. J Pediatr. 1979:95:431-436.
53. Garland SM. Early onset neonatal group B streptococcus (GBS) infection: associated obstetric risk factors. Amt N2J OimeiGjnoecol. 1991:31:11 7-118.
54. Biffer KM, Gcaoft SP. Prevention of early onset group B streptocoecal disease with selective intrapartutn chemoprophylaxis. N EngU MaJ. 1983 14; 1665- 1669.
55. Schuchat A, Deaver Robinson K. Plikaytis BD, Zangwill K, Mohle-Boetani J, Wegner JD. Multistate case-control study of maternal risk factors for neonatal group B streptococal disease. Pediair Infect Du). 1994:13:623-629.
56. Geralds W, Cats BP, Hoogkamp-KotstanJe M. Early neonatal group B streptococcal disease: degree of colonisation as an Important determinant. J Infect. 1985;1:119-124.
57. Steigman AJ, Bottone EJ, Hanna BA. Control of perinatal group B streptococcal sepsis: efficacy of single injection of aqueous penicillin at birth. Mt Sinai J Med. 197445:685-693.
58. Lloyd DJ, Belgaumkar TK, Scott KE, Wort AJ, Aicrman K, Krause VW. Prevention of group-B beta-hemolytic streptococcal sepsicaemia in low-birth-weight neonates by penicillin administered within 2 hours of birth. Lancet. 1979;1:7I1-715.
59. Patel DM, Lebalnt MH. Moitison JC, al. Postnatal penicillin prophylaxis and the incidence of group B streptococcal sepsis in neonates. South Mid J. 1994;87: 1117-1120.
60. Siegel JD, McCracken GH, Thelkeld N. DePasse BM, RosenfeU CR. Single-dose pen lei lien prophylaxis of neonatal group- 8- streptococcal disease. Lancet. 1982;11: 1426- 1 430.
61. Merenstein GB, Todd WA, Brown G, Yost CC. Luiier T. Group B B-hemolytic streptococcus: randomized controlled treatment study at term. Otntci Gynecoi. 1980:55:3 15-318.
62. Gardner SE, Yow MD, Leeds W, Thompson PK. Mason EO Jr, Clatk DJ. Failure of penicillin to eradicate group B streptococcal colonization in the pregnant woman. Am J Obstet Generai. 1979;135: 1062- 1065.
63. Paredes A, Wong P, Yow MD. Failure to eradkate the carrier state of group B streptococcus in infants J PedÍatr. 1976;89:19I-193.
64. Lewin EB, Arretey MS. Natural history of group B streptococcus coloniation and its therapy during pregnancy. Am J Obs Wi Gwcol 1981:139:512-515.
65. Yow MD. Mason EO. Leeds W, Thompson PK, Clark DJ, Gardner SE. Ampiculin prevents lntrapartum transmission of group B streptococcus. JAMA. 1979:241:1245-1247.
66. Matorral R, Garcia-Perea A, Omenaca F, Diei-Enciso M, Madero R, Usandiiaga JA. lntrapamim chemoprohpylaxis of early-onset group B streptococcal disease. Ew J Ofatei Gynecol Reprod BioL 1991:40:57-62.
67. Easmon CSF. Hastings MJC, Deeley J, Bbxham B, Rivers RPA. Marwood R. The effect of intrapartum chemoprophylaxis on the vertical transmission of group B streptococci. Br/OisteiGynaecol. 1983; 138:697 -703.
68 Morales WJ, Lim DV, Walsh AF. Prevention of neonaral grixip B strepiococcal sepsis by the use of a rapid screening test and selective intrapartum chemoprophylaxis.Am JOiaietGjwcol. 1 986i 155:979-983.
69. Tupperainen N, Hall man M. Prevention of neonatal group B sBeptococcal disease: intrapartum detection and chemoprophylaxis of heavily colon iitd parturients, Oiuiei GyricR)!. 1989:73:583-587.
70. Ohlsson A, Myhr TL. Intrapartum chemoprophylaxis of perinatal group B streptococcal infections: a critical review of randomiied controlled trials. Am } Obsttl Gynecol. 1994:170:910-917.
71. Amstey MS, Qibbs RS. Is penicillin G a better choice than ampiciUin for prophylaxis of neonata! group B streptococcal infections! Obstet Gyntcol 1994:84:1058-1059.
72. Burman LG, Christensen P, Christensen K, et a!. Prevention of excess neonatal morbidity associated with group B streptococci by vaginal chlorhextdlne disinfection during labour. Lancet. 1992; 340:65-69.
73. Mercer BM, Dillard K1 Fricke JC. Vaginal chlorhex id ine for reduction of intrapartum infection in a high risk population. Presented at the annual meeting of the Infectious Diseases Society for Obstetrics and Gynecology; August 2-5, 1995: Grand Traverse, Michigan.
74. Group B Streptococci infection m Pregnancy. Washington, DC: American College of Obstetricians and Gynecologists; 1992. AGOG technical bulletin No. 170.
75. Committee on Infectious Diseases and Committee on the Fetus and Newborn. Guidelines lor prevention of group B streptococcal (GBS) infection by chemoprophylaxis. Pediatrics. 1992;90:775-778.
76. Minkoff H, Mead P. An obstetric approach to the prevention of early onset group B beta-hemolytic streptococcal sepsis. AmJ Obstet Gynecol. 1 986; 154:97 3-977.
77. Schuchat A. Prevention of group B streptococci! I disease: a public health perspective. MMWR. In press.
78. Landon MB, Harger J, McNellis D, Mercer B, Thorn E. Prevention of neonatal group B streptococcal infection. Ofcstet Gynecol. 1994;84:460-462.
79. Yancey MK1 Duff P. An analysis of the cost-effect ivenéis of selected protocols for the prevention of neonatal group B streprococcal infection. Ofisiet Gynecoi. 1994;83:367 -371.
80. Rouse DJ, Goldenberg RL. Cliver SP, Cutter GR, Mennemevet ST, Fargason CA. Strategies for the prevention of early-onset neonatal group B streptococcal sepsis: a decision analysis. Obstet G^necol. 1994 B1-.463-494.
81. Mercer BM Ramsey RD, Sibai BM. Prenatal screening for group B streptococcus, I: impact of antepartum screening on antenatal prophylaxis and intrapartum care. Am J ObsKf Gynecol. 1995; 173:83 7-841.
82. Mercer BM, Ramsey RD, Sibai BM. Prenatal screening for group B streptococcus, H: impact of antepartum screening and prophylaxie on neonatal care. Am } Obsiel Gynecol. 1995; 173:842-846.
83. Vogel LC, Boyer KM. Gadöla CA, Gotoff SR Prevalence erf type-specific group B strepiococcal antibody in pregnant women. } Ptdiatr. 1980;96: 1047 -1051.
84. Harrison LH, Dwyer DM, Johnson JA. Emergence of serotype V group B streptococcal infection among infants and adults. ] Infect Dis. 1995:171:513.
85. Baker CJ, Rench MA. Edwards MS, Carpenter RJ, Hays BH, Kasper DL. lmmuniiatlon of pregnant women with a polysaccharide vaccine of group BstrepTOCOCCU4. N EngiJ MoL 19881319'.118Q-UaS.
86. Wessels MR, Paotettl LC, Kasper DL. lmmunogenicity in animals of a poly sacche - ride-protein conjugate vaccine against type 111 group B streptococcus. J CIm Invest. 1991:86:1428-1433.
87. Paoletti LC, Kasper DL. MichonF, et al. An olígosaccharide- tetanus conjugate vaccine against type III group B streptococcus. J Bid Chem. 1990;265: 2 78-283.
88. Paoletti LC. Wessels MR, Rodewald AK. Shroff AA, Jennings HJ, Kasper DL. Neonatal mouse protection against infection with multiple group B strcptucoccal (GBS) serotypes by maternal lmmunliation with a tetravaleot GBS polysaccharidetetanus toxoid conjugate vaccine. Infect \rtaraai. 1994:62:3236-3243.
89. Madoff LC, Paoletti LC, Tal JY, Kasper DL. Maternal Immunization of mice with group B streptococcal type IH polysacchande-beta C protein conjugate elicits protective antibody to multiple serotypes. } Clin lnuest. 1 994:94:286- 292.
90. Glvner LB, Baker CJ- Pooled human IgG hyperimmune for type III group B streptococci: evaluation against multiple strains in vitro and in experimenral disease. J InfectDis. 1991;115:969-971.
91. Raff HV, Siscoe PJ, Wolff EA, Maloney G, Shuford W. Human monoclonal antibodies to group B streptococcus. Reactivity and in vivo protection against multiple serotypes. J Exf. Med. 1988; 168:90 5-9 17.
92. Hill JR, Gomales LA, Knappe WA, Fisher GW. Kelsey DK, Raff HV Comparative protective activity of human monoclonal and hyperlmmune polyclonai antibody against group B streptococci. .Ï in/eel Dis. 1991 i163:792-798.
93. Givner LB, Edwards MS, Baker C). A polyclonai human lgu preparation hyperimmune for type III, group B streptococcus: in vitro opsonophagoeytic activity and efficacy in experimental models. J In/eri Dis. 1988:156:724-730.
Classification of Streptococcaccac Involved In Human Diseases