Streptococcus agalactiae, known as group B Streptococcus (GBS), is a Gram-positive coccus commonly found to colonize the human genital and gastrointestinal tracts, and the upper respiratory tract of infants.1 GBS was identified as a significant cause of perinatal morbidity and mortality in the 1970s.2,3 Since that time, implementation of clinical practice guidelines for intrapartum antibiotic prophylaxis has reduced the incidence of neonatal sepsis due to GBS by approximately 80% in the US.1,4 In pregnant individuals, GBS is a known source of asymptomatic bacteriuria, urinary tract infection (UTI), pyelonephritis, upper genital tract infection, postpartum endometritis, pneumonia, sepsis, and bacteremia.5 In neonates/infants, the rate of early-onset invasive GBS infection in 2016 was 0.22 cases per 1,000 live births, in the CDC surveillance study.4
Colonization of pregnant individuals with high colony counts of GBS continues to be a major risk factor for both maternal and neonatal infections.1 High colony counts are defined as at least 105 (100,000) colony-forming units per milliliter (CFU/mL) in the urine. Low colony counts are less than 105 (100,000) CFU/mL.6 A urine culture obtained using a midstream, clean-catch urine sample at 12 to 16 weeks gestational age is the standard of care and the established method for diagnosing asymptomatic bacteriuria (ASB).6 A confirmatory culture of a second sample to control for contamination or transient ASB is recommended, although not typically implemented.6 A culture result of at least 105 (100,000) CFU/mL of GBS is considered positive, and lower counts indicate anogenital colonization.6 Historically, at least 104 (10,000) CFU/mL was used as a threshold for treatment of GBS ASB during pregnancy.4,6 However, the efficacy of treating ASB with low colony counts is controversial, and the data appear limited or contradictory.
The authors conducted an internal case review in a cohort of 459 individuals seeking prenatal care in a family practice clinic using an interprofessional team including: physicians, residents, NPs, nurse midwifes, and pharmacists. The purpose was to analyze the prevalence and local practice patterns related to GBS ASB, and define best evidence-based practice for management of GBS ASB. To examine the associations between the role of screening and treating GBS ASB during pregnancy, the authors searched MEDLINE, CINAHL, PubMed, Embase, and the Cochrane Database of Systematic Reviews. Criteria in the review included: written in English; published between 2015 and 2020; randomized controlled trials, systematic reviews with or without meta-analyses; clinical practice guidelines and studies that addressed relationships between GBS bacteriuria, morbidity and mortality, and antibiotic resistance. The search terms used were: Group B strep* infection, bacteriuria, pregnancy, urinary tract infection, antibiotic resistance, maternal and neonatal morbidity and mortality.
Risks of UTIs, pyelonephritis, chorioamnionitis, and preterm birth
UTIs are among the most common complications experienced during pregnancy.7 The literature reveals 2% to 10% of all pregnancies are complicated by ASB, defined as bacteriuria in the absence of acute UTI symptoms.6-8 If left untreated, as many as 20% to 35% of pregnant individuals with ASB will develop UTI that may lead to pyelonephritis.7,8 While E. coli is the predominant organism found in symptomatic and asymptomatic UTIs, GBS has been isolated in 2.1% to 30% of ASB cases in pregnant individuals.7,9
During pregnancy, individuals with bacteriuria have an increased risk of developing pyelonephritis.10 In the clinical setting, pyelonephritis is diagnosed by a positive urine culture in the presence of fever, urinary symptoms, nausea/vomiting, flank pain, and/or costovertebral angle tenderness.11
Maternal GBS bacteriuria is a marker for anogenital tract colonization, which poses a risk for an infection of the amniotic fluid, membranes, placenta and/or decidua, known as chorioamnionitis, or intra-amniotic infection.11 Clinical signs include fever, uterine tenderness, maternal and fetal tachycardia, purulent amniotic fluid, and maternal leukocytosis.11 GBS is among the most common bacterial organisms implicated as a cause of chorioamnionitis.12 Although the final diagnosis of chorioamnionitis can only be confirmed after birth by culture and histologic examination of the placenta and membranes, in clinical practice, antibiotic treatment is initiated with the onset of maternal fever during labor with any of the other clinical signs noted above.11,12
In the 1960s, Kass reported asymptomatic GBS bacteriuria was associated with preterm birth (prior to 37 weeks gestation).13 In 1984, Moller and colleagues noted premature rupture of fetal membranes and preterm labor were more common in the GBS bacteriuria group.14 A prospective cohort study in 1995 found bacteriuria was associated with preterm birth in the initial analysis, but after further adjustment for medical, demographic, and social factors, association of bacteriuria and spontaneous preterm birth was no longer statistically significant.15,16 A small trial of 69 individuals by Thomsen and colleagues reported a reduction in rates of preterm labor and preterm rupture of the membranes in those treated with penicillin for GBS ASB at any colony count compared with those who received a placebo.17 Historically, these early studies contributed to the practice of treating GBS ASB at low colony counts and informed the practices of most senior obstetricians in the US.18
Internal case review
After approval by the institutional review board, an internal electronic medical record review was conducted. The prevalence of GBS bacteriuria at any colony count was 7.2% among the 459 pregnant individuals initially included and 6.6% among the 458 individuals with complete outcomes data. The authors noted a local practice pattern for treating asymptomatic GBS bacteriuria with colony counts ranging from 102 (100) to over 105 (100,000) CFU/mL.
Among the 31 individuals with GBS bacteriuria with complete data regarding the gestational age at the time of birth, two cases of preterm birth were found with a prevalence of 6.5%, one of which also had low birth weight. Both individuals were treated with antibiotics at the time of GBS bacteriuria identification during pregnancy and again with intrapartum prophylaxis. In the two cases of preterm birth, the infants were admitted to the neonatal ICU and monitored closely. One of these infants was diagnosed with respiratory distress syndrome due to suspected neonatal GBS disease. Both infants received prophylactic antibiotics and had negative blood cultures for GBS disease.
Among the 30 individuals with GBS bacteriuria with complete data regarding maternal and neonatal complications, there were 25 cases of GBS ASB that were treated with antibiotics at the time of identification and again during labor. One of these 30 cases went on to develop chorioamnionitis, with a prevalence of 3.3%. Interestingly, this individual was treated with cephalexin for GBS bacteriuria in early pregnancy based on >=104 (10,000) CFU/mL culture results and again with penicillin after test of cure noted >=105 (100,000) CFU/mL on urine culture. Of note, one patient moved away immediately after delivery; thus, was lost for follow-up of maternal or neonatal complications beyond discharge from the hospital.
None of these 31 cases of GBS bacteriuria developed pyelonephritis during pregnancy. Twenty-five of the 31 individuals with GBS bacteriuria were treated with antibiotics at the time of identification during pregnancy. The six individuals not treated at the time of GBS identification in the urine had low colony counts ranging from 102 (100) to 25 x 103 (25,000) CFU/mL, and none of those six experienced maternal or neonatal complications. The authors also noted five local cases of clindamycin-resistant GBS bacteriuria, one of which was an individual who was allergic to penicillin.
Literature review
A literature review was conducted to provide clarification regarding treatment of GBS bacteriuria during pregnancy and resolve local practice discrepancies. Early studies revealed most individuals receiving antibiotics to treat or eliminate GBS colonization during pregnancy were recolonized within 3 weeks after the antibiotic was discontinued.19 A search of clinical practice guideline collections among the national and international specialty societies found the threshold for diagnosis and treatment of GBS ASB varied in the literature.4,20 (See Summary of GBS bacteriuria studies.) Variable statistical reporting of prevalence, incidence, relative risk, significance, and/or odds ratios was noted and provided when available.
The early studies supported antibiotic treatment of asymptomatic GBS bacteriuria to reduce the incidence of maternal and neonatal morbidity and mortality.13,14,17 Newer retrospective and prospective studies have not demonstrated a higher risk of preterm birth in individuals with GBS bacteriuria.23,24,28,30 In more recent studies that controlled for potential confounders, GBS bacteriuria was not associated with preterm delivery or significant differences in the rates of endometritis, pyelonephritis, or maternal sepsis.28,30 Newer evidence shows a significantly lower risk of pyelonephritis than found in earlier reviews.30 Additionally, newer concerns regarding antibiotic use have surfaced, including antibiotic resistance and changes to the microbiome, potentially increasing the magnitude of harm.6 Recent information regarding the human microbiome suggests asymptomatic bacterial colonization may play a protective role for both pregnant individuals and neonates.6
The 2010 CDC Prevention of Perinatal Group B Streptococcal Disease guidelines recommend against the use of antimicrobials prior to the intrapartum period.4 The guidelines received formal endorsements from the American Academy of Family Physicians (AAFP), the American Academy of Pediatrics (AAP), the American College of Nurse-Midwives (ACNM), the American College of Obstetricians and Gynecologists (ACOG), and the American Society for Microbiology (ASM).4
The CDC transitioned the 2010 guidelines to ACOG and AAP, and those organizations published updated guidelines in 2019.4,32 ACOG and the Infectious Diseases Society of America both published separate updated guidelines in 2019 in support of treating asymptomatic GBS bacteriuria at the time of identification in pregnant individuals at a threshold of at least 105 (100,000) CFU/mL.32,33 Both note that fewer than 105(100,000) CFU/mL GBS bacteriuria indicates maternal anogenital colonization and does not require treatment at the time of identification in asymptomatic individuals, but that individuals with GBS bacteriuria at any colony count during pregnancy should receive intrapartum antibiotic prophylaxis.32 Penicillin is the antibiotic of choice for GBS in nonallergic individuals, and antibiotic sensitivities should be determined for those allergic to penicillin.32 ACOG guidelines now recommend performing universal vaginal-rectal swab GBS screening between 36 0/7 and 37 6/7 weeks of gestation, if not previously detected in the urine with positive vaginal-rectal culture as an indication for an individual to receive intrapartum antibiotics.32
In 2016 and 2018, the World Health Organization (WHO) issued recommendations for a 7-day antibiotic regimen for all pregnant individuals with ASB to prevent persistent bacteriuria, preterm birth, and low birth weight.34 The WHO's recommendation was based on the evidence from the Cochrane Review that included 14 trials involving more than 2,000 individuals.21 In these trials, bacteriuria was defined as at least one clean-catch, midstream or catheterized urine specimen with more than 105 (100,000) CFU/mL on culture, but some of the included studies defined ASB with lower colony counts.21,31,35 The 2015 and 2019 Cochrane Reviews were limited by the quality of the studies due to numerous different designs, lack of information about randomization methods, different definitions of ASB based on varying colony counts, low statistical power, and bias.21,31,35
GBS remains susceptible to the beta-lactams, but individuals with a penicillin allergy should have antibiotic sensitivities determined.32 The preferred antibiotic should be selected based on drug safety considerations and local resistance patterns. (See Antibiotics for treatment of GBS bacteriuria in pregnancy.) With antibiotic resistance to erythromycin, clindamycin, and cefazolin on the rise, definitive antibiotic therapy should be based on culture and guided by the sensitivity pattern.37-39 Although nitrofurantoin typically has been avoided in the first trimester due to potential risk of birth defects, it may be prescribed in the second and early third trimesters.35,37 Nitrofurantoin and trimethoprim/sulfamethoxazole should be avoided after 36 weeks gestational age when they may increase neonatal jaundice and predisposition to kernicterus if taken in the last week before delivery.35 Additionally, nitrofurantoin has been associated with a low risk of fetal or neonatal hemolytic anemia if the mother has glucose-6-phosphate deficiency. Thus, the drug should be prescribed with caution.37
The CDC considers antibiotic resistance a global threat with no new antibiotics having been produced since the 1980s and has advised that alternative and/or innovative therapies need to be considered.4 New studies are investigating maternal and perinatal outcomes to explore the efficacy, risks, and benefits of potential maternal GBS vaccination.30,40
Conclusion
The authors reviewed the literature to provide evidence-based recommendations regarding the management of GBS bacteriuria in pregnant individuals and found contradictory or inconclusive results with substantial biases. Caution is advised when drawing conclusions for treating GBS ASB at low colony counts. The review clearly noted that treatment at the time of identification of GBS ASB with colony counts lower than 105 (100,000) CFU/mL is not recommended, as this reflects maternal anogenital colonization and may lead to antibiotic resistance. Antimicrobial stewardship is a must. Nontreatment of GBS bacteriuria at colony counts lower than 105 (100,000) CFU/mL in asymptomatic individuals offers an important opportunity to decrease inappropriate antimicrobial use.
REFERENCES