RECURRENT GROUP B streptococcal bacteremia has been well described in infants as well as in adults. However, recurrent group B streptococcal bacteremia in association with menstruation has not been reported. Previous studies have shown the presence of underlying co-morbidities, the two most common being non-hematologic malignancies and diabetes, in patients with recurrent bacteremia. We describe a diabetic patient with two episodes of group B streptococcal bacteremia 12 weeks apart, in whom both episodes were associated with the menstrual cycle. Increases in vaginal colonization during menses and the gradual increase in adherence of the bacteria in the first half of the menstrual cycle may explain the recurrence during menses. The short interval between the two episodes may imply a relapse of the first infection. The fact that she is a diabetic, may have increased her risk for colonization of the genital tract, and may have played a role in her recurrent illness. The development of treatment modalities that eradicate colonization of both the gastrointestinal and genital tracts, thereby preventing future relapses, would potentially benefit such patients.
Streptococci were first recognized as a cause of human disease in the early 1800s. Group B streptococci (GBS) were noted as human pathogens in 1938 when Fry  described three cases of fatal puerperal sepsis caused by the organism. Also known as Streptococcus agalactiae, GBS exhibit B-hemolysis on blood agar. They are common inhabitants of the lower reproductive and gastrointestinal tracts but may also be cultured from skin and deep tissues in disease. The organism is the most common infectious cause of neonatal mortality in the United States , usually attributed to neonatal sepsis and meningitis. Infections in newborns may be the result of in utero infection or may follow rupture of the amniotic membranes. Maternal infections with GBS are also described, notably bacteremia, pyelonephritis, chorioamnionitis, and postpartum endometritis [3–5]. Recent reports indicate that 33%–53% of severe GBS infections are in nonpregnant adults [4–6]. Recurrence of bacteremia also occurs and putative risk factors have been described. Many such patients have primary bacteremia, but in most cases, an underlying focus of infection has been observed [7,8]. To our knowledge, this is the first reported case of a young diabetic woman with recurrent primary bacteremia that appears to be related to menstruation.
The patient was a 28-year-old African American woman, a known diabetic for 6 years, requiring insulin for the past 4 months. She presented with fever, rigors, and chills beginning 2 days before admission. Her symptoms improved initially with acetaminophen, but later began to worsen, leading to her admission to hospital. She denied cough, shortness of breath, abdominal pain, vomiting, diarrhea, dysuria, or vaginal discharge. She was menstruating at presentation. There was no oral contraceptive use. Her medical history was significant for an episode of group B streptococcal bacteremia 3 months earlier that occurred while she was menstruating. At that time, she was admitted to another hospital and was treated with intravenous penicillin G for 2 days followed by oral penicillin for an additional 8 days.
Upon admission to our hospital, her temperature was 100.4°F. The only significant physical findings were morbid obesity and a midsystolic click without a murmur. Two sets of blood cultures on the date of admission were positive for group B streptococcus; one set was positive for the same organism 2 days later. Therapy with intravenous penicillin G was initiated 36 hours after admission at a dose of 2 million units every 4 hours. Her bacteremia cleared and she rapidly became asymptomatic. A transthoracic echocardiogram performed on hospital day 5 failed to reveal any cardiac abnormality.
GBS infections were reported infrequently until the 1960s when several authors indicated that disease attributed to this organism was more common than previously appreciated . GBS colonize the respiratory, gastrointestinal, and genital tract , and a variety of underlying diseases increase the risk for invasive disease . Women 20 years of age or younger with sexual experience or with use of intrauterine devices are known to have increased risk . Diabetes mellitus is also recognized as one of the most common predisposing factors for colonization and infection . Pregestational diabetic women have a higher rate of colonization with GBS compared with gestational diabetics, suggesting that carbohydrate intolerance is an independent risk factor for GBS colonization . Infection may also be related to colonization of the genital tract. However, recurrent GBS bacteremia associated with menstruation has not been described.
Vaginal colonization by GBS is associated with the menstrual cycle. There is a gradual increase in adherence during the first half of the cycle reaching a maximum at day 14, followed by a decrease to a low level that persists throughout the second half of the cycle [12–14]. Correlation is seen with variation in the pyknotic index of the vaginal epithelium . The prevalence of GBS colonization is not related to oral contraceptives. However, when hormone-dependent adhesion of GBS to human vaginal cells was studied, it was found that as oral contraceptives abolish cyclic changes and create a negative feedback to production of gonadotrophic hormones from the pituitary, there is a tenfold reduction in adhesion of GBS to the vaginal cells. This effect correlates with low leutenizing hormone/follicle stimulating hormone levels in the serum. In the same study, it was also seen that GBS type III strains possessed the greatest ability to adhere to vaginal cells and were less affected by acidic pH . The fact that in our patient bacteremia occurred twice at the onset of her menses without any evidence of associated infection strongly suggests a genital tract source of colonization. Gonococcal infection is also known to disseminate during menstruation . Other microorganisms, namely Escherichia coli, viridans streptococcus, and other aerobic streptococci have been shown to cause sepsis in obstetric patients . A good correlation of blood isolates and genital cultures has been shown . The endocervix is probably the site of invasion in patients with bacteremia from the genital tract. This was demonstrated in an animal study that found the uterine wall and amniotic fluid to be sterile before inoculation of bacteria . Endometritis and endoparametritis in obstetric patients were also found to lead to GBS bacteremia in a small subset of women after delivery . This suggests that the organism enters the bloodstream after colonization of the endometrial tissue. It is possible that bacteremia is common during menstruation, much like the bacteremia associated with toothbrushing or straining at stool, but that the organisms are spontaneously cleared from the bloodstream. The absence of vaginal or endometrial cultures in the present case prevent us from offering unequivocal proof of this association. Nevertheless, the sloughing of the endometrial surface associated with menstruation may be considered analogous to the postpartum state and could provide an access point for GBS, which commonly colonize the genital tract. Because both episodes of bacteremia were temporally related to her menses, we speculate that the genital tract was the source of bloodstream invasion in our patient.
Recurrent bacteremia with GBS has been carefully studied. In these patients, the second episodes were most often attributed to either inadequate treatment of endocarditis or osteomyelitis that was not previously recognized. Although in our patient a systolic click was detected on initial examination, a transthoracic echocardiogram failed to disclose a significant lesion. Unfortunately, because of the occurrence of sepsis during her menses, neither cervical nor vaginal cultures were performed. Among those with recurrent bacteremia, identical strains were often isolated from patients whose initial infection was relatively recent (mean duration 14 weeks), whereas a different strain was found in those with remote infection (mean duration 43 weeks). Nevertheless, the high frequency with which identical strains were found in each situation suggests that many cases of recurrent GBS infection are actually relapses of previous infection . Another potential explanation is failure to eradicate carriage with subsequent reinfection. Although the most appropriate antibiotic regimen to eradicate carriage is unknown, it is probable that the 8-day regimen of penicillin our patient received for her initial infection was inadequate. The majority of her initial treatment was via oral medication which might be expected to eradicate intestinal colonization but not other sites. This supports our hypothesis that she maintained carriage in the genital tract. It seems likely that in each instance, with the initiation of menstruation, the organism gained access to the bloodstream through the sloughed endometrium leading to sepsis.
GBS are uniformly susceptible to penicillin G, which is the drug of choice when the diagnosis is established. High doses of penicillin G are recommended because the MIC for GBS is four-to tenfold greater than that for Group A streptococci, and also because the inoculum size influences the in vitro susceptibility to penicillin G . Eradication of mucosal colonization by GBS in infants with the use of rifampin is shown to have been unsuccessful despite the low MIC to rifampin of the isolates tested . Our patient was treated with high dose penicillin G for 4 weeks in an effort to eradicate her infection and also to attempt to eliminate carriage. To our knowledge, she has not had relapse after 1 year of follow-up.
Recurrent group B streptococcal bacteremia, which is seen in patients with underlying co-morbidities, may also occur after the onset of the menses, most likely because of the increases in adherence of the bacterium to vaginal cells during this time in the cycle. Because second episodes of bacteremia are usually caused by deep-seated infections like endocarditis and osteomyelitis, a thorough investigation for a focus of infection is warranted. Even in the absence of a focus of infection, prolonged parenteral therapy may be necessary for complete eradication of colonization from the gastrointestinal and genital tracts.
1. Fry, RM. Fatal infections by hemolytic streptococcus group B. Lancet 1;1938:199–201.
2. Zangwill KM, Schchat A, Wenger JD. Group B streptococcal disease in the United States, 1990: report from a multistate active surveillance system. Mor Mortal Wkly Rep CDC Surveill Summ 1992; 41:25–32.
3. Yancey MK, Duff P, Clark P, Kurtzer BH, Kublis P. Peripartum infections associated with vaginal group B streptococcal colonization. Obstet Gynecol 1994; 84:816–9.
4. Farley MM, Harvey RC, Stull T, Smith JD, Schuchat A, Wenger JD, Stephens DS. A population-based assessment of invasive disease due to group B streptococcus in nonpregnant adults. N Engl J Med 1993; 328:1807–11.
5. Schwartz B, Schuchat A, Oxtoby MJ, Cochi SL, Hightower A, Broome CV. Invasive group B streptococcal disease in adults. JAMA 1991; 266:1112–4.
6. Opal SM, Cross A, Palmer M, Almazan R. group B streptococcal sepsis in adults and infants. Arch Intern Med 1988;148:641–5.
7. Colford JM, Mohle-Boetani J, Vosti KL. Group B streptococcal bacteremia in adults. Medicine (Baltimore) 1995; 74:176–90.
8. Harrison LH, Ali A, Dwyer DM, et al. Relapsing invasive group B streptococcal infection in adults. Ann Intern Med 1995; 123:421–7.
9. Baker CJ. Summary of the workshop on perinatal infections due to group B streptococcus. J Infect Dis 1977; 136:137–52.
10. Baker CJ, Goroff DK, Alpert S, et al. Vaginal colonization with group B streptococcus: a study in college women. J Infect Dis 1977; 135:392–7.
11. Ramos E, Gaudier FL, Hearing LR, Del Valle GO, Jenkins S, Briones D. Group B streptococcus colonization in pregnant diabetic women. Obstet Gynecol 1997; 89:257–60.
12. Botta GA. Hormonal and type-dependent adhesion of group B streptococci to human vaginal cells. Infect Immun 1979; 25:1084–6.
13. Forslin L, Danielsson D. In vitro
studies of the adherence of N. gonorrheae
and other urogenital bacteria to vaginal and uroepithelial cells with special regard to the menstrual cycle. Gynecol Obstet Invest 1980; 11:327–40.
14. Zawaneh SM, Ayoub EM, Baer H, Cruz AC, Spellacy WN. Cyclic variation in the adherence of group B streptococci to human vaginal epithelial cells. Am J Obstet Gynecol 1981; 140:381–6.
15. Gibbs RS, Blanco JD. Streptococcal infections in pregnancy: a study of 48 bacteremias. Am J Obstet Gynecol 1981; 140:405–11.
16. Evaldson G, Malmborg AS, Nord CE, Ostensson K. Bacteroides fragilis, Streptococcus intermedius
and group B streptococci in ascending infection of pregnancy. Gynecol Obstet Invest 1983; 15:230–41.
17. Edwards MS, Baker CJ. Streptococcus agalactiae.
In: Mandell et al. editors. Principles and practice of infectious diseases. 4th ed. Philadelphia, Churchill Livingston; pp. 1835–45.
18. Feldman WE. Concentrations of bacteria in cerebrospinal fluid of patients with bacterial meningitis. J Pediatr 1976; 88:549–52.
19. Fernandez M, Albanyan EA, Rench MA, Baker CJ. Failure of rifampin to eradicate group B streptococcal (GBS) colonization in infants with invasive disease [abstract no. 41]. In: Program and abstracts of the 36th Annual Meeting of the Infectious Diseases Society of America; Colorado. 1998. p. 82.