Correspondence

Prevention of Early-Onset Group B Streptococcal Disease in Neonates

N Engl J Med 2002; 347:1798-1799November 28, 2002

Article

To the Editor:

Schrag et al. (July 25 issue)1 report the results of a retrospective cohort study undertaken in an attempt to provide a definitive answer about whether to screen pregnant women for group B streptococcus. Unfortunately, such an answer can be determined only by randomized, controlled trials.

There were significant differences between the groups — notably, differences in ethnic background, the rate of preterm delivery, and the adequacy of prenatal care, as well as geographic differences in approach. There was a significant difference in antibiotic use between the groups, suggesting that the assumption that 98 percent of obstetricians had a plan to deal with group B streptococcal infection is flawed. Women with group B streptococcal bacteriuria or previous invasive disease were included in the analysis, although adherence to the recommendations of the Centers for Disease Control and Prevention would have removed them from consideration for screening. There was clearly an attempt to control for confounders, but the inability to control for unseen confounders remains a concern.

A more appropriate conclusion would be that the study provides compelling evidence in support of conducting a randomized, controlled trial to discern the best approach for preventing group B streptococcal infection. Our concern is that physicians and groups making recommendations will interpret these data as evidence that screening is the best approach and that this will become the standard of care before the definitive studies can be conducted.

David L. Fay, M.D.
Chris J. Wenninger, M.D.
Medical College of Wisconsin, Waukesha, WI 53188
david.fay@phci.org

1 References

1
Schrag SJ, Zell ER, Lynfield R, et al. A population-based comparison of strategies to prevent early-onset group B streptococcal disease in neonates. N Engl J Med 2002;347:233-239
Full Text | Web of Science | Medline

To the Editor:

Before the current recommendations are changed, there are three important considerations. First, the final aim of intervention should be a reduction in the overall rate of neonatal sepsis rather than a decrease in the rate of sepsis caused by one organism and an increase in that caused by another organism. Second, attention must be paid to the efficacy and feasibility of implementing the screening approach in areas where group B streptococcus is indigenous — such as our hospital, where 30 percent of women have had no prenatal care. Finally, the cost–benefit analysis must be considered. Rouse and colleagues1 estimate that the screening approach costs $15,782 to $50,105 more per 1000 live births (assuming a reduction in group B streptococcal sepsis similar to that found by Schrag et al.). These increased costs do not include the treatment costs associated with the increase in the rate of Escherichia coli sepsis.

Meena Khandelwal, M.D.
Ozgur H. Harmanli, M.D.
Temple University School of Medicine, Philadelphia, PA 19140
mkhandel@nimbus.temple.edu

1 References

1
Rouse DJ, Goldenberg RL, Cliver SP, Cutter GR, Mennemeyer ST, Fargason CA Jr. Strategies for the prevention of early-onset neonatal group B streptococcal sepsis: a decision analysis. Obstet Gynecol 1994;83:483-494
CrossRef | Web of Science | Medline

Author/Editor Response

The authors reply:

To the Editor: Drs. Fay and Wenninger express concern about our study design. Although randomized, controlled trials are viewed as the gold standard, they may be impossible to conduct once competing interventions have been widely adopted. In 1994, direct comparison of the risk-based and screening-based approaches was not considered feasible, because hundreds of thousands of pregnant women would have needed to be randomized.1 Our observational study successfully evaluated the relative effectiveness of the interventions in more than 600,000 births. Although our study was not randomized, the analysis adjusted for differences between cohorts, and subgroup analyses assessed residual confounding. Our results were robust and suggest that the benefits of screening resulted from prophylaxis in colonized women without intrapartum risk factors.

We agree with Drs. Khandelwal and Harmanli that multiple factors influence the development of new recommendations for prevention. A critical review of data from our study, together with other considerations, led the Centers for Disease Control and Prevention to issue new guidelines recommending universal prenatal screening for group B streptococcus in August 2002.2 These guidelines recommend the risk-based approach only for women with unknown culture status at the time of labor. Analyses have suggested that both the screening-based strategy and the risk-based strategy are cost effective; we found that universal screening would prevent significantly more cases. Thus, even at institutions serving a sizable population of women with no prenatal care, the new guidelines recommend that patients who do use such care should have the opportunity to benefit from prenatal screening.

The concern that intrapartum prophylaxis may lead to increases in other causes of neonatal sepsis merits consideration. Our data suggest that full implementation of either strategy would result in a similar rate of antibiotic use; thus, a recommendation for universal screening should not heighten this concern. To date, increases in other causes of neonatal sepsis have been detected only among the small proportion of neonates who are preterm or have very low birth weight.3,4 Such increases have not been causally linked to the intrapartum use of antibiotics. Population-based surveillance has revealed a decline in the overall rate of early-onset sepsis that is consistent with the reduction in the number of cases of group B streptococcus infection.5 We continue to monitor these trends and view prenatal screening and intrapartum antibiotic prophylaxis as an interim approach to the prevention of perinatal group B streptococcal disease, until vaccines become available. Future randomized, controlled trials related to the prevention of group B streptococcal disease may be most useful if they address the efficacy and safety of vaccination rather than strategies for intrapartum prophylaxis.

Stephanie J. Schrag, D.Phil.
Anne Schuchat, M.D.
Centers for Disease Control and Prevention, Atlanta, GA 30333
zha6@cdc.gov

Janet Mohle-Boetani, M.D., M.P.H.
California Department of Health Services, Berkeley, CA 94707

5 References

1
Landon MB, Harger J, McNellis D, Mercer B, Thom EA. Prevention of neonatal group B streptococcal infection. Obstet Gynecol 1994;84:460-462
Web of Science | Medline

2
Prevention of perinatal group B streptococcal disease: revised guidelines from CDC. MMWR Morb Mortal Wkly Rep 2002;51:1-22
Medline

3
Stoll BJ, Hansen N, Fanaroff AA, et al. Changes in pathogens causing early-onset sepsis in very-low-birth-weight infants. N Engl J Med 2002;347:240-247
Full Text | Web of Science | Medline

4
Hyde TB, Hilger TM, Reingold A, Farley M, O'Brien KL, Schuchat A. Trends in the incidence and antimicrobial resistance of early-onset sepsis: population-based surveillance in San Francisco and Atlanta. Pediatrics 2002;110:690-695
CrossRef | Web of Science | Medline

5
Baltimore RS, Huie SM, Meek JI, Schuchat A, O'Brien KL. Early-onset neonatal sepsis in the era of group B streptococcal prevention. Pediatrics 2001;108:1094-1098
CrossRef | Web of Science | Medline

Citing Articles (1)