to the editor: The article by Dr. Keenan on screening for group B streptococcus (GBS) and the accompanying editorial by Dr. Deutchman highlight an important dilemma faced by family physicians who practice obstetrics.^1,2 Failure to diagnose and treat pregnant women with GBS places neonates at risk for life-threatening infection and physicians at risk for litigation.

The Centers for Disease Control and Prevention (CDC) offers two strategies for the prevention of neonatal GBS infection.³ Universal screening with a rectovaginal culture taken at 35 to 37 weeks' gestation detects most maternal carriage of GBS but is comparatively expensive. Relying on risk factors alone is less expensive but offers less sensitivity and specificity in the differentiation between colonized pregnant women and uncolonized pregnant women. GBS culture remains the “gold standard” for the detection of colonization.

The method of collection of GBS specimens that is described by the CDC may present a “convenience” barrier to physicians. Pregnant patients must undress, assume the lithotomy position and have a rectovaginal swab obtained by the caregiver, often with a chaperone in attendance. We offer an alternative. In an unpublished study of 241 pregnant women, we used a self-collection method for the GBS rectovaginal culture and found that women were as likely as their physicians to collect a sample that would show maternal carriage of GBS [79 percent versus 83 percent, Fisher's exact test = 1.263(1), single tailed P = 0.365]. Other investigators have found similar results and have also found that pregnant patients prefer self-collection of samples to collections made by the physician.^4,5

Physicians who prefer screening for GBS with a rectovaginal culture should consider the use of patient-collected specimens obtained when the pregnant patient collects her urine sample. This method does not require the patient to undress, requires no chaperone and may increase the time available for other matters during the office visit. For physicians who do not collect a GBS culture, this method may remove the “convenience” barrier to collection of GBS cultures.

in reply: Any technique that helps in the identification of infants who are at risk for infection with group B streptococcus would be welcome, including self-collection techniques if they are found to have an acceptable degree of sensitivity and specificity after appropriate study.

editor's note: This letter was sent to the author of “Prevention of Neonatal Group B Streptococcal Infection,” who declined to reply.