Screening programs have been credited with the drop in neonatal disease caused by group B streptococci (GBS), but these infections continue to pose significant risks. The identification of GBS carriers must be followed by effective antibiotic therapy. Approximately 27 percent of mothers currently receive antibiotics during labor and delivery to prevent transmission of GBS. Up to 12 percent of these women report allergy to the principal chemoprophylactic agent, penicillin, and are treated with erythromycin or clindamycin. Many strains of GBS have developed resistance to these antibiotics, and the proportion of resistant strains is believed to be increasing rapidly. Manning and colleagues studied women attending a large medical school clinic for obstetric care to determine the extent of resistance to erythromycin or clindamycin in GBS and to identify any factors associated with resistant strains.

Positive GBS isolates were obtained from 103 pregnant women. These isolates were tested for susceptibility to 10 antibiotics: penicillin, ampicillin, levofloxacin, quinupristin-dalfopristin, cefazolin, imipenem, vancomycin, clindamycin, erythromycin, and linezolid. Strains that showed complete or intermediate resistance were retested to ascertain the minimum inhibitory concentration. Data on the mother, fetus, and pregnancy were obtained from medical records, and pertinent variables were correlated with antibiotic resistance. Stepwise logistic regression was used to identify the strongest predictors of colonization with resistant strains.

Thirty isolates (29 percent) were resistant to erythromycin, and 22 (21 percent) were resistant to clindamycin. All of the clindamycin-resistant isolates were also resistant to erythromycin. Eight isolates were resistant only to erythromycin. The isolates were susceptible to the other antibiotics tested, but in eight cases, resistance to penicillin was intermediate. The most common GBS serotypes were V (21 cases), III (20 cases), and Ia (19 cases). Women carrying serotype V strains were 13 times more likely to have an erythromycin- or clindamycin-resistant strain than women carrying serotype II strains (41 versus 3 percent).

In the initial analysis, unmarried women younger than 25 years had a higher rate of resistant GBS strains. Black ethnicity also was associated with resistance. Stepwise logistical regression identified black ethnicity and serotype V strains as statistically associated with resistance to erythromycin or clindamycin.

The authors conclude that a significant percentage of GBS isolates in pregnant mothers are resistant to erythromycin or clindamycin. They recommend that cefazolin be used as the first choice in mothers with contraindications to penicillin.