 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
We appreciate the interest of Drs Edwards and Duff in our article. We reported that previously healthy term infants younger than 90 days who were hospitalized because of serious bacterial infections (SBIs) were more likely to have been exposed to intrapartum antibiotics (IPAs) than were age-matched healthy controls. We also found that the causative organisms in infants with an SBI were more likely to be ampicillin-resistant if the infant's mother had been exposed to an IPA, which in our cases was mostly ampicillin.1 We argue that these findings support ascertaining a history of IPA exposure in young infants who are suspected of having an SBI and for recommending that, consistent with national guidelines, penicillin be used as an IPA rather than ampicillin. Edwards and Duff apparently believe that the American College of Obstetricians and Gynecologists, American Academy of Pediatrics, and Centers for Disease Control and Prevention recommendations for penicillin over ampicillin as an IPA are premature because they are not "evidence-based." They base this, at least in part, on the results of their study that compared the prevalence of ampicillin resistance among maternal genital tract flora 8 to 36 hours after delivery in women who received either ampicillin or penicillin as IPAs. They found that the proportion of women whose genital tract flora became ampicillin-resistant increased in both groups but that the increases between the 2 cohorts were similar.2 They suggest that their study provides "level 1" evidence that penicillin IPA does not decrease the risk to infants of acquiring an SBI with ampicillin-resistant Gram-negative bacteria.
Edwards and Duff seem to be unfamiliar with 2 important concepts that pediatricians consider when they evaluate infants with fever who are younger than 90 days: the illnesses that are included in the designation "SBI" and the difference between early- and late-onset SBI. Unlike cystitis in adult women, a febrile urinary tract infection in an infant in this age group is considered to be an SBI, the treatment of which usually requires hospitalization and intravenous antibiotics and is an indication for additional evaluation including a voiding cystourethrogram and renal ultrasound or other imaging.3,4 Infants who develop an SBI after 7 days are categorized as having "late-onset" sepsis, which differs in several important ways from early-onset sepsis.5
We have several concerns about the results of the study conducted by Edwards and Duff and their colleagues. First, the study was not powered (according to their methods section) to detect a difference of an increase in resistance between the ampicillin and penicillin cohorts unless the difference was 
50%. For an outcome as severe as late-onset SBI, most pediatricians would be interested in knowing whether the difference in resistance could be lowered by a much smaller amount. In addition, the study outcome was based on bacterial culture. Culture-based studies of vaginal flora are inherently limited and unable to fully assess the changes in ecology after antibiotic administration.6 Molecular-based methods that allow quantification of individual bacterial species are preferred. There may have been significant differences in the 2 groups in the Edwards and Duff study (ampicillin versus penicillin) in colonization with anaerobic bacteria including Lactobacillus, Clostridium, and Bifidobacterium species that could influence the acquisition of bacterial flora and normal microbial succession of the infant gut.7,8 These factors could alter the susceptibility of the infant to colonization with antibiotic-resistant flora. Finally, the study did not address long-term effects on maternal bacterial colonization or provide any information regarding infant colonization. We believe it is a considerable stretch to suggest that the outcome of vaginal colonization at 8 to 36 hours can be extrapolated to providing "level 1 evidence" regarding the cause of an SBI in an infant weeks or months later.
Edwards and Duff are correct in noting that the number of mothers in our study who received penicillin was small. We noted that our finding that there was no increase in risk for infant infection after maternal penicillin treatment "should be interpreted cautiously, however, because the numbers of mothers who received penicillin was small, and there may be insufficient power to detect a small but possibly significant difference."1
We disagree with Edwards and Duff's implicit suggestion that, based on their study, ampicillin is an acceptable agent as an IPA. Ampicillin provides no advantage over penicillin for the prevention of early-onset group B Streptococcus infections in newborns, the purpose of IPAs. We believe that our study supports the national recommendations for the use of penicillin. Their study does suggest that there may be risks associated with penicillin IPAs, particularly with respect to early-onset SBI. We suggest that those caring for infants with an SBI should not be reassured by learning that the infant's mother had received penicillin as an IPA.
REFERENCES
Related articles in Pediatrics:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
