The Times They Are A-Changin'

I recently had a call about a patient with postpartum fever and concern for postpartum endometritis. The Obstetrics team wanted an opinion after the blood cultures grew E. coli. The patient was on empiric therapy with gentamicin, ampicillin, and metronidazole, and I thought to myself: maybe it’s time to write about peripartum infections and why it might be time to rethink the traditional empiric regimen of gentamicin, ampicillin, and metronidazole (or clindamycin)—especially now that we have better tools and antibiotics to manage these infections. I also came across a great article that I think is worth mentioning.

What Do We Mean by Peripartum Infections?

To start, it’s important to define what we mean by peripartum infections. Broadly, there are three main clinical syndromes: intra-amniotic infection (chorioamnionitis), postpartum infection (postpartum endometritis), and postabortal infection. These diagnoses are largely clinical (fever, abdominal pain, recent procedure, etc.), often made without pathologic or microbiologic confirmation, and without standardized diagnostic criteria.

Pathophysiology: How Do These Infections Happen?

Most peripartum infections result from ascending bacterial invasion from the lower genital tract. In chorioamnionitis, postpartum endometritis, and postabortal infections, bacteria ascend into what is normally a sterile environment, leading to infection of the fetal membranes, amniotic fluid, placenta, and/or decidua once natural barriers are disrupted. This is the mechanism in the majority of cases. The exception is rare hematogenous seeding from maternal bacteria, most classically Listeria.

Which Organisms Are Involved?

Given this pathophysiology, it makes sense that these infections are usually polymicrobial. The common organisms include:

• Gram positives: Group B Streptococcus, coagulase-negative staphylococci, Enterococcus, Listeria monocytogenes, other Streptococcus spp.

• Gram negatives: Escherichia coli and other Enterobacterales, Neisseria gonorrhoeae, Sneathia, Acinetobacter spp.

• Anaerobes: Bacteroides spp., Gardnerella vaginalis, Prevotella spp., Fusobacterium spp.

• Atypicals: Chlamydia trachomatis, Mycoplasma spp., Ureaplasma spp.

Why the Traditional Regimen?

So, what’s special about these infections compared to other intra-abdominal or gynecologic infections? Historically, empiric treatment has been ampicillin, gentamicin, and clindamycin (later metronidazole, given clindamycin’s declining anaerobic activity). This regimen was developed based on in vitro studies from the 1980s and has been shown to work clinically. But there are very few modern head-to-head comparisons with more contemporary regimens such as beta-lactam/beta-lactamase inhibitor combinations.

Safety Concerns in Pregnancy and Breastfeeding

One reason clinicians may hesitate to change is the special context—pregnancy and postpartum care—where antibiotic choice must consider fetal and neonatal safety. For example, in chorioamnionitis (when delivery isn’t imminent), or in breastfeeding mothers with postpartum endometritis, antibiotic safety is a real concern. But we now have good data supporting the safety and efficacy of penicillins, beta-lactam/beta-lactamase inhibitors, and cephalosporins in these populations.

The NEC Controversy

There is some historical concern about amoxicillin–clavulanate increasing the risk of neonatal necrotizing enterocolitis (NEC), based on the ORACLE I and II trials from the early 2000s. Later studies, however, have shown conflicting results.

• Weintraub et al. found an association between peripartum ampicillin and NEC (PMID: 22157626).

• Kenyon et al. (11 RCTs, 6229 infants) found no overall increase in NEC with prenatal antibiotics, except for amoxicillin–clavulanate (PMID: 24297389).

• Reed et al. reported that antibiotics within 72 hours before delivery may actually reduce NEC incidence (PMID: 29551319).

• A French study more recently questioned the association altogether (PMID: 34954405).

So, while we should recognize the potential consequences of disrupting the microbiome, the evidence does not justify excluding beta-lactam/beta-lactamase inhibitors outright—especially since this concern is more relevant for preterm rupture of membranes and chorioamnionitis, and much less for postpartum endometritis or postabortal infections.

Time to Rethink the Approach

Which brings us to now. As Pek et al. argue in their article, it may be time to update our empiric therapy for peripartum infections. Their recommendation is similar to intra-abdominal infections:

• First line: piperacillin–tazobactam

• Second line: ceftriaxone plus metronidazole

This covers the major organisms. In the rare case of invasive listeriosis, piperacillin–tazobactam should be effective, (studies has shown that piperacillin tazobactam is effective (PMID: 27345176). So with this in mind, I don’t think there is much difference, that merits this regimen maybe except that all the data with these infections are based in ~1980’s studies, a topic for another post regarding listeria) though one could transition to ampicillin plus gentamicin if desired.

Final Thoughts

In short: the gentamicin/ampicillin/metronidazole (or clindamycin) combination is outdated. It adds toxicity without clear benefit, and it’s based on evidence that hasn’t been revisited in decades. Like so many other areas of medicine, it’s time for peripartum infections to catch up with the times.

Musical Coda