乌鸦传媒

For women in labor, epidural anesthesia offers relief from some of the worst pain that anyone can experience. But could it pose a risk for the baby?

Yes, said Chinese researchers who in JAMA Network Open. Infections among newborns occurred at a rate of 4.4% when mothers received epidurals, versus 1.8% without the treatment (risk difference 2.6 percentage points, 95% CI 2.2-3.0).

That worked out to a relative risk of 1.81 (95% CI 1.56-2.11) when adjusted for labor duration, reported Lijie Jia, MD, PhD, of Shanghai Jiao Tong University, and colleagues.

Sepsis rates were also significantly greater in the epidural cohort (adjusted RR 2.51, 95% CI 1.19-5.28) but such events were rare, occurring in only 0.2% of the approximately 15,000 births involving epidurals.

Other neonatal outcomes including pneumonia, necrotizing enterocolitis, and low Apgar scores didn't differ significantly between groups.

But some adverse maternal outcomes were more common with epidurals. These included fever (15.4% vs 3.8%) and chorioamnionitis (7.4% vs 1.8%). Rates of postpartum hemorrhage, however, were nearly identical. (In most cases, adjustment for labor duration attenuated the differences in raw rates somewhat, the researchers noted.)

For the study, Jia and colleagues analyzed records of first-time mothers giving birth at Shanghai's International Peace Maternity and Child Health Hospital from 2013 to 2018. Births occurring prior to 37 weeks' gestation, those ending in cesarean section, and those in women with previous pregnancies were excluded. The investigators ended up with 15,401 births involving epidurals and an equal number of non-epidural controls.

Women could elect to receive the anesthesia when reaching cervical dilation of at least 3 cm. The authors noted that, in Shanghai, epidurals are partly covered by government insurance and therefore cost is not a reason to skip them, although cultural beliefs could be.

The control group was selected on the basis of propensity-matching for neonatal infection risk. The system for doing so was established in a preliminary analysis, via "a weighted cohort of parturient women with and without epidural analgesia use but similar in demographics and maternal comorbidities before labor and delivery," Jia and colleagues explained.

Mean maternal age was about 30, with body mass index averaging 26.0, in both of the final cohorts. About 7% had diabetes, 2% had thyroid disorders, and 4% were positive for group B Streptococcus. Mean newborn birthweight was 3.3 kg in both groups.

Neonatal infection was determined according to Chinese government criteria, which do not necessarily require positive microbial cultures. The most common type of infection diagnosed in the study was "uncharacterized," accounting for more than 80% of all cases in the epidural group. Newborns received this diagnosis when they had elevated inflammatory markers and clinical presentation consistent with infection but without positive cultures or radiographic findings.

Jia and colleagues wrote that this category could reflect "a stage of infectious or inflammatory response rather than a definitive infection, attributable to either bacteriuric intraamniotic infection or sterile intraamniotic inflammation."

In any case, the team acknowledged, "the causes of neonatal infection remain elusive," but speculated that maternal fever appeared to be a mediator.

The investigators acknowledged a number of limitations to the analysis, including lack of placental culture studies to determine when histologic chorioamnionitis reflected infection versus sterile inflammation. Also, the epidural group more commonly received oxytocin and instrument-aided delivery than the controls, and Jia and colleagues did not examine whether this might have contributed to the higher infection rates (but promised such an analysis in a future study). The restriction to first-time deliveries was also a limitation; the researchers said they intend another study including multiparous mothers.

Comment