Lowering Maternal Opioid Doses Doesn't Prevent Severe NOWS, Study Finds

Neither the dose of methadone or buprenorphine a pregnant patient takes nor the amount of the drug measured in her newborn's umbilical cord predicted how severe the baby's withdrawal would be, according to a research letter published in JAMA Network Open. The finding, drawn from 94 newborns at multiple sites, bears directly on a question clinicians face when a patient's dose climbs during pregnancy: whether to lower it late in the third trimester in the hope that the newborn will withdraw less. The investigators say no.

Among the 41 infants who developed severe neonatal opioid withdrawal syndrome, cord concentrations of methadone, buprenorphine and their metabolites looked no different from those in infants who did not. Maternal doses were also statistically indistinguishable between the two groups.

How the researchers measured exposure

The cohort enrolled patients from July 2016 through December 2025. To qualify, a patient had to have taken methadone or buprenorphine without interruption for at least 30 days before delivery, give birth at 34 weeks' gestation or later, and have cord tissue collected at birth.

Cord tissue is an unusual choice and a deliberate one. It accumulates drug exposure across the third trimester the way a tree ring records a season, and it varies less from moment to moment than serum or urine. A certified laboratory measured each medication and its primary metabolite by mass spectrometry — EDDP for methadone, norbuprenorphine for buprenorphine.

Each site assessed newborns using its own protocol, either modified Finnegan scoring or the Eat, Sleep, Console method. An infant counted as having severe NOWS if the baby needed medication, had elevated Finnegan scores, or failed to meet Eat, Sleep, Console criteria.

The 92 pregnancies produced 94 neonates. Fifty-three had methadone exposure and 40 had buprenorphine exposure. Mothers had a median age of 31, most had public insurance, and smoking during pregnancy was common.

Methadone carried a threefold risk of severe withdrawal

Overall, 41 newborns — 43.6 percent — developed severe NOWS. The rate differed sharply by medication: 33 of 53 methadone-exposed infants (55.6 percent) versus 11 of 40 buprenorphine-exposed infants (27.5 percent), a difference that held after adjustment for gestational age, maternal smoking and study site (adjusted odds ratio 3.42; 95% CI, 1.17 to 9.96).

That result is consistent with the pattern the paper describes, and it is the part of the study most likely to be quoted. The more interesting finding is what happened next.

The number that did not move

Within the methadone group, higher maternal doses did correspond to higher cord concentrations of both methadone and EDDP, with correlation coefficients of 0.64 and 0.45. Buprenorphine behaved differently: maternal dose bore no significant relationship to cord levels of buprenorphine or norbuprenorphine at all.

But those cord concentrations told clinicians nothing about severity. Median maternal methadone dose was 127.5 mg among infants with severe NOWS and 120 mg among those without, a difference that was not statistically significant. For buprenorphine the medians were 16 mg in both groups. Peak Finnegan scores correlated with neither maternal dose nor cord drug levels.

The pattern held when the authors narrowed their definition of severe NOWS to the 22 infants who actually received pharmacologic treatment.

The authors wrote that the gap between the two medications "may reflect pharmacologic properties, placental transport, or fetal adaptation rather than absolute exposure alone." Buprenorphine is a partial agonist at the mu receptor; that difference sits inside the authors' explanation, not inside a milligram count.

Why the dose-cutting temptation persists

The logic behind dose reduction runs in one direction: if more drug in the mother means more drug in the baby, and more drug in the baby means worse withdrawal, then less drug should mean a calmer newborn. The first two steps hold up in this cohort. The third does not.

What the trade-off costs sits on the mother's side of it. The investigators recommended against lowering medication for opioid use disorder during pregnancy to protect newborns, and said dosing should keep being tailored to each patient's clinical stability, with the goal of reducing the risk of relapse and overdose.

The study has real limits, which the authors state. The cohort was too small to support causal analysis of why a patient received one medication rather than the other, a decision that often hinges on how stable she was before pregnancy. Racial and ethnic diversity was limited. NOWS evaluation differed by site, though results were similar across definitions.

What it does not have is the thing that would justify dose cutting: a demonstrated link between how much medication reached the fetus and how sick the newborn became. For clinicians weighing a dose change in the third trimester, buprenorphine and methadone treatment during pregnancy now comes with one more reason to leave the dose where the patient's stability put it.