OUD in pregnancy: buprenorphine better than methadone

 A large  study found that infants born to mothers with opioid use disorder on buprenorphine did better than if they were on methadone, though no difference in maternal outcomes (see opioid use disorder pregnancy bup better than methadone nejm2022 in dropbox, or DOI: 10.1056/NEJMoa2203318)

 

Details:

-- data-mining study of 2,548,372 pregnancies of women in Medicaid programs in the US who had opioid use disorder (OUD) and were on buprenorphine (BUP) or methadone (MET), from 2000-2018

-- mean age 29, 83% white/5% Black, 70% used opioids prior to pregnancy, tobacco use 45%

-- baseline anxiety 30%, depression 27%, bipolar 11%; antidepressants 5%,benzos 19%

-- adequacy of prenatal care: inadequate 44%, intermediate 17%, adequate 15%, adequate plus 24%

    -- women on BUP were more often white, from the Northeast or Midwest, and live in nonmetropolitan or rural areas. Also more likely to have depression or anxiety and have nonopioid substance use disorders. And more likely to be on antidepressants and other psych meds

    -- those on MET were more likely to be on prescription opioids

-- early pregnancy (gestation through 19 weeks): 10,704 pregnant persons exposed to BUP and 4387 to methadone (MET)

-- late pregnancy (week 20 through day before delivery):  11,272 were exposed to BUP and 5056 to MET

--confounders assessed in the adjusted models below: opioid-related ED visits/hospitalizations/treatments before pregnancy, nonopioid substances used, medical conditions associated with OUD (hepatitis, HIV, STDs), mental health conditions, chronic medical problems, meds used, health care utilization metrics, proxies for social issues (homelessness, domestic violence), demographics

-- a separate analysis of propensity scoring was done to equalize some of the differences in the women in each group

-- 85% on BUP and 89% on MET continued the same medication throughout the pregnancy

 

Results:

-- neonatal abstinence syndrome (NAS):

    -- BUP: 52.0%

    -- MET: 69.2%

        -- 27% lower risk with BUP, adjusted relative risk aRR 0.73 (0.71-0.75),

--preterm birth:

    -- BUP: 14.4%

    -- MET: 24.9%

        -- 42% lower risk with BUP, aRR 0.58 (0.53-0.62)

-- small-for-gestational-age (SGA):

    -- BUP: 12.1%

    -- MET: 15.3%

        -- 28% lower risk with BUP, aRR 0.72 (0.66-0.80)

-- low birth weight:

    -- BUP: 8.3%

    -- MET: 14.9%

        -- 44% lower risk with BUP, aRR 0.56 (0.50-0.63)

 

-- C-sections, in women exposed in early pregnancy:

    -- no significant difference: BUP 33.6% vs MET 33.1%

-- severe maternal complications in early pregnancy:

    -- no significant difference: BUP 3.3% vs MET 3.5%

-- not much difference if assess these 2 maternal outcomes after exposure in late pregnancy

-- no significant difference if use high-dimensional propensity scoring

-- restricting the analysis to pregnancies with the highest level of prenatal care “slightly attenuated the associations for all neonatal outcomes except neonatal abstinence syndrome”

 

This graph shows that there was a very consistent benefit for infants exposed to BUP vs MET (all with very tight confidence intervals) but no evident benefit for maternal outcomes: circles to the left of the 1.00 vertical line reflect benefit of buprenorphine:

 

 

Commentary:

-- OUD is inceasing in pregnant women, with 2017 data finding OUD in 8.2 women per 1000 deliveries (and 14.6 per 1000 women on Medicaid)

-- both buprenorphine and methadone have been shown to improve adherence to prenatal care, lower preterm birth, reduce the risk of returning to opioids after pregnancy, and have fewer opioid overdoses and deaths

-- this study was the largest one to date and assessed maternal as well as infant outcomes

-- the above study pretty clearly found that buprenorphine was superior to methadone in important ways: decreased neonatal abstinence syndrome (NAS), preterm birth, SGA, and low birth weight, even after adjusting for many potential confounders

-- other studies were smaller, which limited their ability to draw rigorous conclusions (eg, there were large confidence intervals in the MOTHER trial, a small RCT which also found that buprenorphine was associated with less need for morphine for the NAS and shorter hospital stays: see 

N Engl J Med 2010; 363:2320-2331, or DOI: 10.1056/NEJMoa1005359) 

-- there has been concern that there might be a selection bias in other studies, that those on MET were sicker or had more intensive opioid addiction that perhaps could explain the different rates of NAS. in this study there was very similar health care utilization, including ED visits and hospitalizations and the types and numbers or substances used, suggesting that the addiction levels were similar enough to make a selection bias very unlikely. Also, further sensitivity analysis utilizing a high-dimensional propensity-score algorithm to equalize the groups did not materially change the results 

-- another potential confounder is that those on BUP may have had higher quality care, since their receiving BUP tends to be more integrated into medical care than a methadone clinic. But this study did find similar results when limiting the analysis to those receiving high-quality care (per assessment of prenatal adequacy of care index) 

-- great news (finally...): the Feds have just changed the buprenorphine prescribing obstacles: through section 1262 of the Consolidated Appropriations Act, 2023 (the Omnibus bill), they have removed the need to get a special waiver to prescribe buprenorphine (the "X" waiver), and "all practitioners who have a current DEA registration that includes Schedule III authority may now prescribe buprenorphine for Opioid Use Disorder in their practice if permitted by applicable state law" and the Feds are "actively working on implementation of a separate provision of the Omnibus related to training requirements for DEA registration that becomes effective in June 2023": https://www.samhsa.gov/medication-assisted-treatment/removal-data-waiver-requirement .  ie, much better for patients and prescribers... And, i think, some training for providers both in OUD and in the specifics of BUP treatment are important, given the complexities of the prescription meds used

Limitations:

-- as a data-mining study, little granular data on important psycho-socio-behavioral issues: eg, housing, poverty, education, life stressors, feelings of powerlessness/lack of control over one’s life, quality of diet/exercise, smoking/and amount, alcohol/and amount, social supports, etc. These limitations (as well as likely other potential confounders) in a retrospective analysis, renders the conclusions to being associations between the meds and outcomes and do not imply causality (one needs an RCT to really determine causality)

    -- and using homelessness and domestic violence as a proxy for social issues is hardly adequate....

-- not so much diversity: 83% white, only 5% Black; and the study was limited to those on Medicaid and not those with private insurance or no insurance at all, limiting generalizability to these groups

-- there may be limitations in the accuracy of the information put into the assessed large Medicaid database. 

-- there are differences in Medicaid coverage by states: those on buprenorphine, for example, may have been in more comprehensive care, since the states covering  buprenorphine likely had broader coverage and access to care (the researchers did try to compensate for this by limiting the analysis to women who had the highest level of prenatal care per the Adequacy of Prenatal Care Utilization Index, but that is only one marker of the quality/comprehensiveness of health care overall) 

-- we also do not know the doses of BUP or MET prescribed or other interventions done (individual psych therapy, group therapy, etc)

-- we also do not know the outcomes of the neonates with NAS, how severe it was, differences in morbidity/mortality, etc

 

So,

-- an impressive large study of the huge societal problem of OUD, only getting worse now in the times of covid: see http://gmodestmedblogs.blogspot.com/2020/08/covid-severe-psych-substance-use.html 

-- and buprenorphine reasserts itself as a really great drug overall for OUD, but now more clearly so in pregnant women with OUD

-- decreasing neonatal abstinence syndrome is really important (and BUP seems to have won out over MET): "infants with NAS have a 3.5-fold increase in severe neonatal morbidity, and their mothers have substantial increases in mortality and severe morbidity during childbirth", per https://publications.aap.org/pediatrics/article/144/2/e20183664/38503/Neonatal-Abstinence-Syndrome-and-Associated?autologincheck=redirected

-- and, and, and it seems that prescribing and receiving BUP has just gotten lots easier, as per above.....

geoff