Substance use need not be an absolute barrier to breastfeeding, but drug- and patient-specific guidance with counseling free of bias are the keys.
According to the 2016 National Survey on Drug Use and Health, one in 10 people (28 million) over age 12 used an illicit drug in the past 30 days. For young adults aged 18 to 25, use ranges as high one in four for illicit drug use, one in 10 for heavy alcohol use, and two in five for binge alcohol use. Use is primarily driven by marijuana and misuse of opioid pain relievers. Of pregnant women, 6.3% used illicit drugs, 8.3% reported alcohol use, and 4.3% reported binge drinking in the last month.1
Breastfeeding is a major public health strategy because of the well-known benefits, including child spacing, and reduced rates of sudden infant death syndrome (SIDS), childhood infections, and postpartum depression. These benefits may be particularly important for families struggling with substance use. Both the American Academy of Pediatrics (AAP) and the World Health Organization (WHO) recommend exclusive feeding with breastmilk during the first 6 months of life.2,3 The Healthy People 2020 targets are 81.9% for ever breastfeeding and 60.5% for any breastfeeding in 6 months.4 In 2010, The Patient Protection and Affordable Care Act and Fair Labor Standards Act mandated that working mothers be given reasonable break time and a private place to pump that is not a bathroom for up to 1 year after childbirth.5 A woman with substance use (SU) or substance use disorder (SUD) has the same rights and desires as any other mother to receive unbiased counseling and make an informed decision about breastfeeding and is not immune to the medical and societal pressure that “breast is best,” as noted in official statements from organizations regarding breastfeeding and substance use (Table 1).
Screening for drug use
The American Academy of Pediatrics, American College of Obstetricians and Gynecologists (ACOG), and the American Society for Addiction Medicine (ASAM) recommend that all pregnant women be screened for drug use by using a validated screening questionnaire and intervention techniques to counsel abstinence and refer for treatment those who meet criteria for a SUD.6 Further, ACOG states that routine laboratory testing of biologic samples is not required.8 The validated screening questionnaires are linked with education and intervention strategies, and are superior to urine drug screening (UDS) to detect use.9 If a UDS is used, it requires informed consent and should be ordered as a preliminary test with a reflex confirmatory test. Using UDSs to triage breastfeeding has limitations and potential to stigmatize and drive women away from medical care. Substances stay in maternal urine and breastmilk for different lengths of time, assays vary, medications can cross-react, use of reflex confirmatory tests may be inconsistent, false-positives and -negatives can occur, and there may be an arbitrary potentially biased focus on certain drugs and which mothers to test. A negative UDS test does not preclude use, nor does a positive test guarantee that the breastmilk contains harmful levels of a drug.
Hospital protocols vary widely in how hospitalized newborns are triaged to receive their mother’s breastmilk. Some rely on biological testing of urine or milk at delivery. Others do not use biological testing, do not initially withhold breastmilk, and provide education and supportive intervention first before deciding about ongoing breastfeeding recommendations. The latter better supports the ethical framework put forth by ACOG to discourage breastfeeding exclusion and separation of parents from their children solely based on suspected or confirmed SUD.8
Consequences of biologic screening for drug use
Providers have an obligation to be aware that there are potentially serious legal and social consequences of detecting illicit substances in pregnant and lactating women. After the 2010 Child Abuse and Prevention and Treatment Act passed, all states developed policies for health care providers mandating reporting of newborns and other children (not fetuses in federal law) who are exposed to illicit substances under the definition of child abuse or neglect. In 23 states plus the District of Columbia, laws designate substance use during pregnancy to be child abuse. Some states are criminally prosecuting pregnant women for SU during pregnancy. Federal regulation 42 CFR Part 2 protects the privacy of addiction treatment records, however, state laws supersede this protection. Pregnant women should provide written informed consent for UDS and receive no punitive action for refusal. Criminalizing pregnant women for substance use is unethical, counter-intuitive to the physician-patient therapeutic alliance, and does not provide a compassionate solution for the care of motherless children. The American Academy of Addiction Medicine and Amnesty International oppose policies that criminalize substance use by pregnant women.10,11
Breastfeeding rates in women with SUD
Rates of breastfeeding are lower in women with SUD. A 2011 study noted that 14% of mothers who used illicit substances or were on opioid maintenance therapy breastfed vs. 50% in the general population.12 In another 2019 study, the prevalence of breastfeeding was high for those receiving prenatal medication-assisted treatment. Prenatal intention to breastfeed was 87% for buprenorphine and 81% for methadone-treated mothers (P = 0.25). Exclusive breastfeeding at hospital discharge was 31% for buprenorphine and 19.6% for methadone-treated mothers (P = 0.06).13
The perception of breastfeeding “contraindications” is important to mothers with SUD. In a 2003 cohort of 393 low-income inner-city women, 48% never initiated breastfeeding and 16% had a documented contraindication to breastfeeding. Of those who never initiated breastfeeding, 42% with a contraindication to breastfeeding cited “not wanting to pass dangerous things” as the reason for not breastfeeding. Of those with contraindications, 75% used cocaine, 28% had HIV infection, 5% used PCP, and 3% heroin or methadone.14
How ob/gyns can help
Lactation is a bodily function that cannot ethically be regulated or banned by a medical or government authority. Unless a child is removed from the mother, she will have the ability to provide breastmilk to her child regardless of her lifestyle choices. It is more consistent with a therapeutic alliance, autonomy, and beneficence to counsel women that use of certain drugs is not recommended while breastfeeding (and why) instead of the inverse, “You cannot breastfeed if you have used drugs.” Recommendations should be based on an evaluation of the desire for SUD treatment and be free from intrinsic or organizational biases. Alternate nutrition should be considered with maternal permission when SUD treatment and/or mental health stability are in significant jeopardy.
Women with SUD may have fewer role models, lower self-esteem, and may make the assumption that successful breastfeeding is not achievable. A history of abuse may make it difficult. Breastfeeding may trigger flashbacks or shame, making trauma-informed counseling important. Information provided in a self-esteem-building manner about the potential harms of particular substances in the breastmilk may be enough to motivate a woman to stop using substances or practice responsible harm reduction while breastfeeding. Women with chronic drug use resulting in brain dysregulation may find breastfeeding overwhelming or impossible. Maternal behavior may become disrupted where stress becomes heightened by neonate behavior instead of what would normally be rewarding to mothers who do not have a SUD.15 As a result breastfeeding may be more harmful than helpful. The ideal situation for successful breastfeeding is for a mother to be abstinent from substance use, part of a comprehensive substance use treatment program (ideally gender specific), and if indicated, stable on medication-assisted treatment. If abstinence is not possible, but harm reduction strategies are reliably implemented, the benefits of breastfeeding can outweigh the risks. Harm reduction strategies include compliance with provider visits and, as needed, pumping and discarding milk, feeding with donor milk or formula, seeking an alternative childcare provider, and avoidance of co-sleeping with the baby when using drugs.
For breastfeeding success, the provider approach and environment should be optimized. It is critical to treat SUD as a chronic relapsing disease, work to avoid mixed messages, and avoid pejorative language like “junkie,” “drug seeker,” and “addicted baby.” Assess the mother’s comfort level and exposure of breast and body. Ask permission before examination, respect boundaries, provide a breast pump and lactation consultation familiar with mother-infant SUD, and avoid encouraging discontinuation at the first sign of difficulty. Regardless of breastfeeding success, the mother’s progress in recovery is most important for the infant’s health and development.16 Table 2 summarizes how ob/gyns can help facilitate breastfeeding women with SU and SUD. Table 3 summarizes recommendations for specific substances and breastfeeding.17,18
Impact of specific substances on breastfeeding
The concentration of methadone in human milk is low and women on methadone maintenance should be encouraged to breastfeed regardless of methadone dose.19,20 Neonatal abstinence syndrome (NAS) occurs in approximately 70% of neonates born to mothers prescribed methadone and may negatively influence latch and sucking. Because NAS is significantly reduced in neonates whose mothers breastfeed while on methadone maintenance,21 vigorous efforts to encourage and support mothers to breastfeed are needed. Once NAS is resolved, strategies to reduce long-term exposure to the neonate include feeding or pumping before daily dose and or waiting 2 to 4 hours after a dose.
Buprenorphine is a partial opioid agonist with few side effects and has become a successful therapy for medical-assisted treatment of opioid SUD. Levels of buprenorphine are low in human milk and are not likely to cause negative effects in the infant.22 Breastfed infants whose mothers used buprenorphine have less severe NAS and have excellent rates of breastfeeding (76% breastfed at all and 66% were breastfeeding at 6 weeks post-delivery), as is the case with methadone.23 Breastfeeding is encouraged for mothers using buprenorphine.
Other prescription opioids
Short courses of low-dose prescription opiates other than codeine can be used safely during breastfeeding. Codeine use during breastfeeding is discouraged by the US Food and Drug Administration because its metabolite is morphine and some infants may be ultra-rapid metabolizers (URM; carriers of more than two copies of the gene CYP2D6). URM create high blood levels of morphine, which can lead to respiratory arrest. URM occurs in about 2% of the population, but is more common in particular ethnic groups (28% in North Africans, 10% in Caucasians, 3% in African Americans).24 There is little information available on the safety of breastfeeding when moderate to high doses of opiates are used for long periods of time. Opioids with minimal secretion into breastmilk such as hydrocodone are preferred. Infants younger than age 1 month who are being exclusively breastfed by mothers using opiates in high doses or long term should be observed for any signs of excessive somnolence or sedation.16
Heroin (diacetylmorphine) is metabolized into morphine, which has an elimination half-life of 2 to 3 hours. At therapeutic doses, most opioids like morphine are excreted in breastmilk in minimal amounts. Heroin has an elimination half-life of 15 to 30 minutes, is rapidly excreted into breastmilk, and causes dependence in the infant, however, milk plasma ratios are not well known. Breastmilk should be pumped and discarded for 24 to 48 hours after use which may be impractical.25 Mothers using heroin who cannot enter treatment or manage harm reduction strategies should be counseled not to breastfeed.
High concentrations of cocaine are found in breastmilk screening in recreational users and levels are extremely varied and have been found in the urine of breastfeeding infants 24 to 36 hours after maternal use. The elimination half-life is 1 hour, and concentrations are eight times higher in breastmilk than in the plasma of the user. Infants are extremely sensitive to cocaine because of immature inactivation enzymes (plasma cholinesterase). Infants exposed to cocaine in breastmilk exhibit irritability, tremulousness, dilated pupils, hypertension, seizures, vomiting, high pitched crying, and respiratory distress. Infant deaths have been reported.25,26 Breastmilk should be pumped and discarded for 24 hours after use, which may be impractical. Mothers using cocaine who do not enter treatment or cannot manage harm reduction strategies should be counseled not to breastfeed.
Methamphetamine or MDMA
Amphetamine-type stimulants easily cross into breastmilk and are concentrated there in higher amounts (3-8 times) than in the plasma of the user. The elimination half-life is 6 to 12 hours. Infants breastfed by amphetamine users experience irritability, poor sleeping, agitation, and excessive crying. Infant deaths from cardiopulmonary failure have been reported. Milk should be pumped and discarded for 48 to 100 hours after use, which is impractical.25 Mothers using methamphetamine or MDMA who do not enter treatment or cannot manage harm reduction strategies should be counseled not to breastfeed.
The psychoactive compound in marijuana, delta-9-tetrahydrocannabinol (THC), accumulates in breastmilk in moderate to severe amounts depending on chronicity of use. The elimination half-life is 20 to 36 hours and up to 4 days in chronic users. Data are insufficient data on which to base conclusions about the long-term effect of marijuana exposure through breastmilk. As a result, use of marijuana is discouraged during breastfeeding.6,25
One-third of benzodiazepine (BZD) use is illicit. BZDs appear in low concentration in breastmilk at about half of maternal plasma levels. Short- and intermediate-acting BZDs like alprazolam and lorazepam, respectively, provide negligible infant exposure and have not been associated with problems. Long-acting BZDs like diazepam have the potential to cause lethargy, sedation, and weight loss in infants, which are reversible after breastfeeding is discontinued, although abrupt weaning or rapid cessation can precipitate withdrawal in the infant. Breastfeeding is not recommended with long-term or high-dose use of long-acting BZDs or when using multiple sedative drugs. For occasional use, milk should be pumped and discarded for 6 to 8 hours after use.25
Half of breastfeeding women in western countries reportedly consume alcohol at least occasionally. Alcohol interferes with the milk ejection reflex and may reduce production. Human milk levels parallel maternal blood alcohol levels. Long-term effects of alcohol exposure in breastmilk are unknown. Most experts advise limiting alcohol intake to only 8 oz of wine or two beers and to wait for 2 hours after drinking to resume breastfeeding.17
Nicotine is concentrated in breastmilk in concentrations up to three times maternal plasma although only 10% is excreted into breastmilk. The elimination half-life of nicotine is approximately 2 to 4 hours. There is no evidence that nicotine in breastmilk represents a health risk to the infant and breastfeeding benefits are thought to outweigh risks. Other chemicals in secondhand smoke are thought to be worse and increase incidence of respiratory allergy and sudden infant death syndrome (SIDS). Tobacco smokers are encouraged to breastfeed while trying to quit, start nicotine replacement, and eliminate infant smoke exposure.17,25
Neonates of mothers with SU and SUD who can be breastfed stand to gain significant benefits as do their mothers, who may use motherhood and breastfeeding as a chance to change the path of their lives. With appropriate nonjudgmental support, mothers with SU and SUD can achieve breastfeeding success. Caring for these patients also offers ob/gyns potential opportunities for growth and for overcoming intrinsic biases and other barriers.
The authors report no potential conflicts of interest with regard to this article.