Abstract
A recent examination of LactMed® usage data revealed the following surprising facts: between 2020 and 2024, cannabis was the #1 substance searched for by LactMed® users each year, with the exception of 2021 when COVID-19 vaccines slightly edged it out. Cannabis accounted for 17–23% of all LactMed® searches during these years, amounting to over 240,000 searches in both 2023 and 2024. With these facts in mind, this column summarizes what is currently known about cannabis use during breastfeeding. Unless otherwise referenced, the following information comes from LactMed®, where more information and the original references can be found.
Frequency of Usage
A national survey in the United States found that from 2005 to 2018, cannabis use had doubled among new mothers who were breastfeeding. A 2017 survey of perinatal women found that 5.5% of participants reported postpartum marijuana use; among these women, 47% were breastfeeding. 1 In another survey, the overall percentage of nursing mothers who reported using cannabis in 2018 was 5.7%. 2 Women who resided in states with legalized recreational cannabis were more likely to use cannabis in the postpartum period than women in other states. 3 An anonymous online survey conducted from 2018 to 2019 among 1,516 breastfeeding women who used cannabis found that 67% of participants were “not at all” concerned that cannabis use while breastfeeding affected their baby. 4
Reported Harms
Long-term adverse effects
Concern has been expressed about the possible effects of cannabis on neurotransmitters, nervous system development, and endocannabinoid-related functions. Unfortunately, there are no adequate data on long-term infant outcomes and no resolution of these concerns.
A 1-year study found that daily or near-daily use might retard the breastfed infant’s motor development but not growth or intellectual development. This and another study found that occasional maternal cannabis use during breastfeeding did not have any discernable effects on breastfed infants. These old and often-cited studies were underpowered to rule out long-term harm and they were conducted at a time when cannabis products were much less potent than they are today.
More recently, 50 women who reported using cannabis in the prior 14 days donated milk samples for analysis of delta-9-tetrahydrocannabinol (THC), the major psychoactive component, and its major metabolites. THC was detectable in 66% of the samples and below the limit of quantification in 32% of samples. In four infant evaluations between 4–36 months of age, no differences were found in infant adverse reactions, postnatal growth, or neurodevelopmental outcomes between the groups with quantifiable and nonquantifiable THC in breast milk.
A large registry study in New Zealand suggested that postnatal use of cannabis by mothers may increase the risk of autism spectrum disorder, with male infants affected to a greater extent than female infants. However, no information on the breastfeeding status of the mothers was presented. Another study found that paternal use of cannabis increased the risk of sudden infant death syndrome, presumably from smoke inhalation by the infant.
In a recently published study, 199 pregnant women and their infants were followed up prospectively from 24 weeks of pregnancy to 3 years postpartum. THC was measured in maternal urine at about 27 weeks of gestation. At 3 years of age, children with prenatal exposure to THC had experienced a more rapid increase in body mass index than children with no exposure. However, infants who were breastfed for 5 months or longer had weight gains similar to nonexposed infants.
Short-term adverse effects
Some case reports show that heavy cannabis use can result in serious adverse events in breastfed infants. A 6-month-old infant was exclusively breastfed by a mother who was a chronic cannabis user. The infant presented to the emergency department with somnolence after falling off a couch and developing seizure-like activity and minimally responsive dilated pupils. Laboratory values and a head computed tomography scan were normal except for carboxy-THC found in urine and blood. The infant returned to baseline in 72 hours.
A 9-month-old girl was hospitalized for a first tonic-clonic seizure. THC and two of its metabolites were found in the infant’s blood. Diazepam and its metabolites were also found in the infant’s blood and the urine screening found cotinine, showing tobacco exposure. Her mother explained she had smoked cannabis resin since her child was 4 months old, sometimes just before breastfeeding, and on average, five times a day. After the initial episode, her child had three episodes of “spasms.” Breastfeeding was replaced by formula and 1.5 months later, the child had not had a recurrence.
The infant of a woman who used cannabis edibles for anxiety during pregnancy and breastfeeding had several episodes of apnea. At 1 week of age, while undergoing treatment for a urinary tract infection, the infant required intubation because of apneic episodes. At 5 weeks of age, the mother noticed the infant was having irregular breathing and apneic episodes and took the infant to the hospital. There the infant had more apneic episodes and a urine positive for THC, although the mother reported not using cannabis for the prior 3 days.
In an online survey of 1,516 mothers who used cannabis during breastfeeding, only 3% attributed symptoms in their infants to cannabis use; symptoms were perceived as positive or negative. However, 45% altered timing of cannabis use relative to breastfeeding to avoid exposing their infant to cannabis. 4
Cannabis Exposure
Some evidence of risk can be gleaned from measurements of cannabinoids in the milk of exposed mothers and in their breastfed infants’ serum. THC, other cannabinoids, and their metabolites are very fat soluble and persistent in the body fat of users. They are slowly released over days to weeks, depending on the extent of use. This slow release has probably been overrated as a risk factor for breastfed infants. The high fat solubility confers high protein binding (∼97%) to cannabinoids, which in turn greatly reduces their excretion into milk.
Excretion into milk
In perhaps the most “conventional” pharmacokinetic study, 8 exclusively nursing women who were 3–5 months postpartum and cannabis users smoked 100 mg of standardized cannabis. The mean THC concentration in breast milk was 53.5 µg/L (range 8.4–186 µg/L), and the mean peak THC concentration was 94 µg/L (range 12.2–420 µg/L) 1 hour after inhalation. The estimated daily THC intake for infants was 8 µg/kg, which corresponded to a relative infant dose of 2.5% (range 0.4–8.7%). The extremely wide range of milk THC values is typical of studies on the excretion of cannabinoids into milk.
Five other studies reported cannabinoid concentrations in the milk of 179 mothers who reported using cannabis products. The potency, mode and extent of use, and time between cannabis use and sample collection varied among the mothers. Mean and median milk THC concentrations varied widely in studies from 9.47 to 54.5 µg/L, with maximum levels as high as 1,620 µg/L. Three studies estimated median or mean daily infant intakes of 1.4–5.7 µg/kg, somewhat lower than in the pharmacokinetic study. Three studies estimated the half-life values of 11.5 hours, 27 hours, and 17 days. A computer model of THC in breast milk reported a half-life in milk of 39 hours and similarly, manufacturers of dronabinol, a synthetic oral form of THC, report a terminal half-life of 25–36 hours in plasma.
Infant blood levels
A pharmacokinetic model of THC following cannabis smoking predicted that in the worst-case scenario of maternal cannabis use six times daily during lactation, the maximum infant plasma concentration would range from 0.084 to 0.167 µg/L for infants between the ages of 1 and 12 months. These concentrations were two orders of magnitude lower than predicted maternal plasma levels. Neonates up to 1 month of age had the highest THC exposure, probably because of the incomplete maturation of metabolizing enzymes.
In contrast to the computer model, a mother reported smoking about four cannabis resin joints daily, sometimes just before the beginning of breastfeeding when her breastfed infant was 4 months old. After suffering a tonic-clonic seizure at 9 months of age, the infant’s plasma had 2.2 mcg/L of THC as well as 0.4 µg/L of the active 11-OH-THC metabolite. These amounts probably reflect the high maternal intake, the high variability breast milk excretion, and possibly infant inhalational exposure.
Effect on Lactation
The effects of long-term cannabis use on lactation are unclear, with some studies finding no effect on serum prolactin, but hyperprolactinemia and galactorrhea have been reported in chronic cannabis users. The relevance of these findings with respect to nursing mothers with typical usage patterns is uncertain. Mothers who smoke cannabis often also smoke tobacco, which reduces breastfeeding extent and duration.
In one small study, 14 women who used cannabis postpartum by inhalation reported lower milk production in the first, second, fourth, and sixth weeks postpartum than nonusers. A survey in Colorado where recreational cannabis is legal found that both prenatal and postnatal cannabis use were associated with a shorter duration of breastfeeding. Among women who reported using cannabis during pregnancy, 64% breastfed for 9 or more weeks compared with 78% of women who did not use cannabis during pregnancy. Among women with postpartum cannabis use, 58% breastfed for 9 or more weeks compared with 79% of women who did not use cannabis postpartum.
A database study of 4,969 postpartum women found that those who reported using cannabis were more likely to smoke tobacco, experience postpartum depressive symptoms, and breastfeed for less than 8 weeks. Most of the women who smoked cannabis postpartum also used it during pregnancy.
In an online survey of 1,516 mothers who used cannabis during breastfeeding, 86% of mothers reported no changes in their milk supply when using cannabis compared with when they were not using it. 4 However, another study found that mothers with the perception that their use of cannabis is harmful to their infants are likely to discontinue breastfeeding earlier than mothers who do not believe it is harmful. 1
Effects on Milk Composition
Lactose levels were higher and secretory IgA (sIgA) levels were lower in the milk of 14 cannabis users than in the milk of nonusers. However, when adjusted for body mass index, there was no difference in sIgA levels between the groups. Subjects who smoked both cannabis and tobacco cigarettes had lower carbohydrate levels and greater protein levels in their milk.
Milk samples donated to a breast milk repository by 165 women who had consumed cannabis within the past 14 days and had cannabis metabolites in their milk were compared with samples of milk from 472 women who had no cannabis metabolites in their milk. Cannabis-exposed women had higher levels of protein (mean 1.43 versus 1.19 mg/dL) and carbohydrate (median 1.43 versus 1.19 mg/dL) in their milk. Fat and energy content were not significantly different between the groups.
Professional Guidelines and Response
Professional guidelines recommend that cannabis should be avoided by nursing mothers, and nursing mothers should be informed of possible adverse effects on infant development from exposure to cannabis compounds in breast milk. However, surveys of mothers found that only 25–33% of participants reported receiving any health care provider counseling about marijuana either pre- or postnatally.1,5
Summary
Although the amount of information on maternal cannabis use during breastfeeding has increased considerably in recent years, definitive long-term infant outcome data are still lacking. What current studies do show is that infant cannabinoid exposure via milk is usually low, but variable, with occasional extensive exposure. A few case reports indicate that mothers who are heavy cannabis users might excrete enough THC into milk to adversely affect their infants. Confounding factors are the use of cannabis during pregnancy and possible cannabis smoke inhalation by the infant.
Although cannabis might reduce the breastfeeding duration, the effect is not great and concurrent tobacco smoking is a confounding factor. Minor changes in milk composition have also been reported.
Reasonable guidance for nursing mothers is to avoid cannabis use during breastfeeding, especially heavy use. However, if a mother insists on smoking cannabis, she should moderate her use and avoid breastfeeding for at least 2 hours after smoking. If an edible cannabis product is used, breastfeeding avoidance should be longer, especially if taken with food, which slows absorption. THC half-life in milk is in the range of 12–39 hours, which is consistent with complete THC clearance from milk in 2.5–8 days. Avoiding cannabis smoking in the vicinity of the infant by anyone is important because of increased infant exposure and a possible risk of sudden infant death syndrome.
Footnotes
Disclosure Statement
No competing financial interests exist.
Funding Information
No funding was received for this article.