Ethics approval

Ethical approval was not necessary for the research and writing of this review.

Cannabinoid science

There are over 100 cannabinoids and other components in the cannabis plant, with tetrahydrocannabinol (THC) and cannabidiol (CBD) being the most studied.^{Reference Mouhamed, Vishnyakov, Qorri, Sambi, Frank and Nowierski35} THC gives the recreational ‘high’ and CBD to some extent counteracts the psychoactive effect of THC.^{Reference Mouhamed, Vishnyakov, Qorri, Sambi, Frank and Nowierski35} The phytocannabinoids bind to the endocannabinoid receptors CB₁ and CB₂, while also interacting with other neural transmission systems.^{Reference Pertwee36–Reference Aizpurua-Olaizola, Elezgarai, Rico-Barrio, Zarandona, Etxebarria and Usobiaga38} CB₁ and CB₂ are G-protein-coupled receptors. CB₁ receptors have psychoactive properties and are expressed in the central nervous system, gastrointestinal system, adipocytes, liver tissue and skeletal muscle. CB₂ receptors, which are more restricted, are expressed in immune cells located in the tonsils, thymus, spleen and bone marrow, as well as in the enteric nervous system within the gastrointestinal tract.^{Reference Wang, Dow-Edwards, Keller and Hurd39} CB₁ binding exerts its physiological and pathological effects by regulating presynaptic Ca⁺⁺ levels, thus leading to a reduced release of neurotransmitters.^{Reference Sharapova, Phillips, Sirocco, Kaminski, Leeb and Rolle40} At an early stage of human embryonic and fetal development (14 weeks), the endocannabinoid system develops in relation to other neurotransmission systems, and prenatal cannabis exposure could therefore lead to changes in activities of brain areas such as the prefrontal cortex, the mesolimbic system, the striatum and the hypothalamic–pituitary axis, with potential for long-lasting or delayed effects on executive function or regulation of the emotional system and cognition.^{15,Reference Bahji, Stephenson, Tyo, Hawken and Seitz25,Reference Sharapova, Phillips, Sirocco, Kaminski, Leeb and Rolle40}

Medical use of cannabis

The earliest recorded date for cannabis use as a medicine is 4000 bce in China, and in the 19th and 20th centuries it has been used to treat migraine, neuropathic and musculoskeletal pain, and in childbirth.^{Reference Barnes41} Following the discovery of the CB₁ and CB₂ cannabinoid receptors, research showing increasing medicinal properties has heightened interest in the legalisation of cannabis, specifically for medical use and to facilitate further research.^{Reference Barnes41} Some formulations are currently used in treating resistant spasticity in multiple sclerosis, and nausea and vomiting induced by chemotherapy.⁴² CBD oral solution (Epidyolex^®) has been classified as a Schedule 2 controlled drug in the UK for the treatment of seizures associated with Lennox–Gastaut and Dravet syndromes.⁴² Furthermore, in a recent randomised placebo-controlled phase 2 trial to establish the safety of dosing of CBD in non-pregnant individuals with cannabis use disorder, 400 and 800 mg appeared safe.^{Reference Freeman, Hindocha, Baio, Shaban, Thomas and Astbury43} In the UK, medicinal cannabis can be prescribed only by doctors on the General Medical Council's (GMC's) specialist register with special interest in the condition being treated, and under strict prescribing guidelines stipulated by the National Institute for Health and Care Excellence (NICE), GMC and NHS England.^{Reference Freeman, Hindocha, Green and Bloomfield44}

Guidance on use of cannabis

Recent guidance from the UK's Food Standards Agency (FSA) on the increasingly popular oral non-medicinal CBD-containing products such as beverages (beer, spirits, wine, coffee and soda style drinks), oils (tinctures, drops, syrup, olive oils) and confectionary (gum drops, chocolate) limited their use to no more than 70 mg daily. It recommended that non-medicinal CBD products should not be used by pregnant or breastfeeding individuals, extrapolating from data on animal research which showed fetal harm.⁴⁵ Likewise, in the USA, the US Food & Drug Administration (FDA) also highlighted the potential harm from cannabis (also referred to as marijuana) and although it approved the controlled use of medical cannabis (Epidyolex^®), given the risks, which include serious liver injury, it strongly advises against the use of CBD, THC and marijuana in any form during pregnancy or while breastfeeding.^46,47

Although guidance from official sources is relatively clear, there is a risk that information from other sources may provide conflicting advice to consumers. For example, the FDA reported that it has issued warning notices against three companies making unsubstantiated claims about their cannabis-based products’ ‘ability to limit, treat or cure cancer, neurodegenerative conditions, autoimmune diseases, opioid use disorder, and other serious diseases, without sufficient evidence and the legally required FDA approval’.⁴⁶ Moreover, a study conducted in Colorado demonstrated that cannabis dispensaries often gave conflicting and sometimes incorrect information regarding the use of cannabis-based products during pregnancy. A mystery caller from the study team contacted 400 dispensaries stating that she was 8 weeks pregnant and was looking for advice regarding cannabis-based products for morning sickness. It was found that 35.7% of dispensaries endorsed the use of cannabis-based products during early pregnancy, 65% based their recommendation on personal opinion and only 31% recommended speaking to a healthcare provider without prompting.^{Reference Dickson, Mansfield, Guiahi, Allshouse, Borgelt and Sheeder48} This demonstrates the importance of presenting a united front with official regulations and standards of advice, and of raising public health awareness regarding medical use of cannabis during pregnancy that is informed by scientific evidence.

Cannabis and mental illness

Many people with psychological or mental disorders self-medicate with cannabis;^{Reference Osborn, Lauritsen, Cross, Davis, Rosenberg and Bonadio37} however, the evidence on treatment efficacy is limited and often mixed.^{Reference Whiting, Wolff, Deshpande, Di Nisio, Duffy and Hernandez49} For example, a review of research into the use of CBD in treating fibromyalgia reported that, although findings seemingly suggested that CBD is a safe and effective treatment, the studies all suffered from methodological limitations that prevent firm conclusions being drawn.^{Reference Cameron and Hemingway50} A recent systematic review and meta-analysis concluded that there is insufficient evidence that cannabinoids are effective in treating mental disorders, including depressive disorders and symptoms, anxiety disorders, attention-deficit hyperactivity disorder (ADHD), Tourette syndrome, post-traumatic stress disorder (PTSD) and psychosis.^{Reference Black, Stockings, Campbell, Tran, Zagic and Hall51} Another systematic review demonstrated tentative support for the efficacy of cannabis in social anxiety and as an adjunct in schizophrenia; however, the evidence in treating sleep disorders and PTSD was relatively weaker and there was no benefit of using cannabis to treat depression or mania.^{Reference Sarris, Sinclair, Karamacoska, Davidson and Firth52} The authors concluded that the recommendation of cannabinoid-based interventions for psychiatric disorders would be premature at the moment and further warned against using products with high delta-9 THC content, particularly by young people.^{Reference Sarris, Sinclair, Karamacoska, Davidson and Firth52} This is important to note as other studies have suggested that recreational cannabis with high THC content (e.g. skunk) is associated with mental health problems and physiological manifestations, including psychosis, dizziness, euphoria, drowsiness, dry mouth, confusion, somnolence and fatigue.^{Reference Murray and Hall23,Reference Barnes41,Reference Colizzi and Murray53} It is clear that more research is required with human participants (rather than animals) to examine the safety of different strains and components, dosage and routes of administration.^{Reference Barnes41}

Advocates of medicinal cannabis suggest that it could reduce the rates of opioid dependence and deaths from overdose if patients switch from opioid-based pain relief to cannabis, as the risk of dependence is much lower and there are no reports of death from cannabis overdose.^{Reference Barnes41} However, an international survey of 55 240 cannabis users published in 2020 found that the most commonly used classes of cannabis – sinsemilla and herbal (30.3%) and sinsemilla, herbal and hashish (20.4%) – were associated with increased dependence severity, while the class of concentrates and sinsemilla (1.7%) was associated with a record of diagnosed mental disorder.^{Reference Craft, Winstock, Ferris, Mackie, Lynskey and Freeman54} The conflicting evidence also highlights the need for further research.

Impact of cannabis use in pregnancy

Factors associated with the continued use of cannabis during pregnancy include history of cannabis use disorder (2.77 times more likely to continue), frequent cannabis use (daily or weekly compared with monthly), not completing high school, having a psychiatric disorder, biological father's cannabis use and being unmarried.^{Reference Martin, Longinaker, Mark, Chisolm and Terplan6,Reference Marroun H, Tiemeier, Steegers, Jaddoe, Hofman and Verhulst55} Personal (e.g. marital status, ethnicity) and area characteristics (e.g. urban, cosmopolitan or rural environments) as well as lifestyle factors can contribute to drug use; however, researchers warn that these factors often interact so it would be difficult to isolate the influence of any one of these characteristics on cannabis use.¹

Studies have reported short-term risks for individuals who use CBD in pregnancy. This includes impaired short-term memory and motor coordination, altered judgement, paranoia, dependence disorder, psychosis, injuries, motor vehicle collisions and suicide.^{Reference Colizzi and Murray53,Reference Volkow, Baler, Compton and Weiss56} Indeed, maternal suicide remains the leading cause of direct deaths occurring within a year after the end of pregnancy;^{Reference Knight, Bunch, Tuffnell, Shakespeare, Kotnis and Kenyon57} 8% of the women who died during or up to a year after pregnancy in the UK between 2016 and 2018 were at severe and multiple disadvantages, also known as the ‘toxic trio’ of mental health diagnosis, substance misuse and domestic abuse.^{Reference Knight, Bunch, Tuffnell, Shakespeare, Kotnis and Kenyon57}

Cannabis use in pregnancy has also been associated with still-birth, preterm labour, low birth weight, ‘small for gestational age’ and two-fold increased risk of admission to a neonatal intensive care unit.^{Reference Navarrete, García-Gutiérrez, Gasparyan, Austrich-Olivares, Femenía and Manzanares58–Reference Warshak, Regan, Moore, Magner, Kritzer and Van Hook60} Gestational cannabis use may also be linked to maternal fatty liver, obesity, insulin resistance and increased risk of gestational diabetes mellitus (GDM).^{Reference Ayonrinde, Ayonrinde, Van Rooyen, Tait, Dunn and Mehta61} In a more recent study, prenatal cannabis use was associated with a 50% increased likelihood of low birth weight, independent of confounders, although no association was found for small for gestational age, preterm birth and neonatal intensive care unit admission.^{Reference Crume, Juhl, Brooks-Russell, Hall, Wymore and Borgelt62} Furthermore, another recent population-based retrospective cohort study that included 661 617 pregnancies reported that cannabis users (n = 9427) were twice as likely to have a preterm birth compared with non-users, after adjusting for confounding variables, including tobacco use.^{Reference Corsi, Walsh, Weiss, Hsu, El-Chaar and Hawken63}

Cannabis use can also affect breastfeeding as evidence suggests that THC and CBD can accumulate in breast milk.^{Reference Moss, Bushlin, Kazmierczak, Koop, Hendrickson and Zuckerman64} However, this remains a vastly understudied area. As THC is highly fat soluble it can be excreted into breast milk^{Reference Bertrand, Hanan, Honerkamp-Smith, Best and Chambers65,Reference Baker, Datta, Rewers-Felkins, Thompson, Kallem and Hale66} and released slowly over days to weeks, depending on extent of use. This is a concern as evidence suggests that breastfeeding mothers sometimes increase their consumption of cannabis after childbirth.^{Reference Moss, Bushlin, Kazmierczak, Koop, Hendrickson and Zuckerman64} Furthermore, cannabis use while breastfeeding appears to be associated with a decrease in infant motor development at 1 year of age.^{Reference Astley and Little67} However, it is difficult to draw clear conclusions owing to small sample sizes and difficulty in separating the effect of exposure to cannabis in utero which continues into the postnatal period through lactation from exposure that occurs only during lactation. As studies (the majority of which are animal rather than human^{Reference Marchetti, Di Masi, Cittadini, La Monaca and De Giovanni68}) indicate conflicting outcomes, women who are unable to abstain are advised not to breastfeed within 1 h of inhaling or consuming cannabis, with the aim of reducing the infant's exposure to the highest concentration of cannabis in breast milk.^{Reference Ordean and Kim69}

The long-term effect of cannabis use should also be borne in mind as intrauterine exposure to cannabis has been found to increase the likelihood of initiation of cannabis and other substance use in adolescents.^{Reference Frank, Rose-Jacobs, Crooks, Cabral, Gerteis and Hacker70} Prenatal exposure may also be linked to developing psychosis-like and affective disorders^{Reference Paul, Hatoum, Fine, Johnson, Hansen and Karcher71,Reference Marangoni, Hernandez and Faedda72} and ADHD.^{Reference Roncero, Valriberas-Herrero, Mezzatesta-Gava, Villegas, Aguilar and Grau-López24} It is also linked with reduced attention and executive functioning skills, poorer academic achievement and behavioural problems.^{Reference Warner, Roussos-Ross and Behnke73} This might be because maternal prenatal cannabis use adversely affects growth of fetal and adolescent brains,^{Reference Jaques, Kingsbury, Henshcke, Chomchai, Clews and Falconer29,Reference Marroun H, Tiemeier, Steegers, Jaddoe, Hofman and Verhulst55} as demonstrated in the reduction of inhibitory interneurons in the hippocampal formation of adult rats.^{Reference Reid H, Lysenko-Martin, Snowden, Christie and Thomas74} In more serious cases, children raised in families where there is substance misuse can suffer negative effects on long-term health and well-being.⁷⁵ Parents may be unable to supervise their children and meet their needs appropriately, which could lead to emotional abuse and neglect.^{Reference Cleaver, Unell and Aldgate76} There are further concerns about physical abuse of children if parents have difficulty controlling their own emotions. These children can suffer from behavioural, emotional and cognitive problems and experience long-term psychological effects from their experience of being raised in a chaotic household.⁷⁵

Although research suggests that there is a high rate of abstinence from cannabis during pregnancy (with a discontinuation rate of around 77–78%),^{Reference Forray, Merry, Lin, Prah Ruger and Yonkers77,Reference Massey, Lieberman, Reiss, Leve, Shaw and Neiderhiser78} a high number of individuals also relapse, with 41% using cannabis within 3 months after delivery.^{Reference Forray, Merry, Lin, Prah Ruger and Yonkers77} Several characteristics could influence the rate of abstinence and relapse of cannabis use during the perinatal period. One study found that women with a post-secondary education and lower scores of depression and anxiety were more likely to attain abstinence, whereas those aged 21 years and below and with a diagnosis of depression were more likely to relapse in the postpartum period.^{Reference Forray, Merry, Lin, Prah Ruger and Yonkers77} Furthermore, given the high degree of concurrent substance misuse, women may substitute smoking cigarettes for alcohol, marijuana or cocaine use as they perceive cigarettes to be more socially acceptable and less harmful compared with illicit substances and alcohol.^{Reference Forray, Merry, Lin, Prah Ruger and Yonkers77} Thus, taking personal characteristics and the perception of harm into consideration is important for targeted perinatal interventions aimed at managing risks associated with cannabis use.

Ethical considerations of drug use in pregnancy

The use of medicinal cannabis for different conditions has gained in popularity despite evident risks.^{Reference Mead79} Consequently, clinicians are often torn in the moral conflict and the debate around legalisation of medicinal and recreational cannabis use.^{Reference Harris11,Reference Murray and Hall23} Although legalisation and decriminalisation of cannabis across the world indicate a changing perception on the ethics of cannabis use, the stigma from the legacy of prohibition and ongoing restrictions are believed to stall research.^{Reference Glickman and Sisti80,81} However, the principles of beneficence and non-maleficence, respect for autonomy and justice, weighing costs and benefits, and patient-centred decision-making can all support evidence-based practice and can guide clinicians to recommend or prescribe cannabis when indicated.^{Reference Glickman and Sisti80,Reference Lambert, Scheiner and Campbell82}

Claims that a fetus has full rights and that the right to life^{Reference Kluge83} or prenatal care^{Reference Keyserlingk84} override the mother's right to autonomy or inviolability^{Reference Bewley and Dickenson85} have led to jailing of women in the USA who took drugs in pregnancy or after birth if found positive on drug screening.^{Reference Hui, Angelotta and Fisher86} It has been argued that criminalising substance misuse in pregnancy would cause more harm than good as the worry of penalisation may lead mothers away from adequate perinatal care and support for cannabis use disorder.^{Reference Hui, Angelotta and Fisher86} By pitting the rights of the mother and fetus against each other, the discussion can easily become moralistic rather than evidence based and can perpetuate stigma associated with substance misuse during pregnancy. Furthermore, punitive measures place healthcare workers in difficult positions as they have also questioned the policy of drug screening in pregnancy without consent, their role in policing women who take cannabis in pregnancy and the disproportionate impact on women from minority ethnic groups.^{Reference Hui, Angelotta and Fisher86,Reference Hulsey87}

The principle of non-maleficence supports advising against cannabis use in pregnancy. However, for those who choose to use cannabis in pregnancy despite the warning, Bewley argues that as the fetus is a future member of society, society has a ‘legitimate interest’ in its welfare and should aim to support women who take drugs and minimise preventable harms to babies by offering incentives rather than using threats, coercion or punishment.^{Reference Bewley and Dickenson85} Harris also argued in favour of a gender- and equality-based approach to resolving the dilemma rather than perpetuating the age-long conflict-based ethics.^{Reference Harris11} In this model, clinicians faced with ethical dilemmas should attempt to understand pregnant women and their decisions within their broad social networks and communities, ask how the clinician's personal standpoint influences outcomes judged to be ethical and determine whether the clinician's ethical formulations reduce or enhance existing gender, class or racial inequality.^{Reference Harris11} This seems to be a more appropriate solution as there is no evidence to suggest that penalising drug use in the perinatal period leads to better maternal or fetal outcomes.^{Reference Hui, Angelotta and Fisher86} Bridging the gap between different views requires clarification of what constitutes benefits, harm and rights in relation to cannabis use as these are the core issues contended in the mother–fetus drug use debate.^{Reference Kluge83}

Proponents of recreational drug use contend that the government should intervene only when there is a high risk of causing harm to others.^{Reference Walsh88} Prohibitionary laws based on harm reduction have been challenged as punitive^{Reference Stafford89} and it has been proposed instead that harm reduction should be based on four principles: (a) drug use should be viewed neutrally, not moralistically; (b) a drug user is a citizen and member of a community, not a deviant individual or only an object of measures; (c) drug policy should be based on practice and science, not on ideologies and dogmatism; and (d) drug policy should respect human rights and support justice, not trample on them in the name of a ‘war on drugs’ or the goal of a drug-free society.^{Reference Tammi and Hurme90} These principles may also be useful in the context of healthcare, in that clinicians should ensure that women who use cannabis during pregnancy are treated in a neutral, non-stigmatising way and are supported to make healthcare decisions based on evidence, rather than ideology.

Table 1 shows ethical considerations and recommended approaches to navigating ethical dilemmas of cannabis use in pregnancy.

Table 1

Ethical principles and recommended approaches in navigating ethical dilemmas concerning cannabis use in pregnancy

Implications for doctors and other clinicians

It is important for doctors and other healthcare practitioners involved in supporting women of childbearing age to be aware of the benefits of medicinal cannabis and potential risks of cannabis use in pregnancy.^{Reference Philpot, Ebbert and Hurt91} This helps in discussing potential risks with patients and supporting them to make informed decisions.^{Reference Roncero, Valriberas-Herrero, Mezzatesta-Gava, Villegas, Aguilar and Grau-López24,Reference Glickman and Sisti80} With the legalisation of medicinal cannabis^{Reference Barnes41} and ongoing research into other areas of potential benefit, it is perhaps more challenging to correct the common perception that cannabis is relatively harmless.^{Reference Bahji, Stephenson, Tyo, Hawken and Seitz25} For example, in a survey of 51 healthcare providers to pregnant women in the USA, the providers (2nd–4th year obstetrics residents, nurse midwives and practitioners and faculty physicians) reported that they perceive cannabis use in pregnancy as less dangerous than use of other illicit drugs, while expressing the view that many pregnant women do not perceive cannabis as a drug.^{Reference Holland, Nkumsah, Morrison, Tarr, Rubio and Rodriguez92} The perception of low risk associated with cannabis use has also been reported among young people,^{Reference Roncero, Valriberas-Herrero, Mezzatesta-Gava, Villegas, Aguilar and Grau-López24} and more research is required to ascertain the effects of parental cannabis use on parenting and child and adolescent development.^{Reference Paul, Hatoum, Fine, Johnson, Hansen and Karcher71,Reference Berthelot, Garon-Bissonnette, Drouin-Maziade, Duguay, Milot and Lemieux93} Table 2 provides a summary of interventions at different perinatal stages. Some of these interventions are targeted at vulnerable women or at-risk groups, whereas others are indicated for public health.

Table 2

Indicated or targeted interventions by healthcare professional at different perinatal stages

Comprehensive assessment within a multidisciplinary framework

We emphasise the importance of a multidisciplinary approach within health and social services because of the impact of substance misuse on the fetus, the mother and the newborn child.^{Reference Day and George94} We also advocate supporting women at risk to maximise their health and well-being before pregnancy by improving lifestyle and diet. Accurate identification of cases, use of motivational interviewing techniques and access to cognitive–behavioural therapy (CBT) are vital.^{Reference Forray28} Comprehensive assessment, the cornerstone of management, should consider the risk to the physical and mental health of the mother during pregnancy, and ongoing childcare and parenting issues. This should lead to a plan and care package involving well-coordinated, multidisciplinary care with a specialist drug service working collaboratively with the general practitioner (GP), midwife, obstetrician, paediatrician and social worker.^{Reference Day and George94}

GPs’ ongoing relationship with patients enables them to establish rapport in order to discuss preconception health and promote positive changes in health and lifestyle. The World Health Organization (WHO) recommends the use of interventions that are brief, structured and require easy-to-administer tools. These interventions include information and motivational components that have been effective in primary care settings.^81,95 Furthermore, a Cochrane review found that a combination of motivational enhancement therapy (MET) and CBT with abstinence-based incentives had stronger evidence in the treatment of cannabis use disorder compared with drug counselling, social support, relapse prevention and mindfulness meditation.^{Reference Gates, Sabioni, Copeland, Le Foll and Gowing96}

Identifying cannabis withdrawal syndrome (CWS)^{Reference Budney and Hughes97} in regular and dependent users is important as it can confound mental illness or early signs and symptoms of pregnancy such as nausea, headache and fatigue, which is why some patients present late for treatment of mental illness or for antenatal care.^{Reference Day and George94} The DSM-5 diagnosis of CWS requires at least three of the following seven groups of symptoms developing within 7 days of reduced cannabis use: irritability, anger or aggression; nervousness or anxiety; sleep disturbance; appetite or weight disturbance; restlessness; depressed mood; and somatic symptoms such as headaches, sweating, nausea, vomiting or abdominal pain.⁹⁸ This underscores the importance of training for healthcare workers in identifying symptoms and signs of cannabis misuse, particularly in techniques to improve the accuracy of obtaining a history of cannabis misuse in pregnancy.^{Reference Anderson, Elk and Andres99}

Women may hesitate to disclose cannabis use owing to fear of judgement and concerns that their parenting ability may be questioned. Day & George^{Reference Day and George94} described the difficulty of accurately reporting the prevalence of high-risk drug use in pregnancy owing to underreporting resulting from ‘feelings of shame, denial and stigma experienced by the drug user, lack of awareness among professionals in antenatal services, the presence of comorbid psychiatric disorders, and sociocultural barriers that may prevent a thorough assessment’. Asking about type, quantity and impact of substance use by a trained professional is vital in evaluating risk. Wilson and colleagues have suggested questions to consider pertaining to substance misuse in the preconception or perinatal period. These should focus on empathic enquiry to assess the type and number of substances used, care and support available, impact on daily life and potential motivators for change.^{Reference Wilson, Finch, Kerr and Shakespeare100} Roncero and colleagues advocate harm minimisation strategies by early detection and alerting pregnant women to the risks of cannabis use to themselves and the unborn baby.^{Reference Roncero, Valriberas-Herrero, Mezzatesta-Gava, Villegas, Aguilar and Grau-López24} Owing to the likely underestimation from self-report (which may be linked to guilt, fear of legal consequences, stigma and social service referral), multiple urine screens during pregnancy may be a more reliable method to assess and monitor exposure to substances.^{Reference Roncero, Valriberas-Herrero, Mezzatesta-Gava, Villegas, Aguilar and Grau-López24} Urine toxicology in a group of pregnant females identified twice as many cannabis users compared with self-report (4.9 v. 2.5%).^{Reference Young-Wolff, Sarovar, Tucker, Goler, Conway and Weisner101} Maternal hair and the meconium of the newborn have also shown higher detection rates of cannabis use compared with clinical interviews.^{Reference García-Serra, Ramis, Simó, Joya, Pichini and Vall102} The collection window for urine toxicology is 24–48 h (longer in regular consumers of cannabis), whereas maternal hair provides the advantage of giving information on the presence of THC over several months, given that 1 month corresponds to roughly 1 cm of hair growth from the scalp; however, testing is limited by hair type, hair products, amount collected and processing. Meconium cannabinoid testing is more expensive and is limited by the narrow collection window of 72 h.^{Reference García-Algar, Vall Combelles, Puig Sola, Mur Sierra, Scaravelli and Pacifici103}

It is also important to ask patients about their consumption of non-medicinal CBD-containing foods or products, because of potential adverse effects. As stated earlier, there is limited research on the effects of CBD on embryonic development; however, several animal studies suggest that CBD can cause fetal harm.^{Reference Huestis, Solimini, Pichini, Pacifici, Carlier and Busardò104} More specifically, animal studies on rhesus monkeys, dogs and rats showed that CBD exposure resulted in problems such as hepatoxicity, immunotoxicity, reproductive toxicity, changes to organ weights (i.e. increase in weight of liver, kidneys, heart, thyroid, thymus, spleen and adrenal glands and decrease in testicular weight) and alterations to cytochrome P450 drug-metabolising enzymes.^{Reference Huestis, Solimini, Pichini, Pacifici, Carlier and Busardò104} Obtaining relevant information during prenatal or antenatal reviews therefore helps to target interventions using a person-centred approach.^{Reference Roncero, Valriberas-Herrero, Mezzatesta-Gava, Villegas, Aguilar and Grau-López24} Finally, the high rate of relapse of cannabis use 3 months after delivery indicates that ongoing support should be made available to susceptible women and their infants in the postnatal period.^{Reference Forray, Merry, Lin, Prah Ruger and Yonkers77}

Cannabis use during breastfeeding is a key dilemma for clinicians as the benefits of breastfeeding are very well known but early evidence indicates that CBD and THC can be transferred to the infant via breast milk.^{Reference Bertrand, Hanan, Honerkamp-Smith, Best and Chambers65} Current guidelines recommend abstaining from cannabis use during breastfeeding,¹⁰⁵ but clinicians may still be conflicted between encouraging breastfeeding for the benefit of the child versus encouraging abstinence from cannabis-based products despite the potential medicinal benefits for the mother.^{Reference Ryan106} Although research into cannabis use during breastfeeding is in its infancy, it is clear that a system-wide, multidisciplinary approach is needed to inform nursing mothers about the effects of cannabis on infants.^{Reference Graves107} For example, pharmacists and lactation consultants are in unique positions and may be able to identify those who are using cannabis, offer advice on abstinence, support with education on the known risks and benefits of cannabis use during breastfeeding and provide referrals for treatment, if required.^{Reference Davis, Lee, Weber and Bugden108,Reference Skelton, Benjamin-Neelon and Young-Wolff109}

Limitations

The findings of this review should be seen in light of some limitations. This paper is a narrative review aiming to offer an objective analysis in the current context of decriminalisation and licensing of cannabis for medical use and its ethical considerations during pregnancy. A systematic review with rigorous methodological approaches, considering new and emerging evidence assessing the impact of licensing of medicinal cannabis in the UK, may be able to draw more robust conclusions on this topic of significant public health interest.