In pregnancy, the mechanisms underlying the effects of psychostimulants on the developing foetus are complex. Currently available theories are that they block the neuronal reuptake of catecholamines in the mother, resulting in cardiac stimulation and vasoconstriction. This leads to decreased uterine blood flow and thus to a decrease in the transfer of oxygen and other nutrients to the foetus. In addition to these mechanisms, psychostimulants act on serotonergic or noradrenergic transporters expressed in placental cells (Ramamoorthy, Ramamoorthy, Leibach & Ganapathy, 1995). This may increase levels of monoamines in the intervillous space (further adding to the vasoconstrictive effect) and restricting blood flow to the placenta. Elevation of serotonin and noradrenaline levels occurring via this mechanism may also alter uterine contractility.
If the mother continues to use psychostimulants when breast-feeding, the effects on the infant are variable. While risks of infant exposure are extended, this may be offset by the amelioration of withdrawal symptoms in the first month of life.
Although psychostimulants share a common spectrum of pharmacological activity, it cannot be assumed that their impact on a developing foetus is also the same. For this reason, the following section is separated into individual drugs.
CocaineMost of the clinical and animal research conducted into effects of psychostimulants in pregnancy and lactation have focused on cocaine. Cocaine and its metabolites do cross the placenta. Cocaine does accumulate in the placenta (Ursitti, Klein & Koren, 2001), where it may be metabolised by placental microsomes. Whilst this may protect the foetus after bolus administration, placental retention may also prolong foetal exposure.
Cocaine exerts a number of actions on the foetus. Inhibition of noradrenaline reuptake leads to maternal vasoconstriction, which reduces uterine and placental blood flow (Lipton, Vu, Ling, Gyawali et al., 2002; Patel, Laungani, Grose & Dow-Edwards, 1999; Sutliff, Gayheart-Walsten, Snyder, Roberts & Johnson, 1999). Cocaine also has a direct effect on the foetus, resulting in foetal vasoconstriction and other cardiovascular changes (Fomin, Singh, Brown, Natarajan & Hurd, 1999; Shearman & Meyer, 1999; Yakubu, Pourcyrous, Randolph, Blaho et al., 2002). Top of page
Cocaine and teratogenic effectsAlthough a large volume of animal research has examined this issue, there are conflicting results with regard to its teratogenic potential. This is partly due to differences in teratogenic effects between animal species, which highlight the need for caution when extrapolating results from animal research to humans.
In humans, case reports describe a range of congenital malformations occurring in infants exposed in utero to cocaine. These include malformations of the genitourinary tract, heart, limbs and face (Bingol, Fuchs, Diaz, Stone & Gromisch, 1987; Chasnoff, Chisum & Kaplan, 1988; Little, Snell, Klein & Gilstrap, 1989; Viscarello, Ferguson, Nores & Hobbins, 1992).
However, many controlled studies fail to demonstrate such anomalies (Addis, Moretti, Syed, Einarson & Koren, 2001; Frank et al., 2001). In particular, one study reported that women who used cocaine in the first trimester only (which is the teratogenic period), demonstrated similar obstetric outcomes to drug free controls (Chasnoff, Griffith, MacGregor, Dirkes & Burns, 1989). In a large, blinded prospective study (Behnke, Eyler, Garvan & Wobie, 2001), there was no evidence to suggest that cocaine contributes to the development of gross abnormalities in humans. The authors state that "if cocaine does produce human malformations, it seems to do so at a very low rate or only under certain conditions, perhaps related to such events as the amount and timing of the exposure, or to the simultaneous ingestion of other substances" (Behnke et al., 2001).
The number of reports of cocaine-associated malformations is of concern to many clinicians. However, interpreting these studies requires caution due to the lack of a consistent pattern in the anomalies described and the inconsistencies in research results (Buehler, Conover & Andres, 1996). Infants exposed to in utero cocaine may have a higher risk of malformations but evidence to date has failed to consistently link cocaine exposure with organ specific anomalies (teratogenicity).
Other outcomes associated with cocaine exposure during pregnancy
Obstetric complicationsA large amount of research has examined the relationship between cocaine use during pregnancy and obstetric outcomes. Although a range of poor obstetric and neonatal outcomes have been attributed to foetal cocaine exposure, research results are conflicting. It is thought that the cardiovascular effects of cocaine (producing maternal vasoconstriction and direct foetal effects) may be the main drug effects influencing obstetric outcomes.
In human studies, cocaine use during pregnancy has been associated with reduced growth such as lower birth weight, reduced length and reduced head circumference (Bada, Das, Bauer, Shankaran et al., 2002; Richardson, Hamel, Goldschmidt & Day, 1999), still birth related to abruptio placenta (Bauer, Shankaran, Bada, Lester et al., 2002; Bingol et al., 1987; Little, Snell, Trimmer, Ramin et al., 1999), intracranial haemorrhage in the neonate (Spires, Gordon, Choudhuri, Maldonado & Chan, 1989) and sudden infant death syndrome (Durand, Espinoza & Nickerson, 1990). Other studies have found no relationship between cocaine exposure and pre-term births (Savitz, Henderson, Dole, Herring et al., 2002), sudden infant death syndrome (Fares, McCulloch & Raju, 1997) or between cocaine exposure and morbidity associated with pre-term premature rupture of membranes (Refuerzo, Sokol, Blackwell, Berry et al., 2002).
One issue that may influence results of these studies is timing of cocaine exposure. Chasnoff and colleagues (Chasnoff et al., 1989) compared women who had used cocaine in first trimester only with those who continued to use cocaine throughout pregnancy. Those who used cocaine throughout pregnancy had increased rates of pre-term births, low birth-weight infants and intrauterine growth retardation. Those who used cocaine in first trimester only had similar outcomes to a drug free control group.
One of the challenges with interpreting research in this area is that many studies have failed to take into account potential confounding factors (Gressens, Mesples, Sahir, Marret & Sola, 2001). Women who use cocaine whilst pregnant are more likely to use other drugs, especially alcohol and tobacco (Bada et al., 2002; Singer, Arendt, Minnes, Farkas & Salvator, 2000) and are more likely to exhibit other risk factors for poor obstetric outcomes, such as greater levels of maternal distress (Singer, Salvator, Arendt, Minnes et al., 2002), low socio-economic status, low levels of education and poor maternal nutrition (Savitz et al., 2002). As such, only limited conclusions regarding the role of cocaine as the causal agent can be made from such research. One author suggests that nicotine and cocaine produce similar effects on the foetus, but pharmacokinetic characteristics of cocaine and its patterns of use mean that periods of recovery exist and the "eventual consequences are much less severe" when compared with tobacco use (Slotkin, 1998).
One meta-analysis (Addis et al., 2001) attempted to untease the relationship between cocaine exposure and other risk factors. Outcomes assessed included rates of major malformations, low birth weight, premature birth, placental abruption, premature rupture of membranes and mean birth weight, length and head circumference. After adjusting for confounding factors, only the risk of placental abruption and premature rupture of membrane remained attributable to cocaine use.
In a systematic review, Frank and colleagues (Frank et al., 2001) report that there is no convincing evidence that prenatal cocaine exposure confers a greater risk of developmental toxic effects than multiple other factors. They report that although some decrements in measures of physical growth such as birth weight or head circumference are reported after cocaine exposure, once studies have controlled for alcohol or tobacco use, no negative effects of cocaine are observed. Such reports highlight the complexity of interpreting research in this area — many findings once thought to be specific effects of in utero cocaine exposure can be explained in whole or in part by other factors including prenatal exposure to tobacco, marijuana or alcohol and the quality of the child's environment. Birth weights may in fact be improved by the provision of prenatal care (Racine, Joyce & Anderson, 1993). It is likely that there is a complex interaction between dose response effects of the drug, cumulative environmental and other risk factors (Kaltenbach, 2000).Top of page
Neurobehavioural developmentIn addition to cardiovascular effects, cocaine produces a range of neurochemical effects. This has led to concern that prenatal cocaine exposure may influence brain development.
A range of animal studies have been conducted. They report that cocaine exposure may produce abnormal neocortex development (Lidow, Bozian & Song, 2001), lasting metabolic changes within specific brain regions associated with arousal, attention and stress responses (Dow-Edwards, Freed-Malen & Gerkin, 2001), changes in dopaminergic activity (Lipton, Ling, Vu, Robie et al., 1999), changes in circadian activity (Strother, Vorhees & Lehman, 1998) and disruption of short-term memory (Morrow, Elsworth & Roth, 2002). A meta-analysis of research of prenatal cocaine exposure on the development of the nigrostriatal dopamine system in animals (Glatt, Bolanos,Trksak & Jackson, 2000) found that cocaine exposure led to negligible effects on most indicators of dopamine function. Some authors suggest that neurodevelopmental changes observed in animals may explain behavioural changes observed in human studies. However, this has not been adequately explored.
The results of human research into neurobehavioural development have been conflicting. Some clinical research suggests that prenatal cocaine exposure may lead to problem behaviours (Delaney-Black, Covington, Templin, Ager et al., 1998), deficits in attentional processing (Coles, Bard, Platzman & Lynch, 1999), behavioural abnormalities such as jitteriness (Singer et al., 2000), poorer cognitive, motor and language development and reduced emotional responsivity (Singer, Hawkins, Huang, Davillier & Baley, 2001).
In a study examining children at three, five and seven years (Bandstra, Morrow & Anthony, 2001), a stable influence of prenatal cocaine exposure was observed on indicators of sustained attention and task vigilance. These effects were maintained after controlling for prenatal exposure to other substances and additional medical and sociodemographic variables. They also observed a more pronounced effect for children whose mothers had a heavy alcohol intake in addition to cocaine use.
In a prospective study (Morrow, Bandstra, Anthony, Ofir et al., 2001), a range of subtle deficits across the spectrum of neurobehavioural functioning were observed within the first postnatal week in infants with cocaine exposure. These deficits were partly correlated with reduced foetal growth. The deficits in functioning were larger as the number of trimesters of exposure increased. The authors suggest that prenatal cocaine exposure may produce more problematic effects in infants born prematurely and that cocaine exposed full-term infants may be more resilient. Other authors also suggest that any effect of cocaine on longer-term development is an indirect association, mediated by reduced birth weight, head circumference, other drug use or other prenatal issues (Behnke, Eyler, Garvan, Wobie & Hou, 2002; Bendersky & Lewis, 1999). In addition, one controlled study reports that mothers in a cocaine-exposed group had less frequent emotional contact with their infant and tended to have maladaptive coping strategies compared with a non-exposed group (Singer et al., 2001). The authors suggest that interventions targeted at maternal parenting skills may be of some benefit.
Bennett and colleagues (Bennett, Bendersky & Lewis, 2002) found that in utero cocaine exposure was largely unrelated to IQ and adjustment skills at four years, particularly for girls. The systematic review by Frank and colleagues (Frank et al., 2001) reports a lack of relationship between cocaine exposure and cognitive performance, behaviour and affect after controlling for alcohol and tobacco use.
In conclusion, it is likely that cocaine use contributes to infant neurobiological and behavioural outcomes in cumulation with other drug exposure, maternal and environmental factors.
Cocaine and ethanolIn animal studies alcohol alone or cocaine alone may lead to reduced foetal weights or foetal mortality (Ohnaka, Ukita, Yamamasu, Inoue et al., 2001). When alcohol is used concurrently with cocaine, liver esterases transesterify cocaine to produce cocaethylene, which is considered a more potent vasoconstrictor than cocaine. As such, it is thought that exposure to a combination of alcohol and cocaine may be more deleterious to pregnancy and foetal outcome than either drug alone (Randall, Cook, Thomas & White, 1999; Snodgrass, 1994).
Cocaine exposure and neonatal withdrawal syndromesThe literature on prenatal cocaine exposure is unclear whether immediate postpartum effects on the infant are transient, related to either acute toxicity of cocaine, or to a withdrawal effect. Infants born after cocaine exposure may exhibit a range of symptoms including tone and movement abnormalities, brisk or excessive reflexes, jitteriness, irritability and poor feeding (Eyler et al., 2001). One of the difficulties of interpreting this area of research is the lack of consistent measures used to describe infant behaviours and the frequent inclusion of neonates concurrently exposed to opiates in utero.
In a recent prospective study (Eyler, Behnke, Garvan, Woods et al., 2001) examining 154 neonates with prenatal cocaine, no dramatic effects of toxicity or withdrawal were observed. Deficits in neurobehavioural functioning were minor and improved in most infants within the first week. The presence or severity of a neonatal withdrawal syndrome may be influenced by extent and frequency of exposure during the period immediately prior to parturition. Top of page
Cocaine and risk during lactationCocaine and its metabolites pass into the breast milk. Chasnoff and Lewis describe a case of cocaine intoxication in a breast-fed infant (Chasnoff, Lewis & Squires, 1987). The baby experienced irritability, vomiting, diarrhoea, tremulousness and seizures subsequent to maternal cocaine. In light of such case reports, the American Academy of Pediatrics Committee on Drugs questions the appropriateness of breastfeeding a baby where the mother is using cocaine (American Academy of Pediatrics Committee on Drugs, 2001).
Amphetamines and teratogenic effectsAnimal studies have reported that amphetamine use is associated with cardiac malformations (Nora, Trasler & Fraser, 1965) or other malformations (Acuff-Smith, George, Lorens & Vorhees, 1992; Kasirsky, 1971). In humans, there are a number of case reports linking amphetamine exposure with malformations (Gilbert & Khoury, 1970; Matera, Zabala & Jimenez, 1968). One study found a positive relationship between dexamphetamine exposure and heart defects (Nora, Vargo, Nora, Love & McNamara, 1970).
A number of other studies have failed to demonstrate a relationship between malformations and amphetamine exposure (Heinonen, Slone & Shapiro, 1977; Little, Snell & Gilstrap, 1988; Milkovich & van der Berg, 1977; Nora, McNamara & Fraser, 1967). One report from a teratogen information service (Felix, Chambers, Dick, Johnson & Jones, 2000) found that methamphetamine abuse was not associated with increased rates of congenital anomalies. However, exposed infants did have minor physical anomalies, irritability or other signs of neurological dysfunction. This is consistent with theoretical data.
In contrast to these earlier studies, Sherman and Wheeler-Sherman (2000) reported major congenital abnormalities in 16% of infants whose mothers had used methamphetamine, often in combination with alcohol or other illicit drugs. Most of the anomalies found were cardiac defects, but also included gastroschisis and hydronephrosis. It is not established if this rate of anomalies is higher than would be expected. Approximately 5% of infants in the exposed group also had neonatal thrombocytopaenia. Unfortunately, this study was uncontrolled and conducted retrospectively. A recent controlled study examined full-term neonates exposed to methamphetamine in utero (Smith, Yonekura, Wallace, Berman et al., 2003). Although examining foetal growth and withdrawal symptoms were the aims of this study rather than teratogenicity, 134 neonates were assessed to have no malformations or anomalies.
From these studies, it would seem that use of amphetamines in regular low doses (e.g. when prescribed for therapeutic purposes such as ADHD) poses little teratogenic risk. Further research is required to address the possible risk of cardiac malformations and whether dependent or binge patterns of amphetamine use may confer a greater risk to the foetus. Alternatively, any possible increase in negative outcomes associated with amphetamine use may be attributed to causes such as other drug use or environmental factors associated with illicit drug use.
Other outcomes are associated with amphetamine exposure during pregnancy
Obstetric complicationsSimilarly to cocaine, prenatal exposure to methamphetamine may cause cardiovascular alterations including increased maternal and foetal blood pressure, reduced foetal oxyhaemoglobin saturation and a decrease in uterine blood flow (Nora, 1968). At least some of these effects seem to be dose-related (Yamamoto, Yamamoto, Fukui & Kurishita, 1992).
A number of studies have reported an association between amphetamine exposure and outcomes related to growth retardation such as reduced body weight, reduced length and head circumference at birth (Little et al., 1988; Naeye, 1983; Smith et al., 2003). Other adverse effects include stillbirth (Dearlove, Betteridge & Henry, 1992) and intracranial haemorrhage (Dixon & Bejar, 1989).
Neurobehavioural developmentWilliams and colleagues (Williams, Vorhees, Boon, Saber & Cain, 2002) report that rats exposed to methamphetamine between postnatal days 11 to 20 developed behavioural and spatial learning impairments. Methamphetamine may lead to alterations in dopaminergic (Heller, Bubula, Freeney & Won, 2001) or serotonergic (Tavares, Silva, Silva-Araujo, Xavier & Ali, 1996) systems. One review (Frost & Cadet, 2000) suggests that exposure to methamphetamine is likely to produce changes in neural circuitry in a wide variety of brain regions, but assessing the clinical and functional significance of this remains a challenge.
One prospective study examined 65 children who had been exposed to amphetamines in utero (Cernerud, Eriksson, Jonsson, Steneroth & Zetterstrom, 1996). Deficits in learning in children with prenatal amphetamine exposure were observed compared to matched controls. Top of page
Amphetamine exposure and neonatal withdrawal syndromesNeonatal withdrawal syndromes have been reported after methamphetamine exposure (Oro & Dixon, 1987; Ramer, 1974). Symptoms included poor feeding, abnormal sleep patterns, tremors and increased muscle tone. A study on 134 neonates with prenatal methamphetamine exposure reported that 49% of neonates experienced withdrawal symptoms, although only 4% required pharmacological intervention (Smith et al., 2003).
Amphetamines and risk during lactationAmphetamines are excreted into breast milk and, depending on the dose, measurable amounts can be detected in the urine of the infant. In one study of 103 nursing infants whose mothers were taking amphetamines, no neonatal insomnia or stimulation was observed over a 24 hour period (Ayd, 1973).
MDMAMDMA is a substituted amphetamine. In addition to a range of amphetamine-like properties, it has a greater range of serotonergic and potentially hallucinogenic properties. There have been limited animal studies exploring the effects of MDMA during pregnancy and within existing research, results have been mixed.
MDMA and teratogenic effectsAt this stage, there is insufficient evidence to make firm conclusions about the potential teratogenicity of MDMA. One animal study found no effects of MDMA on rates of malformations (St Omer, Ali, Holson, Duhart et al., 1991). Doses utilised were 0, 2.5, or 10 mg/kg of MDMA.
There are only a few studies on the effects of MDMA on human pregnancy. One case of congenital heart disease was reported in an abstract examining 489 pregnancies (Rost van Tonningen, Garbis & Reuvers, 1998). In another abstract examining 38 pregnancies, one case of congenital heart disease and one possible case of omphalocele were reported (van Tonningen, Garbis & Reuvers, 1998).
The UK National Teratology and Information Service (NTIS) collected prospective follow-up data from 1989 to 1998 on 136 pregnancies following primarily first trimester exposure to MDMA (McElhatton, Bateman, Evans, Pughe & Thomas, 1999). 35% of these women had elective terminations (one after prenatal diagnosis of malformations) and 10% had miscarriages. Of the remaining 78 live-born infants, over 15% had congenital malformations, especially cardiovascular and musculoskeletal anomalies. This is five to sevenfold higher than the expected incidence of 2-3%. Although other factors were not well controlled for, the high rate of anomalies reported in this study indicates a possible association between MDMA exposure and congenital anomalies. MDMA shares many pharmacological properties with amphetamines. However, its structural differences (namely the methylenedioxy ring substitution) may mean that it displays a different teratogenic profile. Further controlled studies are required in this area.
Other outcomes are associated with MDMA exposure during pregnancy
Obstetric complicationsAnimal studies have reported that MDMA exposure led to no effect on litter size or birth weight (St Omer et al., 1991) and no effect on body, brain and liver weight (Bronson, Barrios-Zambrano, Jiang, Clark et al., 1994). In contrast, other studies have reported reduced maternal weight gain and litter size (Colado, O'Shea, Granados, Misra et al., 1997) and reduced embryonic motility (Bronson, Jiang, Clark & DeRuiter, 1994).
Although no research has examined the relationship between MDMA and obstetric complications, given that MDMA shares many pharmacological effects with amphetamines, it is likely that MDMA shares similar effects to amphetamines such as intrauterine growth retardation. Top of page
Neurobehavioural developmentAnimal studies have reported that MDMA exposure is associated with increases in serotonergic and dopaminergic markers (Won, Bubula & Heller, 2002) or long-term effects on cerebral function (Kelly, Ritchie, Quate, McBean & Olverman, 2002). Some studies demonstrate vulnerability to MDMA-related neurotoxicity (Meyer & Ali, 2002), whereas Colado and colleagues (1997) reported that although female rats demonstrated signs of neurotoxicity, this was not observed in their offspring.
In another study (Broening, Morford, Inman-Wood, Fukumura & Vorhees, 2001), rats exposed to MDMA in the late stages of pregnancy showed dose-related impairments of sequential and spatial learning and memory, while rats exposed at an earlier time (equivalent to early third trimester) showed no significant impairment. This might suggest vulnerability of the brain to MDMA later in its development but drawing any further conclusions from a single animal study is clearly tenuous.
No studies in humans have been undertaken on the effects of prenatal MDMA exposure on neurobehavioural development.
As discussed in earlier sections, psychostimulant users often display a range of other risk factors associated with poor neonatal outcomes. Ho and colleagues (Ho et al., 2001) compare pregnant women reporting use of MDMA with pregnant women not exposed to MDMA. They report that women using MDMA were more likely to binge drink during pregnancy, use tobacco and other illicit drugs. Pregnancies were more likely to be unplanned and women were more likely to be young, single and experiencing a range of psychological problems. This reaffirms the need to account for a range of confounding risk factors when undertaking research in this area.