10

Prenatal, Perinatal, and Neonatal Exposure to Cannabis

The issue of exposure to cannabis during pregnancy reflects concerns that two different individuals may experience the potential adverse effects of cannabis, which is the illicit drug used most frequently by women of childbearing age. The Substance Abuse and Mental Health Services Administration’s National Survey on Drug Use and Health found that in 2015, 3.4 percent of pregnant women ages 15 to 44 had used marijuana during the previous month (CBHSQ, 2016). This is compared to 0.8 percent of pregnant women who used pain relievers, the next most used illicit drug among pregnant women (CBHSQ, 2016). In part because cannabis is an illicit drug, there is very little information on the physiological effects of cannabis in pregnancy on the mother. Moreover, most of the data reflect cannabis administered by smoking and not cannabis exposure through other routes of administration.

Concern about the fetus and newborn stems from the fact that tetrahydrocannabinol (THC) crosses the placenta (Bailey et al., 1987). A rapidly

growing body of evidence indicates that endocannabinoids play roles in a broad array of critical neurodevelopmental processes, from early neural stem cell survival and proliferation to the migration and differentiation of both glial and neuronal lineages as well as neuronal connectivity and synaptic function (Lubman et al., 2014). Another potentially important issue is that THC is secreted in breast milk and can accumulate to high concentrations (Garry et al., 2009).

This chapter focuses on exposure to cannabis from the beginning of pregnancy through the infant’s first month of life. Thus, the review covers complications of pregnancy, fetal effects, exposure through breast milk, and later effects of fetal exposure. Although the general principle of the overall report is to restrict the literature reviewed to that which has emerged since the publication of Marijuana and Medicine: Assessing the Science Base (IOM, 1999), the last Institute of Medicine report on marijuana, the committee chose to include information concerning longer-term outcomes from two older cohorts released in the 1980s, with the rationale that the identification of late adolescent and young adult outcomes would require that length of follow-up. The committee hand-searched additional literature to examine other prioritized long-term health outcomes not covered in these cohort studies.

The committee identified only one recent good- to fair-quality systematic review (Gunn et al., 2016). This review sought information on a comprehensive set of complications of pregnancy and on fetal and neonatal outcomes up to 6 weeks postpartum. Several lower-quality systematic reviews (Fryers and Brugha, 2013; Irner, 2012; Savitz and Murnane, 2010; Williams and Ross, 2007), narrative reviews (Andrade, 2016; Forray et al., 2015; Hashibe et al., 2005; Huang et al., 2015; Huizink, 2014; Metz and Stickrath, 2015; Schempf, 2007; Viteri et al., 2015), and articles from the grey literature (CDPHE, 2015) were used to identify outcomes not reviewed in Gunn et al. (2016), as was a bibliographic search of materials published from 1999 onward. A literature search was also conducted for outcomes in Gunn et al. (2016), from 2014 to August 2016, to identify any more recent articles. The committee identified 30 primary literature articles that best address the committee’s research questions of interest.

PREGNANCY COMPLICATIONS FOR THE MOTHER

Is There an Association Between Cannabis Use and Pregnancy Complications for the Mother?

Stillbirth and Spontaneous Abortion

Systematic Reviews The committee did not identify a good- to fair-quality systematic review that reported on the association between cannabis exposure and stillbirth or spontaneous abortion.

Primary Literature Varner et al. (2014) used results from a population-based case-control study conducted by the Stillbirth Collaborative Research Network to compare illicit drug use in pregnancies that did and did not result in stillbirth.¹ Among 663 stillbirth deliveries, women who had a stillbirth were twice as likely as those with a live birth to report having been addicted to an illicit drug. Tetrahydrocannabinolic acid (THCA), the most common individual drug reported by the population, was found in 2.9 percent of women with a stillbirth and 1.7 percent of the controls (odds ratio [OR] for stillbirth, 2.34; 95% confidence interval [CI] = 1.13–4.81). However, the authors indicate that the result may have been partially confounded by exposure to cigarette smoking and that they may not have had the statistical power to disentangle this effect.

Warshak et al.’s 2015 study on the association between marijuana exposure and adverse neonatal outcomes included stillbirth in the outcomes they examined and found no association (1.1 percent among 361 cannabis users versus 1.5 percent among 6,107 cannabis nonusers; p = 0.54).

Fetal Distress

Systematic Reviews Gunn et al. (2016) found no association between marijuana use and fetal distress based on two studies (Berenson et al., 1996; Witter and Niebyl, 1990).

Primary Literature The committee did not identify any good-quality primary literature that reported on the association between cannabis use and fetal distress and that were published subsequent to the data collection period of the most recently published good- or fair-quality systematic review addressing the research question.

___________________

¹ Fetal death was defined in the study as 20 weeks of gestation or less (Varner et al., 2014).

Other Complications

Systematic Reviews The assessment of the literature on pregnancy complications for the mother relied primarily on Gunn et al. (2016). Of the possible complications, only the increased risk of anemia had a significant association with exposure to cannabis (pooled odds ratio [pOR], 1.36; 95% CI = 1.10–1.69). Mixed findings about an association with cannabis use occurred in studies of precipitate labor and the manual removal of the placenta. No associations were found between in utero exposure to cannabis and the following health outcomes: maternal diabetes, rupture of membranes, premature onset of labor, use of prenatal care, duration of labor, placental abruption, secondary arrest of labor, elevated blood pressure, hyperemesis gravidarum, maternal bleeding after 20 weeks, antepartum or postpartum hemorrhage, maternal weight gain, maternal postnatal issues, duration of maternal hospital stay, or hormone concentrations (Gunn et al., 2016).

Primary Literature Three further studies were identified: Budde et al. (2007), Leemaqz et al. (2016), and Warshak et al. (2015). These studies examined the association between cannabis exposure and the following outcomes: anemia, precipitate labor, manual removal of the placenta, maternal diabetes, rupture of membranes, premature onset of labor, use of prenatal care, duration of labor, secondary arrest of labor, elevated blood pressure, hyperemesis gravidarum, maternal bleeding after 20 weeks, antepartum or postpartum hemorrhage, placental abruption, maternal weight gain, maternal postnatal problems, and duration of maternal hospital stay.

Findings in Leemaqz et al. (2016) from 313 women who used cannabis during pregnancy and Warshak et al. (2015) from 4,892 women who used cannabis during pregnancy were consistent with there being no significant association between cannabis exposure and gestational diabetes (adjusted odds ratio [aOR], 1.11; 95% CI = 0.52–2.38; p = 0.949 and aOR, 0.87; 95% CI = 0.66–1.04; p = 0.04, respectively) or gestational hypertension/preeclampsia (aOR, 0.443; 95% CI = 0.13–3.54; p = 0.671 and aOR, 0.84; 95% CI = 0.68–1.04; p = 0.12, respectively). Warshak et al. (2015) did not find a statistically significant association between cannabis use and placental abruption (aOR, 1.17; 95% CI = 0.81–1.70; p = 0.25). Budde et al. (2007) reported an increased risk of placental abruption that did not achieve standard statistical significance (OR, 2.83; 95% CI = 0.86–10.78; p = 0.055).

Discussion of Findings

Despite identifying one good- to fair-quality systematic review addressing pregnancy complications for the mother, the findings of the review must be interpreted with caution. The review relied on a primary literature that is limited in the number, quality, and rigor of the studies that have been carried out to date. By and large, the existing studies have been retrospective cohort studies, many of which looked at a large number of outcomes without biological plausibility or a biological mechanism guiding the test of the hypothesis. For example, the association identified between anemia and cannabis use in pregnancy arises in the absence of a clear mechanism by which these factors would be related. In addition, many studies were underpowered to detect relatively rare pregnancy complications. Therefore, though Gunn’s review reports “no association” for the vast majority of conditions selected, it remains unclear whether this represents a type II error. Ethical challenges obviously preclude the ability to conduct randomized controlled trials of cannabis use in pregnancy, thereby precluding the ability to establish causal relationships. Logistical and financial constraints make even prospective cohort studies of adequate size and duration challenging to fund and implement. Even with rigorous study designs, comorbid tobacco and polysubstance use often confound the interpretation of the data. Such considerations markedly diminish the confidence with which the committee can draw conclusions regarding how much risk can be attributed to cannabis in the area of adverse maternal events.

FETAL GROWTH AND DEVELOPMENT

Is There an Association Between Cannabis Use and Fetal Growth and Development?

Birthweight

Systematic Reviews Studies reviewed in Gunn et al. (2016) that examined the effect of cannabis exposure on birth weight reported both mean birth weights and the percentage of infants at low birth weight (LBW; defined as 2.2kg or 5.5 lbs). Gunn et al. (2016) found that in utero exposure to cannabis is associated with a decrease in birth weight among cannabis-exposed infants (pOR, 1.77; 95% CI = 1.04–3.01; pooled mean difference

[pMD], −109.42 grams; 95% CI = −38.72 to −180.12) compared to those without cannabis exposure.

Primary Literature Similar to the findings reported by Gunn et al. (2016), Gray et al. (2010) and Fergusson et al. (2002) also reported lower mean birth weights for infants prenatally exposed to cannabis. Among 9,521 mothers, Fergusson et al. (2002) showed a −84.20 gram difference (95% CI = −174.7 to −6.4; p = 0.005) in birth weight for the children of mothers who had used cannabis at least once per week before and throughout pregnancy versus nonusers. Out of 86 total infants of cannabis-using mothers (independent from tobacco use), Gray et al. (2010) reported a mean birth weight of 3,161 grams (standard deviation [SD], 689; p = 0.051) among 41 infants who had been exposed to cannabis and 3,417 grams (SD, 504; p = 0.051) among 45 infants who had not been exposed to cannabis. In contrast, Schempf and Strobino (2008) found that, when adjusted for other drug use (i.e., cocaine and opiates), there was no significant association between cannabis use and LBW (defined as less than 2,500 grams) (aOR, 0.93; 95% CI = 0.55–1.57).

Birth Length

Systematic Reviews In their systematic review, Gunn et al. (2016) found that for the nine studies that reported neonatal length at birth (measured in centimeters), there was no statistically significant association between neonatal length and prenatal exposure to cannabis (pMD, −0.10; 95% CI = −0.65–0.45).

Primary Literature Birth length was also examined by Fergusson et al. (2002), who found that children who had been exposed to cannabis in utero had a lower birth length than children who had not been prenatally exposed to cannabis. However, after adjusting for various confounding factors (e.g., cigarette smoking during pregnancy, alcohol consumption during pregnancy), the association was no longer significant (p = 0.225). Similarly, Gray et al. (2010) found nonsignificant differences in birth length between 41 infants of cannabis-using mothers (independent from tobacco use) (49.8 cm; SD, 3.8; p = 0.156) and 45 infants of non-using mothers (50.8 cm; SD, 2.2; p = 0.156).

Head Circumference

Systematic Reviews Gunn et al. (2016) found that among the 10 studies they reviewed that measured head circumference at birth, no statistical

association was found between cannabis exposure in utero and neonatal head circumference (cm) (pMD, −0.31; 95% CI = −0.74–0.13).

Primary Literature The committee did not identify any good-quality primary literature that reported on the association between cannabis use and head circumference and that was published subsequent to the data collection period of the most recently published good- or fair-quality systematic review addressing the research question.

Intrauterine Growth Restriction/Small for Gestational Age

There are two ways to describe slower-than-expected growth for a particular duration of gestation. The first is intrauterine growth restriction (IUGR), an obstetric diagnosis based on serial ultrasounds during pregnancy. The second is small for gestational age (SGA), which applies to infants with a birth weight that is less than the 10th or 5th percentile on normative growth curves. The limitation of the latter is that it does not distinguish between those infants with true slow growth and those with normal growth in the lower percentiles.

Systematic Reviews Gunn et al. (2016) addressed two studies that looked at the relationship between in utero cannabis exposure and SGA and concluded that no association can be reported on the association between exposure to cannabis during pregnancy and IUGR/SGA. A pOR was not reported.

Primary Literature Leemaqz et al. (2016) similarly did not find an association between cannabis exposure and SGA (defined as a birth weight less than the 10th percentile) when adjusted for any smoking (aOR, 1.13; 95% CI = 0.80–1.60). In a path analysis of urban black women who reported cannabis use at 50 percent of prenatal visits, Janisse et al. (2014) found a reduction in birth weight for heavy marijuana use alone (−55.2 grams), with a path coefficient of 0.05.² Their analysis suggests that LBW resulting from cannabis exposure reflects fetal growth restriction rather than premature delivery.

Congenital Malformation

In this category the committee considered infants who had malformations or anomalies diagnosed prenatally or after birth. Congenital malformations reflect abnormalities of fetal development in one or more

___________________

² The authors used a z-score of birth weight for duration of gestation residualized.

organ systems and can occur throughout pregnancy. They may be identified before or after birth.

Systematic Reviews Gunn et al. (2016) reported no association between cannabis exposure and chromosomal anomalies. No estimate of effect was provided.

Primary Literature Warshak et al. (2015) analyzed data from among 4,892 cannabis users and 153 marijuana cannabis nonusers and reported no association between cannabis exposure and fetal anomalies (aOR, 1.29; 95% CI = 0.87–1.92). In contrast, Forrester and Merz (2006) found higher rates of cannabis use to be associated with the presence of 19 defects out of a total of 54 selected conditions.³ However, this study only performed bivariate comparisons for exposure/no exposure without considering other substances, confounders, or multiple comparisons.

Two case-control studies of the association of cannabis exposure to specific malformations were found. Using data from the National Birth Defects Prevention Study (1997–2005), van Gelder et al. (2014) examined the association between maternal cannabis use from 1 month before pregnancy through the end of the third month of pregnancy and 20 selected anomalies (n = 13,859 case infants; n = 6,556 control infants). The authors reported an increased risk of the following anomalies: anencephaly (aOR, 2.2; 95% CI = 1.3–3.7), esophageal atresia (aOR, 1.4; 95% CI = 0.8–2.4), diaphragmatic hernia (aOR, 1.4; 95% CI = 0.9–2.2), and gastroschisis (aOR, 1.2; 95% CI = 0.9–1.7). Williams et al. (2004) obtained an (aOR, 1.90; 95% CI = 1.29–2.81) for the risk of isolated ventricular septal defect (VSD) among 122 isolated VSD cases and 3,029 control infants.

Discussion of Findings

The findings for birth weight are consistent with the effects of non-cannabinoid substances in smoked cannabis and cigarette smoking. It has been shown in several studies that the increases in carbon monoxide, with elevated carboxyhemoglobin blood levels, may be up to fivefold higher after marijuana than cigarettes (Wu et al., 1988). In other studies of marijuana exposure during pregnancy, the cause of the fetal growth

___________________

³ The authors found higher rates of association between cannabis use and the following birth defects: encephalocele, hydrocephaly, microcephaly, anotia/microtia, tetralogy of Fallot, ventricular septal defect, atrial septal defect, pulmonary valve atresia/stenosis, hypoplastic left heart syndrome, cleft palate alone, cleft lip with/without cleft palate, pyloric stenosis, anal/rectal/large-intestinal atresia/stenosis, obstructive genitourinary defect, polydactyly, syndactyly, reduction deformity of upper limbs, gastroschisis, and trisomy 21 (Forrester and Merz, 2006).

restriction noted was proposed to be fetal hypoxia due to the shift in the oxyhemoglobin curve caused by carbon monoxide (Frank et al., 1990).

NEONATAL CONDITIONS

Is There an Association Between Maternal Cannabis Use and Neonatal Conditions in the Infant?

Prematurity/Gestational Age

Systematic Reviews Gunn et al. (2016) documented a decrease in gestational age (measured in weeks) associated with cannabis use (pMD, −0.20; 95% CI = −0.62 to −0.22) and increased odds of the risk of preterm delivery (<37 completed weeks) (pOR, 1.29; 95% CI = 0.80–2.08).

Primary Literature Two other studies, Gray et al. (2010) and van Gelder et al. (2014), found no association between cannabis use and shortened gestation. For a total of 86 infants, Gray et al. (2010) reported a median estimated gestational age at delivery of 39 weeks (p = 0.685) both for infants who were exposed to cannabis and for infants who were not exposed to cannabis. van Gelder et al. (2014) found no association between cannabis use and gestational age after adjusting for gestational weight gain (aOR, 0.6; 95% CI = 0.1–2.4; n = 3 exposed; n = 335 nonexposed). The study was likely not to have the power to detect a difference.

Two studies, Dekker et al. (2012) and Leemaqz et al. (2016), reported an increased risk of spontaneous preterm birth associated with cannabis use (aOR, 2.34; 95% CI = 1.22–4.52 and aOR, 2.28; 95% CI = 1.49–3.60; p <0.001, respectively).

Neonatal Intensive Care Unit Admission

Systematic Reviews Gunn et al. (2016) reported increased risk of neonatal intensive care unit (NICU) admission for infants exposed to prenatal cannabis (pOR, 2.02; 95% CI = 1.27–3.21).

Primary Literature Warshak et al. (2015) also found an increased risk of NICU admission among infants born to 4,892 cannabis users and 153 nonusers (aOR, 1.54; 95% CI = 1.14–2.07).

Other Neonatal Conditions

Systematic Reviews Gunn et al. (2016) considered other neonatal conditions and found no association between maternal cannabis use and infant Apgar scores at 1 and 5 minutes. They did not find any differences for jaundice, resuscitation, respiratory distress syndrome, intubation following delivery, hypoglycemia, and sepsis. Studies were mixed as to whether infants exhibited abnormal behavior on neonatal behavioral assessments, in part because different assessment instruments were used in each study.

Primary Literature Warshak et al. (2015) did not find a statistically significant difference in the length of infant hospital stays (aOR, 1.12; 95% CI = 0.95–1.31). Gray et al. (2010) examined Apgar scores at 1 and 5 minutes and found no association between the scores and infant cannabis exposure (p = 0.709 and p = 0.496, respectively).

Discussion of Findings

The literature with regard to prematurity is mixed and needs further study. No neonatal outcomes appeared to be associated with cannabis exposure, but the studies are limited. Findings related to health care use, such as the increase in NICU admissions, need to be treated with caution. This pattern may reflect protocols requiring admission of all infants whose mothers have a history of substance use in pregnancy or failed toxicological screens during labor, rather than the health of the infant per se, particularly as there appears to be no increase in length of neonatal stay.

LATER OUTCOMES

Is There an Association Between Maternal Cannabis Use and Later Outcomes for the Offspring?

Systematic Reviews

The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis exposure and later outcomes for the child.

Primary Literature

As noted above in the introduction of this chapter, examination of later outcomes relied heavily on three cohorts, with some limited results from other hand-searched studies reported below.

The first of these cohorts was the Ottawa Prenatal Prospective Study (OPPS) by Fried and colleagues (Fried et al., 1998). The sample of 698 pregnant women was a convenience sample obtained through advertising in doctors’ offices and in the media. It could be characterized as including low-risk middle-class women of European descent. No gestational criterion was used, but most of the women were in their second trimester of pregnancy. Data collection was by interview about drug use while pregnant, including the use of cigarettes, alcohol, and cannabis, the last of which was characterized in terms of the number of joints per week. Of the original 698 study participants, 140 women reported at least some use of cannabis or drinking at least 0.85 oz. of absolute alcohol per day or smoking at least 16 mg of nicotine per day (Fried et al., 1998). A smaller group of women (n = 50) who did not use any substances during pregnancy were randomly selected as a reference group. Among these women, prenatal maternal cannabis use was categorized into three groups, with levels averaged across pregnancy: (1) no use, (2) mild/moderate use up to six joints per week, and (3) heavy use of at least six joints per week. Offspring were followed until the ages 18 to 22 years, with some attrition as would be expected (Fried et al., 1998).

The second study, started in 1982, was the Maternal Health Practices and Child Development (MHPCD) Study (Day and Richardson, 1991). The sample was recruited from a single inner-city outpatient prenatal clinic in Pittsburgh and thus was of mixed race/ethnicity and lower socioeconomic status. The participants had to be at least 18 years of age and in their fourth month of pregnancy. Of the 1,360 participants who met these criteria and were screened by an interview, pregnant women who used two or more joints per month were then selected for the study, with a random sample of an equal number of women chosen from the remaining non-using subjects, for a total sample of 564 (Huizink, 2014). Prenatal cannabis use was expressed as average daily joints for each trimester of pregnancy separately, although there was some overlap. Follow-up data on offspring have been reported up to the age of 14.

The most recent study was the Generation R study started in 2001, a multiethnic (Dutch, Moroccan, Surinamese, and Turkish) population-based prospective cohort study from fetal life until adulthood in the city of Rotterdam, the Netherlands (Jaddoe et al., 2012). The sample consists of 9,778 mothers with a delivery date between April 2002 and January 2006, and the members of the sample tended to be of higher socioeconomic status (Huizink, 2014). All participating women in Generation R

filled out questionnaires on their substance use at three points in pregnancy corresponding to the three trimesters. In this sample, 220 women reported using cannabis in pregnancy, generally in the first trimester (Huizink, 2014). The study discriminated between cannabis exposure, tobacco smoking, and the use of neither. Data on the resulting children up to age 6 were used in this report.

Sudden Infant Death Syndrome

Systematic Reviews The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis exposure and sudden infant death syndrome (SIDS).

Primary Literature Only one study was identified that examined the association between cannabis use and SIDS. In a case-control study of 428 infants who died of SIDS in southern California between 1989 and 1992, Klonoff-Cohen and Lam-Kruglic (2001) found no association between SIDS and cannabis exposure at conception (aOR, 1.1; 95% CI = 0.6–2.0; p = 0.82), during pregnancy (aOR, 0.6; 95% CI = 0.3–1.6; p = 0.33), or postnatally (aOR, 0.6; 95% CI = 0.2–1.8; p = 0.42). An interesting finding is increased risk of SIDS with paternal cannabis use at conception (aOR, 2.2; 95% CI = 1.2–4.2; p = 0.01), during pregnancy (aOR, 2.0; 95% CI = 1.0–4.1; p = 0.05), and postnatally (aOR, 2.8; 95% CI = 1.1–7.3; p = 0.04).

Breastfeeding

Systematic Reviews The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis exposure and breastfeeding.

Primary Literature One narrative review (Garry et al., 2009) identified two early studies on the effects of cannabinoids in breast milk on subsequent motor function but found no consistency in the results. The authors noted the difficulty in studying this issue since prenatal exposure is also likely among other confounders of cannabis use. The committee’s search identified one study of physical growth (Fried et al., 2001) which makes mention of no difference being found in choice and duration of breastfeeding relative to marijuana use.

Physical Growth

Systematic Reviews The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis exposure and physical growth in the child.

Primary Literature Postnatal growth results were obtained from the OPPS (Fried et al., 2001). Growth was measured for 152 participants at 1 year, 2–4 years, 6 years, 12 years, and 13–16 years. There was a dose–response relationship between head circumference and cannabis exposure (measured as heavy or six or more joints per week, moderate or between zero and six joints per week, and none), with children of heavy cannabis users having the smallest head circumferences (Z-score, 0.84; SD = 1.3; p = 0.08), a finding that persisted through age 12 but was not seen at ages 13–16 (Fried et al., 2001). In addition, infants of heavy cannabis users were the lightest at birth (Z-score, 0.32; SD = 0.9), but they experienced substantial weight gain such that they were the heaviest at 1 year. Furthermore, at ages 13–16 no differences were seen in height, weight, ponderal index, or onset of puberty.

Cognition/Academic Achievement

Systematic Reviews The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis exposure and cognition and academic achievement of the child.

Primary Literature The committee reviewed this literature in terms of preschool cognitive development and later cognitive development. Among the studies that examined cognitive development up to 3 years of age, no difference was found. In addition, two studies (OPPS and MHPCD) looked at cognitive development at 36–60 months. Both studies reported a weak effect on short-term memory.

Six studies out of two cohorts were identified that addressed the association between cannabis and cognitive function between ages 5 and 16 years using a variety of assessment instruments (Bluhm et al., 2006; El Marroun et al., 2010; Fried and Watkinson, 1988, 1990; Goldschmidt et al., 2012; Richardson et al., 1995). No differences in overall cognitive scores were found, but differences with exposure to different levels of prenatal cannabis were seen for some subscale scores, although they were not replicated across studies. In their assessment of school achievement, Goldschmidt et al. (2012) found worse reading scores at age 14 as measured by the Wechsler Individual Achievement Test (WIAT Screener). The

authors found a WIAT Screener basic reading score of 93.8 among nonexposed children, 93.1 among children exposed to less than one joint per day, and 87.8 among children exposed to one or more joints per day (p = 0.001).⁴ No differences with cannabis exposure were seen for cognitive or motor development in Fried and Watkinson (1988), Richardson et al. (1995), or El Marroun et al. (2010).

Behavior

Systematic Reviews The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis exposure and later child behavior.

Primary Literature The committee sought studies linking prenatal marijuana exposure to later child behavior. Of the three cohorts assessed above, only one report dealt with child behavior problems (Bluhm et al., 2006). The remaining reports assessed behavior in testing situations: for example, variability in reaction times and errors on continuous performance tests. Because the committee felt the latter do not really capture the construct of interest, this section reports only on child behavior problems at age 18 months and 3 years. At 18 months, higher aggression scores were seen in girls but not in boys; this effect did not persist at 3 years (El Marroun et al., 2010).

Substance Use and Delinquency

Systematic Reviews The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis exposure and later substance use and delinquency of the child.

Primary Literature The committee identified five reports from two cohorts (OPPS and MHPCD) that addressed the association between prenatal cannabis exposure and substance use and delinquency among offspring between 14 and 22 years of age. In the study addressing delinquency at age 14 years, prenatal cannabis exposure was found to be correlated with an increased risk of delinquent behavior (OR, 1.84; 95% CI = 1.05–2.96) (Day et al., 2015). However, this effect was mediated by depression and attention difficulties at age 10. Three studies addressed prenatal exposure to cannabis on the use of both cigarettes and cannabis in offspring ages 14 to 22 years. In Porath and Fried (2005), prenatal marijuana exposure more

___________________

⁴ This can be accounted for by attention and depression at age 10.

than doubled the risk of the initiation of cigarette smoking (OR, 2.58; 95% CI = 1.11–6.00) and daily cigarette smoking (OR, 2.36; 95% CI = 1.00–5.57). The authors also found that prenatal cannabis exposure also increased the risk of initiation of cannabis use in youth (OR, 2.76; 95% CI = 1.11–6.86) and increased the risk of using marijuana regularly (OR, 0.79; 95% CI = 0.33–1.90). Sonon et al. (2015) found that prenatal cannabis exposure was a predictor of offspring marijuana use (OR, 1.22; 95% CI = 1.02–1.44) at age 22 (Sonon et al., 2015).

Mental Health and Psychosis

Systematic Reviews The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis exposure and later mental health and psychosis in the child.

Primary Literature At age 10, children in the MHPCD study with prenatal cannabis exposure in the first and third trimesters had worse scores on a measure of depressive symptoms. Using the Avon Longitudinal Study of Parents and Children (ALSPAC) cohort study Zammit et al. (2009) found no difference in definite psychotic-like symptoms (PLIKS) as measured by a PLIKS semi-structured interview at 12 years of age between those exposed prenatally and those not exposed (aOR for linear trend, 0.91; 95% CI = 0.49–1.71; p = 0.776). Day et al. (2015), working with the MHPCD cohort at age 22, found that prenatal marijuana exposure was associated with an increased risk of psychotic symptoms as measured by the Diagnostic Interview Schedule (incidence density ratio [IDR] 1.31; p <0.05). In a mediation model, considering the effect of early initiation use of cannabis, the youth risk was essentially the same (IDR, 1.27; p = 0.06).

Discussion of Findings

The literature reviewed above does not support an effect of cannabis exposure on overall cognitive function, although some variation in subscale scores has been seen. Only one study has examined overall child behavior, and it found that the results did not persist. More consistency is seen for adolescent outcomes, with increased delinquency, greater cigarette and cannabis use, and some suggestion of increased mental health symptoms. For the later outcomes, attributing the outcomes to prenatal exposures is particularly difficult. While the studies attempted to control for the child’s environment using standard measures of socioeconomic status as well as a direct assessment of the home environment, these

approaches may be insufficient to detect potentially subtle differences in the family and neighborhood environments of women who smoke cannabis during pregnancy and those who do not. For example, the association of prenatal cannabis exposure and adolescent substance use may reflect family/neighborhood influences and may not be a direct effect of the prenatal exposure. Likewise, maternal distress/depression during pregnancy, which is likely to continue postpartum, may influence both the use of cannabis and child developmental outcomes. In addition, these studies did not address heritable or epigenetic vulnerability.

RESEARCH GAPS

To address the research gaps relevant to prenatal, perinatal, and neonatal outcomes, the committee suggests the following:

• There is a need for systematic inquiry using standardized questions about dose and duration at specific intervals in pregnancy to ascertain the level of prenatal cannabis exposure.
• Capitalizing, where possible, on the increase in toxicological screening at delivery to validate self-report measures.
• With the increased availability of recreational cannabis, observational studies need to be carried out—where ethical—on cannabis use and potential physiologic changes (e.g., blood pressure, etc.).
• Pooling, if possible, to obtain cohorts of women exposed only to THC and not to other drugs.
• A systematic follow-up of children exposed to cannabis prenatally with agreed-upon protocols and tests, with an ascertainment of the home and neighborhood environment regarding concurrent substance use.
• Developing strategies for assessing the effect of cannabis on pregnant women and fetuses through registries or systematic use of administrative data.

SUMMARY

This chapter summarizes the literature on prenatal, perinatal, and neonatal exposure to cannabis that has been published since 1999 and deemed to be of good or fair quality by the committee. Overall, there is substantial evidence of a statistical association between cannabis smoke and lower birth weight, but there is only limited, insufficient, or no evidence in support of any other health endpoint related to prenatal, perinatal, or neonatal outcomes. This may be due to a number of limitations faced by many of the research studies reviewed in this chapter, including an almost exclusive reliance on self-reporting to ascertain cannabis exposure, as is true in many areas of this report. While many studies used standardized questions regarding frequency and duration of cannabis use, others relied on data extracted from the medical record. Also, as with other portions of this report, the potency of cannabis varied across time. The lack of biological validation of self-reporting suggests caution is warranted. Moreover, dosage and timing of exposure in pregnancy is particularly important, as exposures early in pregnancy may affect organogenesis leading to birth defects, whereas later exposures are more likely to affect the growth of the fetus.

Second, even within substantial cohorts, the number of women who used cannabis exclusively was small. These sample sizes may have limited statistical power to detect many outcomes.

Third, cannabis exposure was almost exclusively through smoking and was often confounded by the use of other substances—namely, tobacco and alcohol. Although many authors relied on a variety of statistical techniques to isolate the effects of cannabis exposure, attribution of outcomes to cannabis alone was difficult. Even when cannabis is the sole exposure, it is not straightforward to attribute outcomes to THC alone versus the mode of exposure.

Finally, caution needs to be used in interpreting the numerous findings of “no association” in this chapter. Absent a pooled estimate of effect and confidence intervals, such conclusions may be based on a small number of studies, some of which may even conflict.

The committee has formed a number of research conclusions related to these health endpoints (see Box 10-1); however, it is important that each of these conclusions be interpreted within the context of the limitations discussed in the Discussion of Findings sections.

REFERENCES

Day, N. L., and G. A. Richardson. 1991. Prenatal marijuana use: Epidemiology, methodologic issues, and infant outcome. Chemical Dependency and Pregnancy 18(1):77–91.

Day, N. L., L. Goldschmidt, R. Day, C. Larkby, and G. A. Richardson. 2015. Prenatal marijuana exposure, age of marijuana initiation, and the development of psychotic symptoms in young adults. Psychological Medicine 45(8):1779–1787.

Dekker, G. A., S. Y. Lee, R. A. North, L. M. McCowan, N. A. Simpson, and C. T. Roberts. 2012. Risk factors for preterm birth in an international prospective cohort of nulliparous women. PLOS ONE 7(7):e39154.

El Marroun, H., H. Tiemeier, E. A. P. Steegers, J. W. Roos-Hesselink, V. W. V. Jaddoe, A. Hofman, F. C. Verhulst, W. van den Brink, and A. C. Huizink. 2010. A prospective study on intrauterine cannabis exposure and fetal blood flow. Early Human Development 86(4):231–236.

Fergusson, D. M., L. J. Horwood, and K. Northstone. 2002. Maternal use of cannabis and pregnancy outcome. British Journal of Obstetrics and Gynaecology 109(1):21–27.

Forray, A., B. Merry, H. Lin, J. P. Ruger, and K. A. Yonkers. 2015. Perinatal substance use: A prospective evaluation of abstinence and relapse. Drug and Alcohol Dependence 150: 147–155.

Forrester, M., and R. Merz. 2006. Comparison of trends in gastroschisis and prenatal illicit drug use rates. Journal of Toxicology and Environmental Health, Part A: Current Issues 69(13):1253–1259.

Frank, D. A., H. Bauchner, S. Parker, A. M. Huber, K.-A. Kwabena, H. Cabral, and B. Zuckerman. 1990. Neonatal body proportionality and body composition after in utero exposure to cocaine and marijuana. Journal of Pediatrics 117(4):622–626.

Fried, P. A., and B. Watkinson. 1988. 12- and 23-month neurobehavioural follow-up of children prenatally exposed to marihuana, cigarettes and alcohol. Neurotoxicology and Teratology 10:305–313.

Fried, P. A., and B. Watkinson. 1990. 36- and 48-month neurobehavioral follow-up of children prenatally exposed to marijuana, cigarettes, and alcohol. Developmental and Behavioral Pediatrics 11(2):49–58.

Fried, P. A., B. Watkinson, and R. Gray. 1998. Differential effects on cognitive functioning in 9- to 12-year olds prenatally exposed to cigarettes and marihuana. Neurotoxicology and Teratology 20(3):293–306.

Fried, P. A., D. S. James, and B. Watkinson. 2001. Growth and pubertal milestones during adolescence in offspring prenatally exposed to cigarettes and marihuana. Neurotoxicology & Teratology 23(5):431–436.

Fryers, T., and T. Brugha. 2013. Childhood determinants of adult psychiatric disorder. Clinical Practice and Epidemiology in Mental Health 9:1–50.

Garry, A., V. Rigourd, A. Amirouche, V. Faurox, S. Aubry, and R. Serreau. 2009. Cannabis and breastfeeding. Journal of Toxicology 2009(596149):1–5.

Goldschmidt, L., G. A. Richardson, J. A. Willford, S. G. Severtson, and N. L. Day. 2012. School achievement in 14-year-old youths prenatally exposed to marijuana. Neurotoxicology & Teratology 34(1):161–167.

Gray, T. R., R. D. Eiden, K. E. Leonard, G. J. Connors, S. Shisler, and M. A. Huestis. 2010. Identifying prenatal cannabis exposure and effects of concurrent tobacco exposure on neonatal growth. Clinical Chemistry 56(9):1442–1450.

Gunn, J. K. L., C. B. Rosales, K. E. Center, A. Nunez, S. J. Gibson, C. Christ, and J. E. Ehiri. 2016. Prenatal exposure to cannabis and maternal and child health outcomes: A systematic review and meta-analysis. BMJ Open 6(4):e009986.

Hashibe, M., K. Straif, D. P. Tashkin, H. Morgenstern, S. Greenland, and Z. Zhang. 2005. Epidemiologic review of marijuana use and cancer risk. Alcohol 35:265–275.

Huang, Y. J., Z. Zhang, D. P. Tashkin, B. Fend, K. Straif, and M. Hashibe. 2015. An epidemiologic review of marijuana and cancer: An update. Cancer Epidemiology, Biomarkers, and Prevention 24(1):15–31.

Huizink, A. 2014. Prenatal cannabis exposure and infant outcomes: Overview of studies. Progress in Neuro-Psychopharmacology and Biological Psychiatry 52:45–52.

IOM (Institute of Medicine). 1999. Marijuana and medicine: Assessing the science base. Washington, DC: National Academy Press.

Irner, T. B. 2012. Substance exposure in utero and developmental consequences in adolescence: A systematic review. Child Neuropsychology 18(6):521–549.

Jaddoe, V. W. V., C. M. van Duijn, O. H. Franco, A. K. van der Heijden, M. H. van IIzendoorn, J. C. de Jongste, A. van der Lugt, J. P. Mackenbach, H. A. Moll, H. Raat, F. Rivadeneira, E. A. P. Steegers, H. Tiemier, A. G. Uitterlinder, F. C. Verhulst, and A. Hofman. 2012. The Generation R study: Design and cohort update 2012. European Journal of Epidemiology 27(9):739–756.

Janisse, J. J., B. A. Bailey, J. Ager, and R. J. Sokol. 2014. Alcohol, tobacco, cocaine, and marijuana use: Relative contributions to preterm delivery and fetal growth restriction. Substance Abuse 35(1):60–67.

Klonoff-Cohen, H., and P. Lam-Kruglic. 2001. Maternal and paternal recreational drug use and sudden infant death syndrome. Pediatrics and Adolescent Medicine 155(7):765–770.

Leemaqz, S. Y., G. A. Dekker, L. M. McCowan, L. C. Kenny, J. E. Myers, N. A. Simpson, L. Poston, and C. T. Roberts. 2016. Maternal marijuana use has independent effects on risk for spontaneous preterm birth but not other common late pregnancy complications. Reproductive Toxicology 62:77–86.

Lubman, D., A. Cheetham, and M. Yucel. 2014. Cannabis and adolescent brain development. Pharmacology and Therapeutics (148):1–16.

Metz, T. D., and E. H. Stickrath. 2015. Marijuana use in pregnancy and lactation: A review of the evidence. American Journal of Obstetrics & Gynecology 213(6):761–778.

Porath, A. J., and P. A. Fried. 2005. Effects of prenatal cigarette and marijuana exposure on drug use among offspring. Neurotoxicology & Teratology 27(2):267–277.

Richardson, G. A., N. L. Day, and L. Goldschmidt. 1995. Prenatal alcohol, marijuana, and tobacco use: Infant mental and motor development. Neurotoxicology and Teratology 17(4):479–487.

Savitz, D. A., and P. Murnane. 2010. Behavioral influences on preterm birth: A review. Epidemiology 21(3):291–299.

Schempf, A. H. 2007. Illicit drug use and neonatal outcomes: A critical review. Obstetrical & Gynecological Survey 62(11):749–757.

Schempf, A. H., and D. M. Strobino. 2008. Illicit drug use and adverse birth outcomes: Is it drugs or context? Journal of Urban Health 85(6):858–873.

Sonon, K. E., G. A. Richardson, J. R. Cornelius, K. H. Kim, and N. L. Day. 2015. Prenatal marijuana exposure predicts marijuana use in young adulthood. Neurotoxicology and Teratology 47:10–15.

van Gelder, M. M., A. R. Donders, O. Devine, N. Roeleveld, J. Reefhuis, and the National Birth Defects Prevention Study. 2014. Using Bayesian models to assess the effects of under-reporting of cannabis use on the association with birth defects, National Birth Defects Prevention Study, 1997–2005. Paediatric and Perinatal Epidemiology 28(5):424–433.

Varner, M. W., R. M. Silver, C. J. Rowland Hogue, M. Willinger, C. B. Parker, V. R. Thorsten, R. L. Goldenberg, G. R. Saade, D. J. Dudley, D. Coustan, B. Stoll, R. Bukowski, M. A. Koch, D. Conway, H. Pinar, U. M. Reddy, and the Eunice Kennedy Shriver National Institute of Child Health and Human Development Stillbirth Collaborative Research Network. 2014. Association between stillbirth and illicit drug use and smoking during pregnancy. Obstetrics & Gynecology 123(1):113–125.

Viteri, O. A., E. E. Soto, R. O. Bahado-Singh, C. W. Christensen, S. P. Chauhan, and B. M. Sibai. 2015. Fetal anomalies and long-term effects associated with substance abuse in pregnancy: A literature review. American Journal of Perinatology 32(5):405–415.

Warshak, C. R., J. Regan, B. Moore, K. Magner, S. Kritzer, and J. Van Hook. 2015. Association between marijuana use and adverse obstetrical and neonatal outcomes. Journal of Perinatology 35(12):991–995.

Williams, J. H., and L. Ross. 2007. Consequences of prenatal toxin exposure for mental health in children and adolescents: A systematic review. European Child & Adolescent Psychiatry 16(4):243–253.

Williams, L. J., A. Correa, and S. Rasmussen. 2004. Maternal lifestyle factors and risk for ventricular septal defects. Birth Defects Research 70(2):59–64.

Witter, F. R. and J.R. Niebyl. 1990. Marijuana use in pregnancy and pregnancy outcome. American Journal of Perinatology 7(1):36–38.

Wu, T. C., D. P. Tashkin, B. Djahed, and J. E. Rose. 1988. Pulmonary hazards of smoking marijuana as compared with tobacco. New England Journal of Medicine 318(6):347–351.

Zammit, S., K. Thomas, A. Thompson, J. Horwood, P. Menezes, D. Gunnell, C. Hollis, D. Wolke, G. Lewis, and G. Harrison. 2009. Maternal tobacco, cannabis and alcohol use during pregnancy and risk of adolescent psychotic symptoms in offspring. British Journal of Psychiatry 195(4):294–300.

This page intentionally left blank.