The health and psychological consequences of cannabis use - chapter 6

THIS DOCUMENT HAS BEEN RESCINDED: Chapter 6.5 Reproductive effects of cannabis.

Page last updated: 1994

Please note the information in this publication may no longer be current but is retained on our website for historical or research purposes.

<< Previous Chapter | Contents | Next Chapter>>

6. The chronic effects of cannabis use on health

6.5 Reproductive effects of cannabis

In the mid-1970s there seemed to be good reason to suspect that cannabis use had adverse effects on the human reproductive system. There was some animal experimentation which suggested that cannabis adversely affected the secretion of gonadal hormones in both sexes, and the foetal development of animals administered crude marijuana extract or THC during pregnancy (Bloch, 1983; Institute of Medicine, 1982; Nahas, 1984; Nahas and Frick, 1987; Wenger et al, 1992). Cannabis was being widely used by adolescents who were undergoing sexual maturation, and by young adults who were entering the peak age for reproduction (Linn et al, 1983). The suspicion that cannabinoids had adverse effects on the human reproductive system was first raised by case reports of breast development (gynecomastia) in young men aged 23 to 26 years of age, all of whom had a history of heavy cannabis use (Harman and Aliapoulios, 1972). The suspicion seemed confirmed by human observations published shortly after by Kolodny et al (1974), who reported that males who were chronic cannabis users had reduced plasma testosterone, reduced sperm count and motility, and an increased prevalence of abnormal sperm.

In the light of these observations, the widespread use of cannabis among young adults which began in the early 1970s and continued well into the mid-1980s raised understandable fears that fertility would be impaired in men, and the rates of unfavourable pregnancy outcomes would increase among women using cannabis during in their reproductive years. These outcomes could possibly include greater foetal loss, lower birth weight, and an increased risk of birth defects and perinatal deaths. Later, concerns were also raised about the possibility of adverse effects upon the subsequent behavioural development and health of children exposed to marijuana in utero. Evidence relevant to each of these concerns will be reviewed in this section.

6.5.1 Effects on the male reproductive system

In animals, marijuana, crude marijuana extracts, THC and certain other purified cannabinoids have been shown to "depress the functioning of the male reproductive endocrine system" (Bloch, 1983, p355). If used chronically, cannabis reduces plasma testosterone levels, retards sperm maturation, reducing the sperm count and sperm motility, and increasing the rate of abnormal sperm (Bloch, 1983, National Academy of Science, 1982; Wenger et al, 1992). Although the mechanisms by which cannabis produces these effects are uncertain, it is likely that they occur both directly as a result of action of THC on the testis, and indirectly via effects on the hypothalamic secretion of the hormones that stimulate the testis to produce testosterone (Wenger et al, 1992).

The small number of human studies on the effects of cannabis on male reproductive function have produced mixed results. The findings of the early study by Kolodny et al (1974) which reported reduced testosterone, sperm production, and sperm motility and increased abnormalities in sperm were contradicted shortly thereafter by the results of a larger, well controlled study of chronic heavy users, which failed to find any difference in plasma testosterone at study entry, or after three weeks of heavy daily cannabis use (Mendelson et al, 1974). Other studies have produced both positive and negative evidence of an effect of cannabinoids on testosterone, for reasons that are not well understood (Institute of Medicine, 1982). Hollister (1986) has conjectured that reductions in testosterone and spermatogenesis probably require long-term exposure. Even if there are such effects of cannabis on male reproductive functioning, their clinical significance in humans is uncertain (Institute of Medicine, 1982) since testosterone levels in the studies which have found effects have generally remained within the normal range (Hollister, 1986).

The putative relationship between cannabis use and gynecomastia now seems very doubtful. The magnitude of reductions observed in the positive studies are too small to explain the case reports of gynecomastia among heavy male cannabis smokers (Harman and Aliapoulios, 1972), and a small case-control study failed to find any relationship between cannabis use and gynecomastia in 11 cases and controls (Cates and Pope, 1977).

The chronic effects of this study did not exclude a four-fold higher risk of gynecomastia among cannabis smokers, studies in humans and animals have not shown any increased secretion of the hormone prolactin, the most likely mechanism of such effects in males. As Mendelson et al (1984) have argued, if chronic cannabis use caused gynecomastia, one would expect many more cases to have been reported in the clinical literature, given the widespread use of cannabis among young males during the past few decades.

Hollister has argued that the reductions in testosterone and spermatogenesis observed in the positive studies are probably of "little consequence in adults", although he conceded that they could be of "major importance in the prepubertal male who may use cannabis" (p10). He cited a case of growth arrest in a 16-year-old male who began heavy cannabis use at the age of 11, and who experienced a retardation of growth and the development of secondary sexual characteristics which partially remitted after three months abstinence from cannabis (Copeland, Underwood and Van Wyck, 1980). The possible effects of cannabis use on testosterone and spermatogenesis may therefore be most relevant to males whose fertility is already impaired for other reasons, e.g. a low sperm count.

Top of page

6.5.2 Effects on the female reproductive system

The experimental animal studies suggests that cannabis use has similar effects on female reproductive system to those found in males. The acute effects of cannabis or THC exposure in the non-pregnant female animal is to transiently interfere with the hypothalamic-pituitary-gonadal axis (Bloch, 1983). Chronic cannabis exposure delays oestrous and ovulation by reducing leutinising hormone and increasing prolactin secretion.

There have been very few human studies of the effects of cannabis on the female reproductive system because of fears that cannabis use may produce teratogenic and genotoxic effects in women of childbearing age who would be the experimental subjects in such studies (Rosenkrantz, 1985). Two studies have been reported with conflicting results. In an unpublished study, Bauman (1980 cited by Nahas, 1984) compared the menstrual cycles of 26 cannabis smokers with those of 17 controls, and found a higher rate of anovulatory cycles among the cannabis users. Mendelson and Mello (1984) observed hormonal levels in a group of female cannabis users (all of whom had undergone a tubal ligation) under controlled laboratory conditions. They failed to find any evidence that sub-chronic cannabis use affected the cycling of the sex hormones, or the duration of the cycle. In the absence of any other human evidence, both Bloch (1983) and the Institute of Medicine (1982) argued on the basis of the animal literature that cannabis use probably had an inhibitory effect on human female reproductive function which was similar to that which occurs in males.

6.5.3 Foetal development and birth defects

Given evidence that THC affects female reproductive function, one might expect it to have a potentially adverse effect on the outcome of pregnancy. The possibility of adverse pregnancy outcomes is increased by evidence that THC crosses the placenta in animals (Bloch, 1983) and humans (Blackard and Tennes, 1984). This raises the possibility that THC, and possibly other cannabinoids, are teratogens, i.e. substances that may interfere with the normal development of the foetus in utero.

The animal evidence indicates that in sufficient dosage cannabis can "produce resorption, growth retardation, and malformations" in mice, rats, rabbits, and hamsters (Bloch, 1983, p406). Growth esorption and growth retardation have been more consistently reported than birth malformations (Abel, 1985). There are also several caveats on the evidence that cannabis increases rates of malformations. The doses required to reliably produce malformations have been very high (Abel, 1985), and such effects have been observed more often after the administration of crude marijuana extract than pure THC, suggesting that other cannabinoids may be involved in producing any teratogenic effects. There have also been doubts expressed about whether these teratogenic effects can be directly attributed to THC. Some have argued, for example, that the malformations may be a consequence of reduced nutrition caused by the aversive properties of the large doses of cannabis used in these studies (Abel, 1985; Bloch, 1983).

Hollister (1986) has also discounted the animal research data, arguing that "virtually every drug that has ever been studied for dysmorphogenic effects has been found to have them if the doses are high enough, if enough species are tested, or if treatment is prolonged" (p4). Similar views have been expressed by Abel (1985) and by Bloch (1983), who concluded that THC was unlikely to be teratogenic in humans because "the few reports of teratogenicity in rodents and rabbits indicate that cannabinoids are, at most, weakly teratogenic in these species" (p416). Human studies

The findings from the small number of epidemiological studies of the effects of cannabis use on human foetal development have been mixed for a number of reasons. First, both the adverse reproductive outcomes and the prevalence of heavy cannabis use during pregnancy are relatively rare events. Hence, unless cannabis use produces a large increase in the risk of abnormalities, very large sample sizes will be required to detect adverse effects of cannabis use on foetal development. Many of the studies that have been conducted to date have been too small to detect effects of this size (e.g. Greenland et al, 1982 a,b; Fried, 1980).

There are also likely to be difficulties in identifying cannabis users among pregnant women. The stigma associated with illicit drug use, especially during pregnancy, may discourage honest reporting, compounding the usual problem of women accurately recalling drug use in early pregnancy, when they are asked about it late in their pregnancy, or after the birth (Day et al, 1985). If, as seems likely, a substantial proportion of cannabis users are misclassified as non-users, any relationship between cannabis use and adverse outcomes will be attenuated, requiring even larger samples to detect it (Zuckerman, 1985).

Even when large sample sizes have been obtained, there are difficulties in interpreting any associations found between adverse pregnancy outcomes and cannabis use. Cannabis users are more likely to use tobacco, alcohol and other illicit drugs during their pregnancy. They also differ from non-users in social class, education, nutrition, and other factors which predict an increased risk of experiencing an adverse outcome of pregnancy (Fried, 1980, 1982; National Academy of Science, 1982; Tennes et al, 1985). These sources of confounding make it difficult to unequivocally attribute any relationship between reproductive outcomes and cannabis use to cannabis use per se, rather than to other drug use, or other variables correlated with cannabis use, such as poor maternal nutrition, and lack of prenatal care. Sophisticated forms of statistical control provide the only way of assessing to what degree this may be the case, but its application is limited by the small number of cannabis smokers identified in most studies.

Given these difficulties, and the marked variation between studies in the proportion of women who report cannabis use during pregnancy, the degree of agreement between the small number of studies is more impressive than the disagreement that seems at first sight to such be a feature of this literature. There is reasonable consistency (although not unanimity) in the finding that cannabis use in pregnancy is associated with foetal growth retardation, as shown by reduced birth weight (e.g. Gibson et al, 1983; Hatch and Bracken, 1986; Zuckerman et al, 1989), and length at birth (Tennes et al, 1985). This relationship has been found in the best controlled studies, and it has persisted after statistically controlling for potential confounding variables by sophisticated forms of statistical analysis (e.g. Hatch and Bracken, 1986; Zuckerman et al, 1989).

Uncertainty remains about the interpretation of this finding. Is it because the "marijuana products were toxic to foetal development", as argued by Nahas and Latour (1992)? Is it because THC interferes with the hormonal control of pregnancy shortening the gestation period, as has been reported by Gibson et al (1983) and Zuckerman et al (1989)? The fact that the lower birth weight among the children of women who used cannabis disappears after controlling for gestation length is supportive of the latter hypothesis. Is it because cannabis is primarily smoked, since tobacco smoking has been consistently shown to be associated with reduced birth weight (Fried, 1993)?

The findings on the relationship between cannabis use and birth abnormalities are more mixed, and conclusions accordingly less certain. Early case reports of children with features akin to the Foetal Alcohol Syndrome born to women who had smoked cannabis but not used alcohol during pregnancy (e.g. Milman, 1982, p42) suggested that cannabis may increase the risk of birth defects. Subsequent controlled studies have produced mixed results. Four studies have reported no increased rate of major congenital abnormalities among children born to women who use cannabis (Gibson et al, 1983; Hingson et al, 1982; Tennes et al, 1985; Zuckerman et al, 1989).

One study has reported a five-fold increased risk of children with foetal alcohol like features being born to women who reported using cannabis (Hingson et al, 1982). The significance of this finding is uncertain because the same study also found no relationship between self-reported alcohol use and "foetal alcohol syndrome" features. This is doubly surprising because of other evidence on the adverse effects of alcohol, and because the epidemiological data indicates that cannabis and alcohol use are associated (Norton and Colliver, 1988). An additional study reported an increase in the crude rate of birth abnormalities among children born to women who reported using cannabis. This result was no longer statistically significant after adjustment for confounders (Linn et al, 1983), although the confidence interval around this adjusted risk (OR=1.36) only narrowly included the null value (95 per cent CI: 0.97, 1.91).

The study by Zuckerman et al provides the most convincing failure to find an increased risk of birth defects among women who used cannabis during pregnancy. A large sample of women was obtained, among which there was a substantial prevalence of cannabis use that was verified by urinalysis. There was a low rate of birth abnormalities among the cannabis users, and no suggestion of an increase by comparison with the controls. On this finding, one might be tempted to attribute the small increased risk in the positive study (Linn et al, 1983) to recall bias, since the report of cannabis use during pregnancy was obtained retrospectively after birth, when women who had given birth to children with malformations may have been more likely to recall cannabis use than those who did not. However, given the uncertainty about the validity of self-reported cannabis use in many of the null studies, it would be unwise to exonerate cannabis as a cause of birth defects until larger, better controlled studies have been conducted.

Top of page

6.5.4 Chromosomal abnormalities and genetic effects

Teratogenesis - interference with normal foetal development - is not the only way in which cannabis use might adversely affect human reproduction. Cannabis use could conceivably produce chromosomal abnormalities or genetic change in either parent which could be transmitted to their progeny. Although possible, there is no animal or human evidence that such events occur. The experimental evidence indicates that "in vivo and in vitro exposure to purified cannabinoids or cannabis resin failed to increase the frequency of chromosomal damage or mutagenesis" (Bloch, 1983, p412). Marijuana smoke exposure, by contrast, "has been ... associated with chromosomal aberrations ... [such as] hypoploidy, mutagenicity in the Ames test ... " (Bloch, 1983, p413). The latter fact is more relevant to an appraisal of the risk of cannabis users developing cancers from exposure to cannabis smoke rather than to the risks of transmissible genetic defects in their offspring.

Hollister (1986) discounted the evidence from cytogenetic studies that cannabinoids may be mutagenic, as did the Institute of Medicine (1982). He also argued that assessing chromosomal damage was "more of an art than a science", as indicated by poor inter-observer agreement, and that the clinical significance remained unclear because "similar types and degrees of chromosomal changes have been reported in association with other drugs commonly used in medical practice without any clinical evidence of harm ..." (p4). Hollister concluded that "even if a small increase in chromosomal abnormalities is produced by cannabis, the clinical significance is doubtful" (p4).

6.5.5 Post-natal development

A further possibility which needs to be considered is that cannabis use by the mother during pregnancy and breast feeding may affect the post-natal development of the child. This could occur either because of the enduring effects of developmental impairment arising from in utero exposure, or because the infant continued to be exposed to cannabinoids via breast milk. These are not well investigated possibilities, although there are a small number of animal studies which provide suggestive evidence of such effects (Nahas, 1984; Nahas and Frick, 1987).

The most extensive research evidence in humans comes from the Ottawa Prospective Prenatal Study (OPPS), which studied developmental and behavioural abnormalities in children born to women who reported using cannabis during pregnancy (Fried and colleagues, 1980, 1982, 1983, 1985, 1986, 1989, 1990, 1992). In this study, mothers were assessed about their drug use during pregnancy and their children were measured on the Brazelton scales after birth, neurologically assessed at one month, and assessed again by standardised scales of ability at six and 12 months. The results indicated that there was some developmental delay shortly after birth in the infants' visual system, and there was also an increased rate of tremors and startle among the children of cannabis users.

The behavioural effects discernible after birth had faded by one month, and no effects were detectable in performance on standardised ability tests at six and 12 months. Effects were subsequently reported at 36 and 48-month follow-ups (Fried and Watkinson, 1990) but these did not persist in a more recent follow-up at 60 and 72 months (Fried, O'Connell, and Watkinson, 1992). These results are suggestive of a transient developmental impairment occurring among children who had experienced a shorter gestation and prematurity. There is a possibility that the tests used in later follow-ups are insufficiently sensitive to the subtle effects of prenatal cannabis exposure, although they were able to detect effects of maternal tobacco smoking during pregnancy on behavioural development at 60 and 72 months (Fried and Watkins, 1990, 1992).

Attempts to replicate the OPPS findings have been mixed. Tennes et al (1985) conducted a prospective study of the relationship between cannabis use during pregnancy and postnatal development in 756 women, a third of whom reported using cannabis during pregnancy. The children were assessed shortly after birth using the same measurement instruments as Fried (1980), and a subset were followed up and assessed at one year of age. The findings failed to detect any differences in behavioural development between the children of users and non-users after birth; i.e. there was no evidence of impaired development of the visual system, and no increased risk of tremor or startle among the children of users. There was also no evidence of any differences at one year. More recently, Day et al (in press), have followed up children at age three born to 655 women who were questioned about their substance use during pregnancy. They found a relationship between the mothers' cannabis use during pregnancy and the children's performances on memory and verbal scales of the Stanford-Binet Intelligence Scale.

There is suggestive evidence that cannabis use during pregnancy may have a more serious and life threatening effect on post-natal development. This emerged from a case-control study of Acute Nonlymphoblastic Leukemia (ANLL), a rare form of childhood cancer (Neglia et al, 1991; Robinson et al, 1989). The study was not designed as a test of relationship between cannabis use and ANLL; it was designed to examine the possible aetiological role of maternal and paternal environmental exposures to petrochemicals, pesticides and radiation. Maternal drug use, including marijuana use before and during pregnancy, were assessed as possible covariates to be statistically controlled in any relationships observed between ANLL and environmental exposures.

An unexpected but strong association was observed between maternal cannabis use and ANLL. The mothers of cases were 11 times more likely to have used cannabis before and during their pregnancy than were the mothers of controls. The relationship persisted after statistical adjustment for many other risk factors. Comparisons of cases whose mothers did and did not use cannabis during their pregnancies showed that cases with cannabis exposure were younger, and had a higher frequency of ANLL with cell types of a specific pathological origin than did the cases without such exposure. The authors argued that these differences made it unlikely that the relationship was due to chance.

Reporting bias on the part of the mothers of cases is an alternative explanation of the finding that is harder to discount. The reports of cannabis use were obtained retrospectively after diagnosis of the ANLL, so it is possible that the mothers of children who developed ANLL were more likely to seek an explanation in something they did during their pregnancies, and hence, may have been more likely to report cannabis use than were mothers of controls. The authors investigated this possibility by comparing the rates of cannabis use reported in this study with the rates reported in several earlier case-control studies of other childhood cancers that they had conducted using the same methods. The rate was lower among controls in the ANLL study, but even when the rate of cannabis use among the controls in these other studies was used the odds ratio was still greater than three and statistically significant. Nonetheless, since this was an unexpected finding which emerged from a large number of exploratory analyses conducted in a single study, it should be replicated as a matter of some urgency.

6.5.6 Conclusions

On the balance of probabilities, high doses of THC probably disrupt the male and female reproductive systems in animals by interfering with hypothalamo-pituitary-gonadal system, reducing secretion of testosterone, and hence reducing sperm production, motility, and viability in males, and interfering with the ovulatory cycle in females. It is uncertain whether these effects also occur in humans, given the dose differences, the inconsistency in the literature on human males and the absence of research on human females. Even if cannabinoids have such effects in humans, their clinical significance in normal healthy young adults is unclear. They may be of greater concern among young adolescents who are now more likely to use, and among males with fertility impaired for other reasons.

Cannabis use during pregnancy probably impairs foetal development, leading to smaller birthweight, perhaps as a consequence of a shorter period of gestation. It is possible although far from certain that cannabis use during pregnancy produces a small increase in the risk of birth defects as a result of exposure of the foetus in utero. Prudence suggests that until this issue is resolved, we should err in the conservative direction by recommending that women not use cannabis during pregnancy, or when attempting to conceive (Hollister, 1986).

There is not a great deal of evidence that cannabis use can produce chromosomal or genetic abnormalities in either parent which could be transmitted to offspring. The available animal and in vitro evidence suggests that the mutagenic properties of cannabis smoke are greater than those of THC, and are probably of greater relevance to the risk of users developing cancer than to the transmission of genetic defects to children. There is suggestive evidence that infants exposed in utero to cannabis may experience transient behavioural and developmental effects during the first few months after birth. There is also a single study which raises concern about an increased risk of childhood leukemia occurring among the children born to women who used cannabis during their pregnancies.