Marijuana & Pregnancy
(American College of Obstetricians & Gynaecologists)
“Medical marijuana is not safer than recreational marijuana.”
ACOG_2020_Poster_Marijuana_and_Pregnancy.pdf
Cannabis in Pregnancy – Rejoinder, Exposition and Cautionary Tales
October 10, 2020 Albert Stuart Reece, MBBS, FRCS, FRCS, FRACGP, MD , Gary Kenneth Hulse, BBSc, MBSc, PhD
Cannabis use during pregnancy is associated with a host of negative outcomes.
The recent paper by Stanciu discussing cannabis use in pregnancy1 makes several useful and highly salient points. With a more complete understanding of the published literature further important patterns in the data emerge. They aid our understanding of the pathobiology of in utero cannabis exposure and thereby powerfully inform the community on the most appropriate manner in which to regulate cannabis and cannabinoids from an improved evidence base.
It is well known that cannabis use has been liberalized across the United States as a result of well-financed and orchestrated campaigns.2 Stanciu is correct that most epidemiological studies point towards harmful associations, that cannabis use in pregnancy is becoming more common, that it is widely recommended in pregnancy by cannabis dispensaries, and that increased rates of low birth weight, premature and stillbirths, and increased neonatal intensive care admission are well recognized associations. It is correct that all 4 longitudinal studies of children born after prenatal cannabis exposure (PCE) show increased adverse neurodevelopmental outcomes including impaired executive function, visuomotor processing deficits, heightened startle responses, impulse control, heightened susceptibility to addiction in later life, emotional behaviors, and motor defects.3-5 Well-documented impacts on the glutamatergic, GABAergic and dopaminergic signaling in the brain are of concern as they represents major neurotransmitters in the central nervous system [CNShttps://pubmed.ncbi.nlm.nih.gov/17162495/“>A large Hawaiian study found an increased incidence of microcephaly (R.R. = 12.80, 95%C.I. 4.13-36.17)8 and the CDC have twice reported elevated rates of anencephalus (adjusted O.R. 1.7, C.I. 0.9-3.4) and (posterior O.R. 1.9 (C.I. 1.1, 3.2).9,10 This sets up a clear spectrum of severity from mild neurodevelopmental impairment, to microcephaly, to anencephalus and then fetal death. In the context of dose-response relationships and strong geotemporospatial associations issues of causality necessarily arise.
Stanciu’s observation that preclinical studies in experimental animals are important to understand the likely effects of PCE in individuals, not least due to the problem of the frequent exposure to multiple substances clinically, is also correct. This issue was studied in detail long ago in the 1960s and 1970s, and succinctly summarized by Graham’s telling observation: “oedema, phocomelia, omphalocoele, spina bifida, exencephaly, multiple malformations including myelocoele. This is a formidable list.”11
However, a reasonable question might be: “Why don’t we see such a broad teratological spectrum clinically?”
Stanciu’s remark that there are “no overt birth defects” is an oft-repeated myth and is in error, as well as obviously being at odds with several preclinical studies, especially in the most predictive species for human teratology (ie, hamsters and white rabbits).12,13
A recent paper from the Centers for Disease Control (CDC) noted that 4 defects, anencephalus, gastroschisis, diaphragmatic hernia and esophageal atresia were more common following PCE.9 The American Academy of Pediatrics (AAP) and the American Heart Association (AHA) issued a joint position statement that both ventricular septal defect (VSD) and Ebsteins anomaly were also elevated by PCE.14
The review of 17 years of birth defects from Hawaii found 21 defects to be elevated after PCE and featured prominently cardiovascular defects (atrial septal defect (ASD), VSD, hypoplastic left heart syndrome, tetralogy of Fallot (ToF) and pulmonary valve atresia or stenosis), chromosomal defects such as Downs syndrome, body wall defects such as gastroschisis, limb defects including syndactyly and upper limb reduction defects, facial, bowel and genitourinary system defects with calculated rate ratios ranging from 5.26 (C.I. 1.08-15.46) to 39.98 (C.I. 9.03-122.29).8
In September and October 2018 Colorado released 2 datasets of congenital anomalies across the period of its cannabis legalization program from 2000 to 2013 and 2000 to 2014 and reported 87,772 and 64,463 major defects respectively (which are obviously contradictory).15 Based on 4830 and 4026 major anomalies in the year 2000 this represents a case excess of 20,152 (29.80%) or 11,753 anomalies (22.30%) respectively. During this period the use of tobacco and alcohol was declining and other drug use was not rising. Only cannabis use rose. Importantly, models quartic in time indicated a non-linear response of total birth defects to rising cannabinoid exposure. Estimated exposure to several cannabinoids including cannabinol, THC, and tetrahydrocannabivarin was shown to be positively associated with major defect rates and to be robust to adjustment for other drug use. CNS defects (microcephalus, neural tube defects), cardiovascular defects (ASD, VSD, patent ductus arteriosus (PDA)), total chromosomal anomalies including Downs syndrome, musculoskeletal, respiratory and genitourinary anomalies all rose dramatically.
Defects described as being cannabis-related (by the Hawaiian, CDC, AAP and AHA investigators) rose more quickly than cannabis-unrelated defects (P<0.003). As fetal cardiac tissue and the central great vessels have high numbers of cannabinoid receptors from early in fetal life it is easy to understand why this pattern might emerge. Since ASD, VSD and PDA are the most common cardiovascular congenital anomalies it is understandable that total cardiovascular anomalies increased in Colorado.
A recent review of total congenital anomalies in Canada showed that they were 3 times more common in the northern territories which consume more cannabis, and that these effects were robust to adjustment for other drug exposure and for socioeconomic variables.16 Total cardiovascular defects, Downs syndrome and gastroschisis were noted prominently in this series. Neural tube defects including anencephalus and spinal bifida and meningomyelocoele were falling across Canada from 1991 to 2007, although it was not clear whether the decline was due to dietary folate supplementation or increased antenatal early termination of pregnancy for anomalies (ETOPFA).17 Notwithstanding this it was recently shown that within each of 3 periods (the pre-folate period, the transitional period and the post-folate period) neural tube defects across Canada were becoming more common.17
An Australian dataset found greatly elevated relative rates of cardiovascular (PDA, ASD, VSD, ToF, transposition of great vessels), body wall (gastroschisis, exomphalos, diaphragmatic hernia), chromosomal (Downs syndrome, Turners syndrome, Edwards Syndrome (trisomy 18)), genitourinary, hydrocephalus, neural tube defects, and bowel defects with borderline results for anencephalus (ETOPFA data unavailable) in a high cannabis use area in Northern New South Wales compared to Queensland state-wide data.18
Transposition of the great vessels was previously linked with paternal cannabis exposure.19
The presence of Downs syndrome on the list of cannabis-associated anomalies in Hawaii, Colorado, Canada and Australia is important as it necessarily implies megabase-scale genetic damage.8,15,16,18 Since cannabis interferes with tubulin metabolism and thus the separation of the chromosomes which occurs in mitotic anaphase it is easy to see how PCE-induced chromosomal mis-segregation errors might occur.20 Studies of PCE in rodents show that cannabis induces major alterations of gene expression widely with 8% alteration in DNA sperm methylation patterns, changes which are transmissible to subsequent F1 generations.21
Stanciu’s comment about a so-called “cannabis phenotype” is provocative. It is true that a “fetal cannabis syndrome” (FCS) has not been described in the way that a “fetal alcohol syndrome” (FAS) has. Fetal alcohol syndrome of course is a very diverse and pleomorphic group of clinical presentations and a wide spectrum of presentations is described. Importantly the fetal alcohol has been described as being mediated by the cannabinoid type 1 receptor (CB1R’s) and is mediated epigenetically.22-26 The suggestion that alcohol can work epigenetically via CB1Rs but cannabinoids cannot defies the bounds of credulity. Moreover, as noted above, there is as yet no objective marker of gestational cannabinoid exposure. Once such a biomarker has been derived (say epigenetically and / or glycomically27) then an objective measure will exist to allow genotype-epigenotype-phenotype correlative studies to be performed so that we can usefully investigate if a fetal cannabis syndrome phenotype spectrum might exist. However, if researchers do not believe it might exist then it is clear that one will not be described. It is our view that once an objective biomarker is established it will only be a matter of time before a diverse and highly variable FCS is also defined and enters the clinical diagnostic compendium.
For complete and extensive data go to Psychiatric Times