Introduction
Illicit drug, alcohol, and tobacco use in pregnancy has known risks. As it is the rare exception that a woman will initiate use of these substances during pregnancy, the best strategy to reduce the risk of substance exposure in newborns is to provide comprehensive screening and counseling to women prior to pregnancy. Behaviors that may pose very little health risk prior to pregnancy, such as occasional alcohol consumption, become much more worrisome once a woman conceives.
The particular challenges of understanding the consequences of fetal exposure to various illicit and recreational drugs are worth noting. Randomized controlled trials that include exposing pregnant women to drugs with suspected harm and no purported benefit are clearly unethical. Data in animal models may not represent the full spectrum of toxicity in the human pregnancy. Understanding this, the information that is available to clinicians regarding effects in human pregnancies is reliant on observational studies. However, there are several limitations to extrapolating the results of these studies.
First, enrollment in observational studies relies on self-report and/or biologic screening. Self-reporting of substance use in pregnancy is likely to lead to underreporting
1 and, therefore, skewed data. Biologic screening in pregnancy is also fraught with bias as the rate at which women clear substances from the bloodstream or urine is variable per substance, so screening at any one point in time may underrepresent use of illicit drugs, such as opiates, anxiolytics, and stimulants, and overrepresent cannabis use.
2,3,4
In addition, exposure to illicit drugs at different times in pregnancy likely leads to different consequences. Many studies, however, study drug exposure in a binary way (exposed/unexposed) and do not clearly differentiate differences in timing of exposure. Similarly, although a dose-response relationship is not definitively known for many drugs, it stands to reason that different levels of exposure or routes of consumption may cause varying effects. However, many studies use exposed/unexposed as qualifiers rather than having any measure of quantification.
Researchers have considered that timing and dose may alter the effects of these substances, but these measurements are often difficult or impossible to obtain. Most women either quit or reduce their use of potentially harmful substances during pregnancy.
5 Those who do not abstain from use typically still engage in some form of harm reduction. Although this is good news for clinical care, it likely causes an underestimation or misunderstanding of how drug exposure in later pregnancy may affect outcomes. Attempts to quantify drug use throughout pregnancy also likely underestimate exposure because women are reticent to admit ongoing drug use due to social desirability bias.
Additionally, there are a number of potentially significant confounders that may affect the outcomes of pregnancies in women who use drugs. Many women who use drugs have concurrent mental health disorders, which are known to increase adverse maternal and neonatal outcomes.
6 Women who use drugs also have a high prevalence of previous abuse and are also at high risk of intimate partner violence, which can influence their health and pregnancy.
7 There are high rates of polysubstance use with multiple illicit substances as well as high rates of concomitant tobacco use, which can have synergistic effects and make differentiation of which substance causes which effect quite challenging.
There is evidence that engagement in prenatal care can significantly improve the outcomes of pregnancies in women who use drugs.
8,9,10,11 This may mean that prenatal care is effective in helping women with harm reduction, or it may mean that many of the risks attributed to drug use are actually caused by lack of access to or engagement in care. Notably, punitive action against women who use drugs in pregnancy has not been shown to decrease drug use or improve outcomes, but has been shown to deter women from seeking care.
11,12 This highlights the importance of improving access to compassionate, nonjudgmental care to all women to allow for the best possible maternal and neonatal outcomes.
Women with substance use disorders are more likely to present later for obstetric care. This is likely multifactorial with issues such as late recognition of pregnancy due to irregular menses, lack of access to care, psychosocial issues, and fear of judgment or reprisal from members of the healthcare system. Because one in four people with a substance use disorder also suffer from a mental health disorder, often undiagnosed,
13 increased access to substance use treatment, mental health services, and comprehensive preconception counseling and contraception is paramount.
Alcohol
Alcohol exposure
in utero is a leading cause of preventable mental disabilities. According to the Centers for Disease Control and Prevention
14 and the US Surgeon General, there is no safe amount of alcohol in pregnancy.
14 Given that over half of pregnancies are unplanned and that it is possible that alcohol exposure prior to recognition of pregnancy could cause birth defects, the Centers for Disease Control and Prevention recommends that women of childbearing age limit their alcohol intake.
14 Pregnant women are advised to avoid alcohol during routine preconception and antenatal care. However, 11.5% of pregnant women admit to consuming alcohol within the past 30 days and 3.9% admit to binge drinking.
15
The worldwide prevalence of fetal alcohol spectrum disorder (FASD) is estimated to be approximately 7.7 in 1000 births.
16 The term fetal alcohol syndrome (FAS) was originally coined to describe 11 children of mothers who consumed large quantities of alcohol during pregnancy with characteristic features (
Table 8.1).
17,18 In 1980, specific criteria for FAS were proposed by the Fetal Alcohol Study Group of the Research Society on Alcoholism, requiring that at least one feature from each of the three categories be present for the diagnosis to be made.
19 In 1996, the National Academy of Medicine (then Institute of Medicine) defined five categories of alcohol-related birth effects
20 (
Table 8.2).
It is difficult to construct a dose-response curve for levels of alcohol exposure necessary to produce FASD because it is difficult to accurately quantify alcohol intake; the amount of alcohol used is invariably underestimated. In addition, specific features of FASD may be caused by exposure to alcohol at different times in gestation. For example, craniofacial anomalies may occur upon alcohol exposure during embryogenesis, disorders of central nervous system function may be brought about by exposure later in gestation, and growth disturbances may reflect alcohol exposure over a broad range of gestational ages.
The prevalence of FASD among offspring of heavy drinkers (for women, consuming ≥8 drinks per week) ranges from 2.5% to 10% in prospective studies.
14,21,22,23,24 Estimates of FASD prevalence in the overall US population in 2017 were 15 to 20 per 1000.
16 Although it is difficult to obtain more specific data, it is clear that all cases of full-blown FASD occurred in mothers with chronic alcohol abuse and in mothers who drank heavily throughout
pregnancy. It is also clear that binge drinking (≥4 drinks during a single occasion) increases the risks of affected children having IQs in the range that is considered intellectual disability.
22
Studies have attempted to identify a threshold level of maternal ethanol ingestion above which effects are seen and below which effects are not seen, without notable success. An analysis by the Spanish Collaborative Study of Congenital Malformations determined that even low and sporadic alcohol ingestion during pregnancy increases risk of congenital anomalies.
25 Other works have demonstrated an association between alcohol consumption in pregnancy and increased neurodegeneration via apoptosis (programmed cell death) during synaptogenesis,
26 defects in neuronal migration and decreased numbers of neurons in the mature cortex,
27 and effects on myelination leading to decreases in white matter volume.
28
The pathophysiologic mechanisms for the adverse effects of alcohol on the developing fetus are poorly understood, and what is known is based on studies in animal models
29 Several potential pathways have been proposed
30,31,32,33:
Ethanol effects the production of normal cellular morphologic differentiation modulators and inhibits normal neuronal growth and differentiation. Changes in neuronal development/migration, apoptosis, and myelinization are likely to be at the root of the neurobehavioral changes seen in children exposed to ethanol in utero. Acetaldehyde, a toxic metabolic product of ethanol, behaves similarly.
Genetic variance in the enzymes that metabolize alcohol, including alcohol dehydrogenase in the cytosol fraction of the liver, the hepatic microsomal ethanol-oxidizing system, and the peroxisomal catalase system, may lead to differential enzyme induction and, thus, divergent susceptibility to cell injury or disruption by alcohol.
Ethanol or acetaldehyde may interfere with placental transport of vital nutrients, such as amino acids, leading to fetal malnutrition. Either interference with normal cellular development or fetal malnutrition may result in the observed 150 g lower birth weight of babies exposed to one to three drinks per day.
34
Alcohol ingestion by the gravida decreases fetal breathing movements but not gross body movements or fetal heart rate.
35 Alcohol withdrawal effects have been described, even in infants without the stigmata of FASD.
36 The fetal effects of alcohol appear to be multifactorial (
Table 8.1).
There is no treatment or cure for FASD or alcohol-related birth defects. The best strategy is prevention, and the first component of prevention is educating women of childbearing age of the risks of drinking during pregnancy. Since the recognition of FASD nearly 50 years ago, many educational campaigns have been carried out in the United States. A seminal epidemiologic study in Seattle reported that any alcohol use in early pregnancy fell from 81% in 1974 to 1975 to 42% in 1980 to 1981, and this was attributed to educational programs.
37 Although the prevalence of heavy drinking (defined in this study as 30 mL or more of absolute alcohol per day) before the discovery of pregnancy was unchanged at 6% to 7% during each time period, such use after pregnancy fell from 2.3% to 0.8%.
Another critical piece to preventing FASD is identifying mothers at risk. The CAGE (Cut down, Annoyed, Guilt, Eye-opener) (
Table 8.3), T-ACE (Take [number of drinks], Annoyed, Cut down, Eye-opener)
38 (
Table 8.4), and TWEAK (Tolerance, Worried, Eye-opener, Amnesia, K/Cut down)
39 (
Table 8.5) questionnaires include basic questions that are effective tools in patient management. Two or more positive answers on the CAGE questionnaire indicate a high probability of being a risk drinker. In the T-ACE questionnaires, the ability
to consume a six-pack of beer or a bottle of wine scores two points on the tolerance question; affirmative answers on the other questions each score one point. A cumulative score of more than 2 indicates a high probability of being a risk drinker. A total of ≥2 points on the TWEAK questionnaire also indicate a high probability of problem drinking. One of these questionnaires should be administered to every patient when obtaining a history.
Once an individual at increased risk for having a child with alcohol-related birth defects is identified, attention should be turned to intervention. Counseling programs have proven effective in reducing chronic and binge drinking. Rosen and coworkers
40 found that 67% of 49 pregnant problem drinkers reduced their alcohol intake when enrolled in a counseling program. Work in both the United States and Finland found that cessation of heavy alcohol intake before the third trimester benefits the fetus.
41,42 A randomized trial, although not specific to pregnancy, demonstrated that two brief (10-15 minutes each using scripted advice and educational information) interventions decreased both chronic and binge drinking.
43 Antioxidants have been explored to prevent FASD in animal models and
in vitro, but these findings have not been confirmed in humans.
44 The drug disulfiram is used to increase the motivation of patients with alcohol use disorder to avoid consumption of alcohol; however, it is generally contraindicated in pregnancy.
45 Although not a proven teratogen, when disulfiram is taken in combination with alcohol, it leads to very high circulating acetaldehyde concentrations. Even when a history of alcohol consumption has been identified, interruption of pregnancy is not a recommended intervention because the risks of adverse perinatal outcomes associated with low-level alcohol consumption, or even occasional binges of marked consumption, have not been clearly quantified to date.