In Depth

Cocaine, Pregnancy and Risk of Intrauterine Death

Richard M. Pauli, M.D., Ph.D.

Teratogenesis, fetotoxicity and fetal death. In assessing the potential harm that may arise secondary to cocaine use in pregnancy, clear distinction needs to be made between teratogenic and fetotoxic effects. A teratogen is a chemical or physical agent that can cause abnormalities in an embryo or fetus. In general this term is restricted to birth defects, structural aberration etc. and usually is related to effects during the period of organogenesis (i.e. in general prior to about 10 weeks gestation). In contrast, while a fetotoxin may cause adverse effects at any time in pregnancy, this term usually refers to effects that arise following the period of organogenesis. The measurable endpoints of fetotoxicity include death in utero. Thus, any given agent may be only a teratogen, only a fetotoxin, or both.

Alcohol is a now well recognized teratogen. It causes the fetal alcohol syndrome in 10% or more of appropriately exposed infants and fetal alcohol effects in others. Perhaps less well recognized are its fetotoxic effects. Studies investigating the frequency of intrauterine death vs. alcohol consumption [e.g. Harlap & Shiono, Lancet 2:173-176, 1980] demonstrated that the relative risk for intrauterine death is significantly increased by as little as 1-2 alcoholic drinks a day. Except for alcohol (and tobacco), cocaine is now the most widely used drug during pregnancy. Are there similar risks of birth defects, intrauterine death and other adverse outcomes because of cocaine exposure?

Epidemiology of cocaine abuse. Around 15% of all Americans admit to having tried cocaine and around 2-3% are regular users of cocaine in some form. The vast majority now use the crystallized free base of cocaine (‘crack’) administered by inhalation (smoking).

Pregnant women are no exception to the epidemic proportions of cocaine use in this country. In reality, cocaine use is higher in young adults and higher in poorer socioeconomic populations — two factors that are also correlated with a high probability of pregnancy. General population estimates of cocaine use by pregnant women range from around 3% to 15% (although it is as high as 45-60% in certain communities). Most abusers of cocaine are not without other factors that place a pregnancy at risk. Indeed, as well described by Burkett et al. (1994) an entire social milieu results in a pattern of risk not only related to polydrug use, but to increased risk of sexually transmitted diseases, of undernutrition, of violence, of lack of prenatal care etc. all of which can place the fetus at risk

Cocaine pharmacology. Cocaine is an alkaloid derived from the leaves of Erythoxyulon coca which grows in various parts of South America. Whether smoked (as the crystallized base) or absorbed through the nasal passageways, or otherwise administered, it is rapidly taken up and rapidly excreted, with a half life in humans of about 40-80 minutes.

Its primary pharmacologic effect is to inhibit re-uptake of dopamine and serotonin in the brain. Accumulation of these neuro-transmitters then results in a transient euphoria. That euphoria is sufficiently rewarding, evidently, that cocaine has marked psychological addictive potential. The excess neurotransmitters also result in activation of the sympathetic nervous system resulting in physical manifestations including rapid heart rate, hypertension and vasoconstriction.

Cocaine is of low molecular weight and is lipophilic. These properties mean that it easily diffuses across the placenta. In fact, fetal levels lag behind maternal blood levels by only around 5 minutes. The fetal levels are virtually identical to those found in the mother. So, whatever physiologic effects arise in the abuser, likewise one can anticipate in the fetus.

Since cocaine reaches the fetus, harmful effects may arise either indirectly, through effects on the mother and placenta, or directly. Maternal hypertension and direct vasoconstriction of uterine vessels can compromise blood flow to the placenta. That in turn means decreased delivery of oxygen to the fetus as a whole. Alternatively, vasoconstrictive effects within the fetus can result in cycles of relative ischemia and rebound vasodilation and hyperperfusion within various organs.

Adverse effects of cocaine in pregnancy. The first report of an adverse pregnancy outcome thought to be caused by cocaine was in 1983. Two instances of placental abruption associated with maternal cocaine use were recorded. Curiously, before this it was thought that cocaine was utterly without harmful effect in pregnancy.

Subsequently, large numbers of studies have been published. Many are anecdotal case reports or case series. Some are efforts to isolate cocaine effects from all of the potential confounders using various kinds of population assessments.

Few have, instead, assessed the risk of the entire social milieu of cocaine abuse. Burkett et al. (1994) is a quality example of this last approach. Studies such as this can assess what we might expect in the real world. That is, by purposefully including all possible confounders, these researchers are asking a different question than have most other studies. Rather than asking, can we isolate, identify and quantify the risks of cocaine per se, Burkett et al. ask what sequelae can one expect in a population of cocaine abusers and what characteristics place subsets of these abusers at highest risk. They showed that cocaine binging embraces a multiplicity of risks (not just cocaine use itself, but other drug use, lack of nutrition etc. etc.). Together that lifestyle appears to markedly increase the risks of complications of pregnancy. Furthermore, the probability of most such complications is directly correlated with the frequency of cocaine binges. The one significant exception to that correlation was the risk for fetal death (see below).

Trying to isolate the effects of cocaine. If one wants to know about the fetotoxic or teratogenic potential of some particular agent, then the effects of that agent must be defined related to exposure and separate from whatever other environmental variables may be present. This has proven to be particularly difficult in studies of cocaine.

First, documentation of exposure is not all that easy. Current methods detect use only within about the preceding 7 days. Certainly self-reporting is going to be inaccurate (probably even more inaccurate than the notoriously unreliable self-reporting of alcohol use in pregnancy since cocaine use can result in legal sanctions).

Secondly, as already noted, there are a series of confounding variables that need to be taken into account. Users of cocaine may well (and usually do) differ from non-users by age, nutritional status, use of other licit and illicit drugs, by their level of prenatal care and so forth. To isolate the effects of cocaine itself, all of these confounders need to be controlled. That is a daunting task requiring large population groups beyond the scope of virtually all clinical studies.

Finally, there is an inherent bias of publication and reporting — studies that have demonstrable abnormal results are more likely to be published than those with negative results. On the one hand, then, considerable caution needs to be taken in reading the available literature. On the other, too harsh an assessment of the problems in that literature may tempt us to reject all of the tentative associations that have been found. Neither approach is prudent.

The effects of vasoconstriction. There is no doubt that cocaine results in vasoconstriction in the mother, in the placenta and in the fetus. It is likely, but generally unproven, that many of the adverse outcomes arise from such vasoconstrictive effects. Vasoconstriction of placental arteries by cocaine demonstrably causes decreased oxygen saturations in the fetus and probably also results in decreased nutrient supply.

Cocaine and abruption. Epidemiologic evidence is strongest for the association of cocaine use and abruption of the placenta. Various studies have placed the risk for abruption in cocaine users at around 10-19%, with relative risk estimates of from 5 to 20 fold over risks for a control group (with other confounding variables held constant). Nonetheless, only a minority of exposed pregnancies will end in placental abruption, suggesting that other factors may be important and that abruption may arise only within a unique set of environmental circumstances.

Instances of cocaine use immediately preceding the placental abruption suggest a direct and dramatic effect. It has been suggested that vasoconstriction might cause disruption of placental adherence to the uterine wall and thereby directly result in abruption.

Abruption, of course, results in secondary risks, including intrauterine death.

Cocaine and other adverse outcomes. Although the epidemiologic data are weaker than for the association with abruption, there is convincing evidence that cocaine use increases the probability of intrauterine growth retardation, preterm labor, premature rupture of membranes, pregnancy induced hypertension, precipitous delivery and for certain birth defects. There is strong correlative evidence that cocaine (without the usual accompanying poor nutrition) can lower birthweight, but the effect seems to be modest — perhaps only 60-200 g on average. Prematurity secondary to preterm labor is frequently reported anecdotally, but demonstrating an unequivocal effect of cocaine alone has been difficult. Similar difficulties attend demonstrating causality in premature rupture of membranes, precipitous delivery etc.

Whether cocaine is a true teratogen (i.e. can cause structural birth defects) has been hotly debated. On the one hand, controlled studies have not shown a particularly high rate of malformations overall in cocaine exposed infants. On the other hand, the particular defects that have been reported are consistent with a biologically plausible explanation of cocaine’s mechanism. Most birth defects reported to be associated with cocaine exposure are just those that are thought to be caused by hypoxia, ischemia, vascular disruption or hemorrhage. These include, for example, certain structural urinary tract anomalies, limb reduction defects, ischemic brain abnormalities, intestinal atresias, hemifacial microsomia and so forth. It seems likely that intermittent vasoconstriction and rebound vasodilation (with or without hemorrhage) arising as a direct result of cocaine use can be teratogenic. Cocaine, then, is probably a vasoactive teratogen, but a relatively weak one, resulting in birth defects in only a small minority of exposed infants.

Note that while many exposed infants have neonatal neurologic abnormalities, long term followup suggests an excellent prognosis so long as the cocaine exposed child is provided with proper parenting and appropriate educational support. The earlier dire warning of an epidemic of crack babies who would flood our schools as special needs children is not becoming a reality.

Cocaine and fetal death. In the ‘real world’ of cocaine abusers, there is a marked excess risk of stillbirth. Indeed, Burkett et al. found that risk to be from 4-25% depending on the pattern of use. Those risks, of course, include the risks of all of the confounding variables of that lifestyle. They also showed that the highest risk was in erratic users (unlike virtually all other adverse outcomes which correlated with frequency of use), probably because erratic users more frequently were those using the most drug at any one time.

Controlled studies suggest that the relative risk of intrauterine death directly attributable to cocaine is probably in the range of 2 to 3 fold. If one assumes a 5% user frequency in the general population of pregnant women, this would imply that currently perhaps as many as one in every ten intrauterine deaths in the United States results from use of cocaine.

Most likely, most of these intrauterine deaths are related to vasoconstriction of umbilical and uterine blood vessels, resulting in hypoxemic sequelae or through the effects of inducing placental abruption. So, not surprisingly, in most instances, intrauterine death is a proximate outcome of recent use of cocaine.

Additional Reading:
[The following articles are relatively recent ones which are particularly relevant to consideration of perinatal effects of cocaine and, most specifically, discuss the risks of intrauterine death associated with cocaine exposure.]*

Burkett G, Yasin SY, Palow D, LaLoie L, Martinez M (1994) Patterns of cocaine binging: Effect on pregnancy. Am J Obstet Gynecol 171:372-379.

Holzman C, Paneth N (1994) Maternal cocaine use during pregnancy and perinatal outcomes. Epdemiologic Rev 16:315-334.

Kain ZN, Rimar S, Barash PG (1993) Cocaine abuse in the parturient and effects on the fetus and neonate. Anesth Analg 77:835-845.

Martinez A, Larrabee K, Monga M (1996) Cocaine is associated with intrauterine fetal death in women with suspected preterm labor. Am J Perinatol 13:163-166.

Slutsker L (1992) Risks associated with cocaine use during pregnancy. Obstet Gynecol 79:778-789.

Young SL, Vosper HJ, Phillips SA (1992) Cocaine: Its effects on maternal and child health. Pharmacotherapy 12:2-17.

*Copies of these and other relevant articles are available for personal use by request from WiSSP.