The puzzle of cocaine's effects following maternal use during pregnancy: Are there reconcilable differences?

Neurotoxicologyand Teratology,Vol. 15, pp. 281-286, 1993

0892-0362/93$6.00 + .00 Copyright©1993PergamonPress Ltd.

Printedin the U.S.A. All rightsreserved.

OPEN PEER COMMENTARY PERSPECTIVE

The Puzzle of Cocaine's Effects Following Maternal Use During Pregnancy: Are There Reconcilable Differences? D O N A L D E. H U T C H I N G S l

N e w York State Psychiatric Institute, Department o f Developmental Psychobiology, N e w York, N Y 10032 Received 7 D e c e m b e r 1992; A c c e p t e d 15 J u n e 1993 HUTCHINGS, D. E. The puzzle of cocaine's effects following maternal use during pregnancy: Are there reconcilable d(fferences? NEUROTOXICOL TERATOL 15(5)281-286, 1993.-This is a selectivereview of the clinical and epidemiological literature. It attempts to reconcile disparate and contradictory f'mdings dealing with the morphologic, growth, and neurobehavioral effects reported to occur in neonates and young children exposed prenatally to cocaine. A history of cocaine use in the United States is briefly presented followed by impressionistic observations of some of the events that transpired during the cocaine epidemic of the 1980s. Based on the collective research findings, it is tentatively suggested that the teratogenic effects of prenatal cocaine may be produced only in those infants exposed to the highest doses reported in the literature. It remains unknown, however, whether or not these effects may be dependent on the concurrent abuse of alcohol and/or dgarcttes. As to growth and neurobehavioral outcomes, effects attributable primarily to cocaine alone and not other substances of abuse appear to be only marginal and transitory. The data to support these conclusions are particularly tenuous and are thus offered only as working hypotheses. Because of the intractable methodological and interpretive problems inherent in human devdopmental research on substance abuse, any attempt to draw definitive conclusion is admittedly premature. Because these methodological problems also complicate the efforts of ongoing studies, answers to these persistent questions may not be readily forthcoming. Cocaine Prenatal Experimental confounds

Birth defects Neurobehavioral effects Growth Head circumference Polydrug abuse Genitourinary tract malformations

THE past two decades marked an historical watershed that witnessed an unprecedented accumulation of epidemiological and basic research findings on the effects of drug abuse during pregnancy. Both human and animal researchers sought to identify sources of experimental confounds and through conceptual explorations and debate honed these insights into methodological imperatives that produced more rigorously designed studies and substantive conclusions. Epidemiologists became more successful at resolving the complex problems of studying the compulsive use of drugs embedded in the morass of poverty, disease, and psychosocial turmoil. Basic research-

ers continued to refine animal models so that they became both better controlled and more relevant to the pharmacologicad peculiarities of the abuse compounds. By the beginning of the 1990s our knowledge of the hazards to the offspring from maternal use of opioids, cannabinoids, and ethanol had advanced appreciably both in precision and reliability. Despite these notable achievements, when cocaine burst on the scene in the mid-1980s, drug abuse researchers and practioners alike found themselves thrust suddenly to the murky perimeters of their science. Though the public assumed that scientists and health-care workers generally agreed that

Requests for reprints should be addressed to Donald E. Hutchings, Box 42, 722 West 168th Street, New York, NY 10032. Supported by Grants DA03544 and DA07822 from the National Institute on Drug Abuse. Portions of this Perspective were adapted from Hutchings, D. E. A contemporary overview of behavioral teratology: A perspective from the field of substance abuse. In: H. Kalter, ed. Issues and reviews in teratology. New York: Plenum Press; 1993:125-167. 281

282

HUTCHINGS

cocaine committed horrific effects on the fetus and newborn infant, few outside the field were aware of the disparate views that actually prevailed. As this new drug story unfolded, unrelenting media reports, like a runaway firehose, sent the scientific community into rampant confusion. Although here I describe my own impressions about these events, let me make clear at the outset that I am neither an historian nor purport to have expertise on the interpretation of complex social events, how they are reported in the media, or what ultimate impact they might have on the way in which public health research is conducted and interpreted. I include this information because of my belief that many of these historical occurrences provide an important context that gave both shape and impetus to the developmental cocaine research as it evolved. A BRIEF HISTORYOF COCAINEIN AMERICA Over the past century, American society has had a tempestuously ambivalent attitude toward cocaine (for a historical review of cocaine in the United States, see ref. 14). Its early virtues-euphoria and central nervous system stimulationwould subsequently be perceived contrarily as harmful, addictive, and life-threatening. In 1890, Coca-Cola, the softdrink icon of America, was touted as a therapeutic nerve and brain tonic because it contained "the wonderful properties of the Coca plant." Not only had cocaine become a popular overthe-counter hay fever remedy, it was promoted by Dr. William A. Hammond, better known to the history of medicine as one of the founders of modern neurology. He thought it a safe and effective palliative for serious melancholia and "feeling blue." He dismissed colleagues concerns of cocaine's addictive potential and proudly concocted what he considered the perfect "coca w i n e " - a pint of wine laced with two grains of c o c a i n e - a mixture which he drank regularly with his meals. By 1900, public attitude was dramatically shifting from the perception of cocaine as a health-promoting brain elixir to a highly addictive drug sought after by increasing numbers of "pleasure-users" in whom it impaired judgment, produced agitation, and promoted criminal violence. By 1903, cocaine had been removed from the Coca-Cola formula and in 1914, the Harrison Act severely restricted its availability except by medical prescription. In a period of some three decades, public sentiment changed from unbridled enthusiasm to a dark view that both feared cocaine and reviled those who used it. By 1940, the recreational use of illicit cocaine had become restricted largely to the underworld and it was not until the late 1970s that it once again gained a new popularity, particularly as a modern day snuffing powder of the rich and famous. Costs of $1000 or more per ounce, a relatively limited availability, and images of it being snorted through rolled-up $100 bills only added to its mystique. It quickly gained a reputation for endowing its users with feelings of supreme confidence, fearless self-assurance, as well as potent sexual prowess and performance. During the early 1980s, its use, particularly as reported in the popular press was commonly, if not somewhat innocently, associated with wealthy show business notables and the nouveau riche "yuppies" of high finance. By the mid-1980s the cocaine drug scene changed dramatically and its ugly shadow side, reminiscent of the early 1900s, emerged once again. The upsurge in its popularity brought with it increasingly alarming reports of its potent addictive properties and a host of pernicious side effects not the least of which were acute hypertension, seizures, stroke, cardiac arrythmia, and sudden death. Cocaine-related emergency room admissions were on the rise in major metropolitan cities

and cover stories of dally newspapers, the major weekly magazines, nightly television news segments and special programs chronicled in lurid detail the ravages that cocaine addiction had wrought on the lives of the rich and poor, young and old. In 1985, the first in a series of studies from several laboratories reported on the maternal use of cocaine during pregnancy and the effects on the the fetus and newborn infants were, indeed, alarming (for comprehensive reviews, see 5,15,19,20). These first reports described increased spontaneous abortion, abruptio placenta, premature labor and delivery, intrauterine growth retardation, reduced head circumference, sudden infant death syndrome, and risk of congenital malformation of the genitourinary tract (1,11). There were also reports of hypospadias, craniofacial defects, and missing digits. Neurobehavioral effects seen among these infants included increased tremulousness, irritability, lack of consolability, and poor state regulation. Especially disturbing were reports of a significant risk of perinatal cerebral infarction (2) and the fear that these cocaine-exposed neonates would suffer permanent and severe neurological damage. Most of the women in these first studies were taking cocaine by the intranasal route (i.e., "snorting") although a few were also reported to be IV users. Despite the relatively small sample size of these preliminary studies and a host of methodological problems that compromised clear interpretation, these reports were generally, and to a large extent, uncritically accepted among many in the scientific research community. One possible explanation for the suspension of scientific judgment about the soundness of these findings is that the adverse effects described for the infants may have been seen as a logical result of the severe toxic effects already described among adult users; if it is so toxic to adults, how could fetuses and neonates be spared? In the midst of what appeared to be an epidemic of cocaine use and associated catastrophic health effects on both the adult and conceptus, a major change was taking place in the illicit marketing of cocaine. The increased demand for cocaine was met with abundant supplies from what had become by official estimate, the world's most powerful drug trafficking operation, the Colombian-based Medellin Cartel. It supplied the vast bulk of marijuana and cocaine worldwide but the United States was its most profitable market, yielding earnings from cocaine alone estimated at about 8 billion dollars per year. Coca processing centers located in Colombia's Amazon River basin converted coca leaf mash into crystalline cocaine in quantities of tons per week (described in ref. 6, pp. 261274). Amidst this avalanche of illicit drugs entering the United States, a new and far cheaper form of cocaine became available that met with immediate and enormous popularity. Creative drug entrepreneurs desalted the powder to return it to a crystallized free-base, packaged it in plastic vials in small single dose amounts and sold it cheaply-usually a few dollarsas "crack." The user heated it to vaporization and inhaled the volatilized mixture, producing a potent euphoria that was experienced within seconds. This easily purchased, powerfully addictive form of cocaine spread like wild-fire, particularly among poor and disadvantaged populations. And along with it, two new media stars emerged, the "crack addict" and "crack babies". In the latter part of the 1980s, two of the most common scenes to be shown on the evening news were cocaine addicts smoking crack and so-called "crack babies." News video crews were invited into neonatal intensive-care units where they trained their cameras on obviously very sick, underweight, and distressed babies attached to various monitoring devices and an assortment of tubes, who were described as the inno-

COCAINE AND HUMAN PREGNANCY cent victims of addict mothers who used crack cocaine during pregnancy. This unrelieved litany of anecdotal reports fueled public outcry for action against the drug merchants as well as the users, and there was particular outrage against the perceived immorality of mothers who would abuse drugs while pregnant. One result was that in several states, zealous prosecutors began prosecuting pregnant women under criminal child abuse and neglect laws or drug trafficking laws that were never intended to apply to a woman's conduct during pregnancy. In one case in Florida, for example, a prosecutor argued that a woman was guilty of delivering an illegal substance to a minor because cocaine was delivered to her fetus through the umbilical cord (17). A legitimate concern of health care providers and particularly those who provided treatment for substance abusers was that those women who would have sought treatment were seriously conflicted about seeking such services for fear of detection and prosecution. Most agreed that the place for these women was in a treatment facility for drug rehabilitation, not in jail for punishment. The deplorable fact, however, was that beds for poor uninsured addicts were in scarce supply and few treatment facilities were either capable or willing to treat addicts who were pregnant (9). This was, however, a most confusing and exasperating time for many substance abuse researchers, particularly those who were interested in developmental effects and worked in neonatology, epidemiology, and developmental psychology. While all were acutely aware of the published reports citing serious developmental effects reported for cocaine use during pregnancy and could hardly escape the daily deluge of anecdotal reports in the press, many of these researchers were not confirming these observations in their own hospital units. Many substance abuse practitioners and scientists at major metropolitan hospitals were seeing large numbers of neonates who tested positive for cocaine but were not observing the more serious toxic effects-malformations and "withdrawal" sympt o m s - t h a t had been given such prominence in the first reports. Though there appeared to be an association of cocaine use in the third trimester with premature labor, most exposed infants presented in the nursery as remarkably normal and asymptomatic. Studies published in the ensuing years may be shedding some light on these bewildering contradictions. MATERNAL/FETAL EFFECTS

Of all the abuse substances, cocaine exerts unusually potent effects on materual/fetal physiology. In humans it has been shown to produce complex cardiovascular effects, direct stimulatory effects on myocardium, and vasculature as well as local anesthetic effects. While the effects of cocaine on the maternal/fetal cardiovascular system have not been systematically studied in clinical populations, two reports have appeared which identify these effects in the pregnant lamb model (13,21). These studies demonstrated a dose-response relationship between the concentration of IV cocaine and decreased placental blood flow, increased uterine vascular resistance, increased fetal heart rate, and decreased fetal oxygen content. Fetal tachycardia and hypoxia were of greater magnitude when cocaine was administered via maternal circulation than when delivered directly to the fetal circulation (21). In the fetal ewe, hypoxia causes a release of catecholamines in fetal circulation which would be expected to act synergistically with cocaine at peripheral catecholamine synapses. (For a recent review of the developmental toxicity of cocaine and possible mechanisms by which it produces adverse fetal effects, see ref. 20.)

283 RECENT HUMAN STUDIES

From 1980 to 1989, Koren et al. (8) examined the fate of abstracts submitted to the Society for Pediatric Research related to prenatal exposure to cocaine. Comparing acceptance and rejection decisions, a statistically significant bias was found. Of studies that reported adverse effects associated with cocaine, 58% were accepted whereas only 11% of those that found no effects enjoyed a similar fate. Moreover, the noeffect studies were more likely to be rejected even though they might be better controlled (e.g., larger sample size, better verification of drug use, less polydrug abuse). Although these observations represented the selection practices of abstracts for one society, it certainly raised the larger issue of selection bias related to study outcome and the disturbing notion that if the findings of a study went against "conventional wisdom," it was considered by some as less credible or of less scientific interest, despite the use of a more rigorous methodology. At about the same time, studies that evaluated both this literature as well as studies that failed to replicate the earlier reports began to appear. To arrive at an overall quantitative evaluation of prenatal cocaine effects, Lutiger et al. (10) used the statistical technique of meta-analysis to analyze reports that appeared up to 1989. This procedure utilizes pooled data from several studies sharing similar design features and characteristics to arrive at an overall quantitative estimate of drug effects. As the authors point out, meta-analysis does not overcome the limitations of individual studies (i.e., accurate assessment of cocaine exposure, differences in cigarette and alcohol use, polydrug use, etc.) but suggest that it does provide a more accurate estimate of risk because random errors tend to cancel out, leaving a clearer picture of "real" or relatively unconfounded effects. Data were combined from 20 research papers on the effects of cocaine on human pregnancy and included comparisons of the following groups: 1. 2. 3. 4.

No drug use Cocaine alone Polydrug users who did not use cocaine Polydrug users who also used cocaine

The findings are complex, vary according to comparison groups, and are somewhat tentative due to the small number of cases in each of the categories. First, comparing those studies that examined cocaine only with drug-free controls showed that there was a significant increase in in utero death. And though there was no overall increase in malformations, a significant increase specifically for genitourinary tract (GU) malformations was found among the cocaine only group. A similar relationship was found when comparing polydrug/cocaine with drug-free controls. However, in addition to an increase in GU malformations, there was also a significant increase in spontaneous abortion among the polydrug/cocaine groups. Moreover, comparing polydrug/cocaine users with polydrug/no cocaine users similarly showed an increased risk of GU malformations. Other comparisons yielded complex effects depending on which groups were compared. Comparing cocaine/polydrng users to drug-free controls revealed only a moderate or "medium mean effect size" on head circumference, gestational age, birth weight, and length. However, comparing polydrug/ cocaine users with polydrug/no cocaine users failed to reveal differences on these same variables. Thus, only GU malformations were consistently found to be associated with maternal cocaine use. None of the comparisons, however, revealed an increased

284 risk o f abruptio placenta, cardiac malformations, or the sudden infant death syndrome, despite the fact that these are among the most commonly cited reproductive and developmental effects o f cocaine. These discrepancies led the authors to suggest that many adverse effects reported to be associated with maternal use of cocaine are the result of the clustering o f other reproductive risk factors in cocaine users that do not occur in nondrng users or even other polydrug abusers. They point out, for example, that comparing cocaine users to nondrug users may fail to control for the lack of prenatal care or the occurrence of sexually transmitted disease. Moreover, comparing those who abuse substances other than cocaine may not be an appropriate comparison because the cocaine users may be consuming disproportionately higher amounts o f alcohol, marijuana, and cigarettes. The four studies reviewed below were published after Lutiger et al. (10) published the meta-anaiysis results and are reviewed here in detail because their collective findings afford some additional information with respect to the developmental toxicity o f cocaine. In one of the best controlled studies, Richardson and Day (18) prospectively studied 34 women who were first interviewed during their fourth prenatal month and again at seven months about their substance use both prior to and during pregnancy. Standard demographic and obstetric information was also recorded. The "cocaine" subjects used light to moderate amounts intranasally during pregnancy and tended to decrease their use of cocaine beyond the first trimester. Their infants were compared with a control group of 600 women who reported no cocaine use during, or in the year prior to, pregnancy. Infant growth parameters, morphology, and behavioral assessment using the Neonatal Behavioral Assessment Scale (NBAS) on postnatal day (PND) 2 was used. This scale evaluates behavior o f the neonate (e.g., habituation, orientation, motor activity, state regulation) in response to the environment. All testing, carried out using "blind" procedures, found no group differences on any of these measures. Neuspiel et al. (16) carried out a study that is of particular interest because of the form o f cocaine predominantly used. The mothers and their infants were selected from a public urban hospital in a poor neighborhood o f the Bronx section of New York City, notorious for the widespread use o f crack cocaine. The study prospectively examined 51 cocaine-exposed infants. Of the women who admitted cocaine use, 72% smoked crack cocaine, 59% used the intranasal route, and 1 subject admitted to IV use. Although frequency of exposure was asked in the interview, these data do not appear in the results. The cocaine-exposed group was compared with 60 unexposed controls that served as the comparison group. Subjects were selected from an initial group of 525 mothers by the use of a structured interview administered by trained staff to determine substance abuse and other pertinent demographic and obstetric information. In some but not all, urine tests were used to classify the mother with respect to cocaine use. Standard growth measures were recorded at birth and the NBAS was administered at 1-3 and 11-30 days of age. At 716 weeks of age, the Nursing Child Assessment of Feeding Scale (NCAFS) was used to evaluate mother-infant interaction during feeding. This scale measures both maternal behavior (e.g., sensitivity to baby's cues, response to distress, fostering of cognitive growth, etc.) and the infant's behavior (e.g., clarity o f cues and responsiveness to the mother). Tests were administered by trained testers blinded to the infants' exposure status.

HUTCHINGS The control group, though unexposed to cocaine, used other compounds; 22% used cigarettes, 10% alcohol, 5% marijuana, and 2% opiates. However, the cocaine users showed significantly greater exposure to these substances; 84% used cigarettes, 49% alcohol, 26% marijuana, and 5% opiates. Compared with the controls, birth weight, length, and head circumference were significantly lower among the cocaineexposed infants, observations that replicated findings of previous work. However, no differences were found at any test age on either the NBAS or NCAFC behavioral measures and it was the NBAS that was so powerfully affected in the early reports and gave rise to the serious concern about permanent CNS damage. Parenthetically, while it is understandable for a neurobehavioral study to use standard criteria to exclude infants with complications that might confound the results (e.g., <2000 g birthweight, congenital malformation, serious illness), it would have been informative to know how many such infants were excluded and whether any of the infants were malformed, especially if any showed GU malformations. The authors summarize their findings by noting that although some of the cocaine-exposed infants showed adverse outcomes at birth, most were relatively healthy, term infants. These growth and behavioral findings were replicated by Coles et al. (4) who studied infants and their mothers selected postpartum to identify those that had used cocaine. The majority of the women admitted to at least weekly use by smoking crack, snorting, and IV administration, and in most cases, multiple forms and routes were common for individual users. Two cocaine groups were studied: Assignment to one cocaine group was made on the basis of any one of the following: (a) maternal report of cocaine use in the last prepartum week, Co) positive maternal urine screen at labor/delivery, and (c) positive infant urine screen. A second cocaine group used cocaine early in pregnancy but then discontinued its use; these mothers and infants had a negative urine screen at recruitment. A n "alcohol" group admitted to heavy use of alcohol during pregnancy, also used cigarettes and marijuana but not cocaine; this was verified by urine screen. For all groups, only full-term healthy infants without medical or genetic problems and mothers without significant medical complications (e.g., sexually transmitted disease) were selected. Based on selfreport, it is important to note that both cocaine groups also used cigarettes and admitted to marijuana and alcohol use. All drug groups reported an average of approximately 90 cigarettes/week. All infants were examined for malformations and none were observed among any of the drug-exposed or control groups. Compared with drug-free control mothers and infants (N = 30) who denied prenatal drug use and were negative upon drug screening, the cocaine neonates had lower birth weight, head circumference, and length. However, the other two drug groups also showed adverse effects on growth, thus it is not clear to what extent cocaine was responsible for these observations. The NBAS administered by testers blinded to exposure status at 2, 14, and 28 days of life failed to reveal any consistent differences between the cocaine-exposed infants and drug-free controls. Although the cocaine-exposed infants did show abnormal reflexes at 28 days and some disturbance of state regulation at 14 and 28 days, all scored within the normal clinical range. Withdrawal symptoms including tremors, hyperacusis, irritability, hypertonicity, etc., were also assessed after each NBAS test and none of the groups differed on this measure. However, duration o f alcohol exposure was significantly related to the number of with-

COCAINE A N D H U M A N P R E G N A N C Y drawal symptoms; there was a similar trend for cocaine exposure that just missed significance ( p < 0.06), suggesting that the longer the conceptus is exposed to drugs a n d / o r the greater the overall exposure, the more likely the infants will show signs o f heightened CNS arousal. Although both Nenspiel et al. (16) and Coles et al. (4) found that cocaine exposure, particularly its use throughout pregnancy, appears to contribute to reductions on birth weight, head circumference, and length, it remains unclear to what extent the concurrent use of alcohol and cigarettes contributed to these effects. With respect to the neurobehavioral f'mdings, Coles et al. (4) pointed out that their study as well as Neuspiel et al. (16) afford some encouragement because they indicate that the prognosis for the cocaine-exposed infant may not be as ominous as originally feared from the initial reports. Of all the issues surrounding the use of cocaine during pregnancy, none is more pressing than whether the exposed children will show neurobehavioral deficits as they mature. Chasnoff et al. (3) reported on growth and development from birth until 2 years of age for groups exposed to polydrug/ cocaine (N = 106), polydrug/no cocaine (N = 45), and no drug controls (N = 81). It is important to note, however, that over the course of the study, a significant number of subjects were lost to follow-up so that by 24 months, the number o f infants in the polydrug/cocalne, polydrug/no cocaine, and no-drug controls were reduced to only 29, 14, and 50, respectively. The drugs most frequently used by the polydrug mothers were alcohol and marijuana. By Week 15 of pregnancy, the mothers were enrolled in a health care program at a university hospital and received both obstetric care and drug counceling. Drug use was documented by use o f random urine toxicology analyses throughout pregnancy and all testing was carried out using "blind" procedures. At birth, compared with the no drug controls, the polydrug/cocaine infants had a significantly lower birth weight, length, and head circumference. Although their weight and length caught up to the controls by 12 months of age, head circumference remained significantly smaller, a difference that persisted until 24 months o f age. The polydrug/no cocaine infants, while showing reductions on all three of these measures at birth, did not differ significantly from controls. However, between 3 and 24 months of age, their head circumference, like the polydrug/ cocaine exposed, was significantly smaller than the controls. Results o f the Bayley Scales of Infant Development administered at regular intervals between 3 and 24 months of age, showed that both polydrug groups scored consistently below the controls at all ages but the differences only reached statistical significance for both groups at 6 months o f age. Although small head size remains a concern, the Bayley scores at 24 months failed to indicate any severe developmental problems and may even allow some cautious optimism that these children may continue to score close to if not within normal limits. Continued follow-up to ages that assess more complex functioning of language and other cognitive processes will reveal whether these infants will develop normally. However, with such a substantial number of subjects lost to follow-up, these can only give us some informations with respect to the better functioning, more motivated mothers who do not leave the program. As for those mothers who leave the study and presumably resume a lifestyle of substance abuse, substantial literature informs us that these children will be at risk for any number o f adverse neurobehavioral outcomes. As we approach nearly a decade of research and clinical

285 experience, what can we conclude about the effects o f cocaine during pregnancy? Based solely on the published evidence, a cautiously objective judgment dictates that definitive answers wait on better controlled, prospective studies. But however flawed the literature, the combined studies along with animal studies o f its pharmacological action, allow for some informed guesses and tentative suggestions. To this observer, the central issues are (a) dose-response effects of cocaine and (b) the possible additive a n d / o r synergistic effects of other drugs, chiefly alcohol, marijuana, and cigarettes. As to the range o f effects that have been described for cocaine during pregnancy, my impression is that the findings across studies that range from no observable effects to the occurence o f GU malformations, appear to represent a crude dose effect relationship. Such an argument is nearly impossible to support because specific information on the precise amount of cocaine used as well as amounts o f other substances is usually unknown. Some minimal information, however, is available so that one can at least speculate as to whether some of the effects reported may be related to the quantity of drug used. A t the low-dose end of the exposure continuum, Richardson and Day (18), characterized the cocaine users in their study as "light" users who averaged 2 to 3 g per month during the first trimester, with a range of 1/30 g used once, to 1 g used dally; the mean level used dropped substantially during the second and third trimesters. This dose, which might arbitrarily be characterized as "recreational" or "occasional use," would appear to be at a "no observable effect level" for growth parameters, morphology, and early neonatal behavior. Chasnoff and co-workers (1,2,11) have consistently reported the most adverse effects for cocaine. Their subjects were drawn from a drug treatment program that treats heavy cocaine users and these women likely represent the high-dose end of the exposure continuum. In the one study in which amount o f cocaine used was described (1), frequency ranged from 2 to 5 times per week and the amount per use ranged from 0.5 to 5.0 g. Moreover, in the group that was studied, 5 of the 23 women also used cocaine IV. These women were thus using in 1 week what the cocaine users studied by Richardson and Day (18) used in 1 month and in the latter study, mean exposure levels diminished in the last two trimesters. Possibly, some of the heavier or "compulsive" users in the Chasnoff samples had high plasma levels of cocaine maintained over relatively prolonged periods in the maternal/ fetoplacental unit. Such a usage pattern may increase the risk of premature labor as well as other effects related to the vasoconstriction of fetal vessels (i.e., GU malformations). This high dose may also be associated with symptoms o f hyperarousal (e.g., tremulousness, irritability) during the early neonatal period but these symptoms, as suggested by Neuspiel and Hamel (15), are more likely the result o f persistent, pharmacologically active levels of cocaine in the neonatal CNS; they do not represent cocaine "withdrawal" in the pharmacological sense of the term. Less well understood is how these effects associated with cocaine might be modified by the presence of other compounds. Many agree that the abuse of cocaine to the exclusion of all other psychoactive drugs is probably so rare that virtually all cocaine users are more accurately characterized as polydrug abusers who also use cocaine. Except for animal studies we are unlikely to ever be sure to what extent particular outcomes may be a result of either the additive or synergistic effects o f cocaine with other compounds. For example, given the heavy use of cigarettes by polydrug/cocalne users would

286

HUTCHINGS

suggest that these individuals, in addition to the other compounds, have increased plasma levels of carbon monoxide along with vasoconstriction from nicotine. These potent physiological effects, known to be associated with fetal hypoxia and growth retardation, are then coupled with the additional vasoactive effects of cocaine. Many would agree that hypertensive/hypoxic episodes in the maternal/fetal placental unit place the fetus at risk for CNS injury. The current status of our knowledge concerning the effects of cocaine during pregnancy is well summarized in a commentary appropriately entitled "The Problem of Cocaine Exposure: A Rush to Judgment" (12). Mayes et ai. voice concern about the premature conclusion that appears to have gained widespread support in both the scientific and lay community that cocaine is specifically responsible for producing serious long-term neurobehaviorai effects in exposed children. (For the results of a survey that confirms the perception of cocaine as a potent teratogen, see ref. 7). They point out the absence of any supporting data from well-controlled prospective studies that examined a large cohort of children and the fragmented literature that does not allow valid predictions as to long-term outcome. Numerous problems that have plagued these studies are n o t e d - small, poorly defined samples, inadequate measures of actual cocaine exposure and the use of other abuse c o m p o u n d s - a n d the complex, intangible role of

the child's family, community, parenting behaviors as well as the contribution of poverty, violence, abandonment, homelessness, multiple short-term foster placements and inadequate or abusive parenting as these might contribute to the overall functioning of the child are mentioned. In particular, they lament the unfair labeling of these children, enabling society and especially schools to stigmatize them as brain-damaged, handicapped, uncontrollable, and unedueable. SUMMARY The research evidence appears to suggest that cocaine is a relatively weak teratogen that increases the risk of GU and possibly other vascular malformations. These effects occur in a relatively small group of individuals who may have a unique but as yet undefined genetic and/or physiological susceptibility and who use cocaine by any number of routes but in high doses, nearly daily, or during episodic binges. To what extent the occurrence of these adverse effects are dependent on or modified by the concurrent use of other compounds remains unknown. ACKNOWLEDGEMENTS This research was supported in part by Grants DA03544 and DA07822 from the National Institute on Drug Abuse.

REFERENCES 1. Chasnoff, I. J.; Burns, W. J.; Schnoll, S. H.; Burns, K. A. Cocaine use in pregnancy. N. Engl. J. Med. 313:666-669; 1985. 2. Chasnoff, I. J.; Bussey, M. E.; Savich, R.; Stack, C. M. Perinatal cerebral infarction and maternal cocaine use. J. Pediat. 108:456459; 1986. 3. Chasnoff, I. J.; Griffith, D. R.; Freier, C.; Murry, J. Cocaine/ polydrug use in pregnancy: Two-year follow-up. Pedlar. 89:284289; 1992. 4. Coles, C. D.; Platzman, K. A.; Smith, I.; James, M. E.; Falek, A. Effects of cocaine and alcohol use in pregnancy on neonatal growth and neurobehavioral status. Neurotoxicol. Teratol. 14: 23-34; 1992. 5. Dow-Edwards, D. L. Cocaine effects on fetal development: A comparison of clinical and animal research findings. Neurotoxicol. Teratol. 13:347-352; 1991. 6. Johnson, H. Sleepwalking through history: America in the Reagan years. New York: Norton; 1991. 7. Koren, G.; Gladstone, D.; Robeson, C.; Robieux, I. The perception of teratogenic risk of cocaine. Teratology 46:567-571; 1992. 8. Koren, G.; Shear, H.; Graham, K.; Einarson, T. Bias against the null hypthesis: The reproductive hazards of cocaine. Lancet 2: 1440-1442; 1989. 9. Larson, C. S. Overview of state legislative and judicial responses. The Future of Children 1:72-84; 1991. 10. Lutiger, B.; Graham, K.; Einarson, T. R.; Koren, G. Relationship between gestational cocaine use and pregnancy outcome: A meta-analysis. Teratology 44:405--414; 1991. 11. MacGregor, S. N.; Keith, L. G.; Chasnoff, I. J.; Rosner, M. A.;

12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

Chisum, G. M.; Shaw, P.; Minogne, J. P. Cocaine use during pregnancy: Adverse perinatal outcome. Am. J. Obstet. Gynecol. 157:686-690; 1987. Mayes, L. C.; Granger, R. H.; Bornstein, M. H.; Zuckerman, B. The problem of prenatal cocaine exposure: A rush to judgement. J. Am. Mud. Assoc. 267:406--408; 1992. Moore, T. R.; Sorg J.; Miller, L.; Key, T. C.; Resnick, R. Hemodynamic effects of intravenous cocaine on the pregnant ewe and fetus. Am. J. Obstet. Gynecol. 155:883-888; 1986. Musto, D. F. Evolution of American attitudes toward substance abuse. N.Y. Acad. Sci. 562:3-7; 1989. Neuspiel, D. R.; Hamel, S. C. Cocaine and infant behavior. Dee. Behav. Pedlar. 12:55-64; 1991. Neuspiel, D. R.; Hamel, S. C.; Hochberg, E.; Greene, J.; Campbell, D. Maternal cocaine use and infant behavior. Neurotoxicol. Teratol. 13:229-233; 1991. Paltrow, L. M. Perspective of a reproductive rights attorney. The Future of Children 1:85-92; 1991. Richardson, G. A.; Day, N. L. Maternal and neonatal effects of moderate cocaine use during pregnancy. Neurotoxicol. Teratol. 13:455-460; 1991. Slutsker, L. Risk associated with cocaine use during pregnancy. Obstet. Gynecol. 79:778-789; 1992. Volpe, J. J. Effects of cocaine use on the fetus. N. Eng. J. Med. 327:399-407; 1992. Woods J. R.; Plessinger, M. A.; Scott K.; Miller, R. K. Prenatal cocaine exposure: A sheep model for cardiovascular evaluation. Ann. N.Y. Acad. Sci. 562:267-279; 1989.