Cocaine use by pregnant women in Toronto

Gideon Koren, MD, FRCPC

September, 1996

Canadians often take pride in the fact that our better health care system protects us from the social and health problems in the United States. During the last decade, the issue of fetal exposure to drugs raised serious concerns about the welfare of a generation of unborn babies. Because most of the data came from the United States, Canadians assumed they did not have the problem. However, new research from Motherisk dispels this myth and provides an alarming perspective on the Canadian situation.

For centuries, the placenta was assumed to serve as a barrier, preventing fetal exposure to foreign agents in the mother's circulatory system. The thalidomide disaster more than 30 years ago painfully showed that a therapeutic compound that did not harm mothers crossed the placenta and could damage fetuses. This marked the beginning of a focus on the adverse effects of various factors on fetal growth and development.

In addressing the potential effects of foreign agents on developing fetuses, it is important to estimate the extent of fetal exposure in terms of dose and time. Most commonly, researchers obtains information on chemicals the mother consumed during pregnancy by interviewing her during pregnancy or after birth or by reviewing medical charts. However, our studies have shown that women giving birth to children with medical problems tend to under-report the amount of alcohol they consumed compared with reports they gave during pregnancy.¹ Also, we have recently shown that, while nonpregnant smokers have excellent correlation between their reported amount of smoking and concentrations of nicotine in their hair, no such relationship is found among pregnant smokers.² This discrepancy comes from inaccurate reporting by mothers, probably stemming from guilt and fear.

A common way to estimate the extent of fetal exposure to a drug is by measuring its concentration in umbilical cord blood at birth. The problem is that drug concentrations in umbilical blood reflect only very recent exposure, and absence of drug does not rule out fetal exposure earlier in the pregnancy. Measurement of drugs in neonatal urine might reflect accumulation in the last few days of pregnancy, but not necessarily before. It has recently been shown that drugs can be found in meconium, a substance that accumulates a variety of substances starting around 16 weeks of pregnancy. However, the implications of different meconium drug concentrations on fetal outcome are poorly understood.

In 1988, with increasing evidence that we could measure accumulation of foreign substances in adult hair to estimate systemic exposure, we hypothesized that we could do the same with fetal hair. Because neonatal hair grows during the last 3 to 4 months of pregnancy, we assumed that agents in the fetal circulation would be deposited in fetal hair. Our work was preceded by a study by Marsh et al,³ which showed a clinically important correlation between concentrations of mercury in maternal hair and fetal damage.³

During the last 2 decades, we have witnessed a dramatic increase in cocaine use by women of reproductive age (along with increased use in the general population). Scores of epidemiologic studies have suggested that cocaine use by pregnant women is associated with a multitude of health issues, such as prematurity, intrauterine growth retardation and death, congenital malformations, placental abruption, and developmental delay. A serious issue in interpreting these findings is the coexistence of other health hazards in these women, including cigarette and alcohol consumption, poor prenatal care, sexually transmitted diseases, and lower socioeconomic class.

In 1989, we examined concentrations of a metabolite of cocaine in the hair of seven newborns whose mothers were known to be cocaine users and found that we could measure this metabolite in the hair. In the same study, we found that the hair of 16 adult cocaine users tested positive for the metabolite of cocaine while urine screening test results were negative.⁴ None of the non-user control subjects tested positive in the hair analysis.

Our results were published along with an editorial questioning how one could separate systemic exposure to cocaine or its metabolite, as confirmed by hair analysis, from external contamination of hair by "crack" cocaine smokers. To address this, we showed that exposure to burning crack results in accumulation of cocaine but not its metabolite in hair, whereas cocaine use leaves the drug and the metabolite in hair. External contamination with crack smoke can be washed off; systemic exposure cannot. For unborn babies, purely external contamination is unlikely anyway, because fetal hair is in contact with amniotic fluid, which is swallowed by unborn babies.

From 1990 to 1992, we measured fetal exposure to cocaine in 600 babies born in three nurseries in Toronto using positive maternal history, the urine test for the metabolite of cocaine, and hair analysis.^5,6 In all, 37 babies (6.25%) tested positive for cocaine. The hair test detected 33 cases but failed to identify four babies who had low concentrations of cocaine metabolites in their urine. The urine test failed to identify 76% of cases. In downtown Toronto, the overall rate of fetal exposure to cocaine was 12.5% (25/200), significantly higher than in the two suburban nurseries (3%; 12/400). Our figures imply that more than 5000 babies a year in the greater Toronto area are cared for postnatally by mothers regularly using cocaine.

Comparison of cocaine-positive and cocaine-negative mother-infant pairs revealed similar maternal age, racial distribution, and obstetric history. However, mothers who used cocaine had a much higher risk of vaginal bleeding (16% versus 6%) and of being hepatitis B carriers (25% versus 8%). Cocaine-using mothers were also much more likely to smoke cigarettes (29% versus 10%). Infants exposed to cocaine in utero had lower average birth weights (by 100 to 200 g) and birth lengths. But further examination of babies exposed to cocaine and to maternal tobacco smoking accounted for most of this variation. Babies exposed to cocaine in utero were more likely to need initial medical support or resuscitation (52% versus 30%). Because fetal exposure to cocaine is usually not known to treating physicians, this difference probably reflects a genuine biological phenomenon and not a bias.

Our study indicates that exposure to cocaine, determined by hair analysis, is associated with substantial perinatal risk. Some risks probably are caused by a clustering of other factors, such as maternal smoking and hepatitis B carrier state. If cocaine exposure were not detected through hair tests, most of these cocaine-exposed infants would not only have perinatal complications, but would also go home to be cared for by drug-dependent mothers, further increasing their health risk. Ample evidence shows that drug-dependent women find it difficult to provide optimal conditions for their infants.

Unless this epidemic of prenatal cocaine exposure is addressed medically and socially, we should anticipate great increases in the rates of long-term physical and neurodevelopmental complications in large numbers of Canadian infants and children.

References

1. Feldman Y, Koren G, Mattice D, Shear H, Pellegrini E, MacLeod SM. Determinants of recall and recall bias in studying drug and chemical exposure in pregnancy. Teratology 1989; 40:37-46.
2. Eliopoulos C, Klein J, Phan MK, Knie B, Greenwald M, Chitayat D, et al. Hair concentrations of nicotine and cotinine in women and their newborn infants. JAMA 1994;271:621-3.
3. Marsh DO, Myers GI, Clarkson TW, Ami Zaki L, Tikriti S, Majeed MA, et al. Dose-response relationship for human fetal exposure to methylmercury. Clin Toxicol 1981;18:1311-8.
4. Graham K, Koren G, Klein J, Schneiderman J. Detecting gestational exposure to cocaine by hair analysis. JAMA 1989;262:3328-30.
5. Forman R, Klein J, Meta D, Barks J, Greenwald M, Koren G. Prevalence of fetal exposure to cocaine in Toronto 1990-1991. Clin Invest Med 1994; 17:206-11.
6. Forman R, Klein J, Meta D, Barks J, Greenwald M, Koren G. Maternal and neonatal characteristics following exposure to cocaine in Toronto. Repro Toxicol 1993; 7:619-22.