Intrauterine Cocaine, Lead, and Nicotine Exposure and Fetal ... | NCBI

Intrauterine Cocaine, Lead, and Nicotine Exposure and Fetal Growth Daniel R Neuspiel MD, MPH, Morn Markowitz, MD, and Ernest Drucker, PhD

Introduction Gestational cocaine exposure has been associated with fetal growth1 and infant behavioral deficits,2 yet possible confounding from other risk factors3'4 has not been fully explored. Tobacco and lead have also been associated with impaired fetal growth and brain development. Cocaine users may be exposed to lead in older substandard housing associated with poverty and may have both active and passive tobacco exposure. Further, tobacco and lead may interact: active smokers57 and their children-11 have higher lead levels, with a dose-response gradient between parental smoking and child lead levelslZl3; newborn lead levels have been correlated with the duration of maternal smoking.14 This investigation hypothesized that cocaine-exposed newborns have more lead and tobacco exposure and that cocaine-associated effects on the growth of newborns are partly explained by lead and/or tobacco exposure.

Methods Over 2 months in 1992, parturient women in a public hospital provided urine samples for routine clinical testing, the discarded portions of which were separated in two for drug and cotinine measurement. At delivery, umbilical cord blood was placed in a heparinized leadfree tube for lead assay. After study staff obtained clinical information from hospital records, the study urine and blood samples were given a number code and other identifiers were removed. Urine samples were tested in women proceeding to singleton deliveries. When more than one urine sample was provided, the earliest sample was used. One urine sample was assayed for cocaine, benzoylecgonine, opiates, methadone, cannabinoids, and amphetamine by the EMIT method.15 Two randomly selected comparison dyads with negative toxicology were chosen for each drugpositive case. All women giving birth were routinely asked about cocaine and tobacco use during pregnancy by clinical

staff, independent of this study. The other study urine sample was assayed for cotinine (a metabolite of nicotine) by radioimmunoassay, and creatinine was determined to correct for variation in urine

volume.16 Whole blood lead was measured by atomic absorption spectrophotometry in a laboratory participating in proficiency testing programs of the Centers for Disease Control and New York State. The 95% confidence limit for blood lead is ±1 xg/dL.17 Statistical analysis was done with SYSTAT software.18 This investigation was approved by the Institutional Review Board of Montefiore Medical Center and by the New York City Health and Hospitals Corporation. Because study data were collected anonymously, informed consent was waived.

Resuids Discard urine specimens were obtained from 237 (41%) of the 581 mothers having singleton deliveries during the study period. Cocaine-positive cases were identified from study assays (10 of 237 tested), clinical tests (7 additional cases of 47 tested), or maternal history (1 additional case). Five case subjects with study urine samples positive only for drugs other than cocaine were excluded. Fortysix comparison subjects were chosen from 220 dyads with both negative study toxicology and no self-report of drug use. Of 18 cocaine-positive dyads, 6 were tested for both cotinine and lead, 6 only for cotinine, and 6 only for lead. Of 46 cocainenegative dyads, 20 had both lead and cotinine levels, 13 had only cotinine, and Daniel R. Neuspiel is with the Department of Pediatrics and the Department of Epidemiology and Social Medicine, Mom Markowitz is with the Department of Pediatrics, and Ernest Drucker is with the Department of Epidemiology and Social Medicine, Albert Einstein College of Medicine and Montefiore Medical Center, Bronx, NY. Requests for reprints should be sent to Daniel R. Neuspiel, MD, MPH, NCBH 4M-08, 3424 Kossuth Ave, Bronx, NY 10467. This paper was accepted February 23, 1994.

September 1994, Vol. 84, No. 9

Public Health Briefs 12 had only lead. There were no differences between the subjects with and without anonymous urine samples either in the proportion of infants clinically tested for drugs (6% and 9%, respectively) or in the proportion of clinical tests that were positive for cocaine (29% and 32%, respectively); the chi-square test was nonsignificant for each. Demographic, pregnancy, and newborn information is in Table 1. Of 64 mothers, 56% were Hispanic, 31% African-American, 5% White, 6% Asian, and 2% other ethnicity. The mean lag time from urine collection to testing was similar in both exposure groups. Lead and cotinine levels (Table 2) were higher in the cocaine group, in which cotinine levels were elevated both among cigarette smokers and nonsmokers. Lead levels differed between admitted smokers (mean = 5.2 ,g/dL; 95% confidence interval [CI] = 4.5, 5.9) and nonsmokers (mean = 3.5 p,g/dL; 95% CI = 3.0, 4.0; P = .001). Multiple linear regression (Table 3) was used to evaluate the independent effects of cocaine, lead, and cotinine on fetal growth after control for gestational age. The apparent effects of cocaine on birthweight and head circumference were diminished after consideration of cotinine levels; smoking history explained fewer cocaine effects than cotinine and was a weaker confounder. Lead exposure explained some of the cocaine-related decrement in all three growth measures, but it had less effect than cotinine on weight and head circumference. Separate models demonstrated no confounding of cocaine effects on fetal growth by race, maternal age, gravidity, or parity. Nine regressions were conducted with each of the three dependent variables.

Discussion Many methodologic difficulties exist in studying cocaine effects on the fetus.2 The effects of concurrent exposures are of particular importance because cocaine users share many other potential perinatal risk factors. Cocaine-exposed newborns may be exposed to more ambient lead. Tobacco exposure may also raise lead levels by small-airway damage and more hand-tomouth behavior.13 Nutritional factors may predispose cigarette smokers and cocaine users to higher lead levels via inadequate iron or calcium intake. In addition, crack smoking increases lung permeability due to chronic epithelial damage,19 which may September 1994, Vol. 84, No. 9

TABLE 1-Characteristics of Mothers and Newborns, by Cocaine Exposure

Group Cocaine Exposed

(n 18) =

Matemal age, y

Gravidity Previous live births Involuntary pregnancy terminations Elective pregnancy terminations Maternal race, % Black Smoking report this pregnancy, % Female births, % Gestational age, wk Birthweight, g Length, cm Head circumference, cm

Comparison (n = 46) 95% Cl Mean

P

Mean

95% Cl

30.5 5.6 1.8 0.9

28.6,32.4 4.2,7.0 1.2,2.4 0.4, 1.5

23.8 2.7 0.9 0.3

22.3,25.3 2.2,3.2 0.6,1.2 0.1,0.5