Original Articles
Estimates of illicit drug use during pregnancy by maternal interview, hair analysis, and meconium analysis*,**

https://doi.org/10.1067/mpd.2001.111429Get rights and content

Abstract

Objective: To compare the sensitivity and specificity of maternal interview, maternal hair analysis, and meconium analysis in detecting perinatal exposure to cocaine, opiate, and cannabinoid. Design/methods: The use of cocaine, opiate, and cannabinoid during pregnancy was determined prospectively in 58 women by 3 methods: structured maternal interview, maternal hair analysis, and meconium analyses. The results of the 3 methods were compared with one another. Results: The maternal interview showed the lowest sensitivity in detecting cocaine and opiate exposures (65% and 67%, respectively), but it had the highest sensitivity in detecting cannabinoid exposure (58%). Both hair and meconium analyses had high sensitivity for detecting cocaine or opiate exposures. Hair analysis had a sensitivity of 100% for cocaine and 80% for opiate detection. However, it had a false-positive rate of 13% for cocaine and 20% for opiate, probably as a result of passive exposure. Meconium analysis had a sensitivity of 87% for cocaine and 77% for opiate detection, but unlike hair analysis, it had no false-positive test results for cocaine. Both hair and meconium analyses had low sensitivity in detecting cannabinoid exposure (21%-22.7%), most probably because of the sporadic use of cannabinoid. Conclusion: Meconium and hair analyses had the highest sensitivities for detecting perinatal use of cocaine and opiate, but not for cannabinoid. The principal drawback of hair analysis is its potential for false-positive test results associated with passive exposure to drugs. Maternal interview is a time-consuming test of low sensitivity. The high sensitivity of meconium analysis and the ease of collection make this test ideal for perinatal drug screening. (J Pediatr 2001;138:344-8)

Section snippets

Methods

All women who initially presented for prenatal care at the Hutzel Hospital antenatal high risk clinic were routinely interviewed regarding drug and alcohol use. From this population, 75 pregnant women were enrolled in the study based on their use (or non-use) of cocaine, opiate, and cannabinoid. By power analysis (see below), a total of 40 subjects were required for the study, but an additional 35 subjects were also recruited to compensate for possible dropouts. Informed consent was obtained at

Results

A total of 75 subjects were initially enrolled in the study. Nine were withdrawn because of non-return to the clinic (n = 1) or voluntary withdrawal from the study (n = 8). Of the remaining 66 subjects, 6 had voluntary or spontaneous abortions and 2 were further excluded: 1 infant had inappropriate collection and storage of meconium and 1 infant was missed at delivery. A total of 58 mothers and 59 infants (one set of twins in the control group) were included in the study. A mean of 2.8 ± 0.4

Discussion

We compared the sensitivity, specificity, and limitations of maternal interview, hair analysis, and meconium analysis for the detection of antenatal exposure to cocaine, opiate, and cannabinoid. Maternal interview is probably most prone to error because of potential denial by the subject of any drug use. Thus to increase the accuracy of the maternal interview, the study was started as early in gestation as possible to allow for repeated interviews (mean = 7.4 interviews per subject), the

References (18)

There are more references available in the full text version of this article.

Cited by (138)

  • Drug testing in support of the diagnosis of neonatal abstinence syndrome: The current situation

    2023, Clinical Biochemistry
    Citation Excerpt :

    Therefore, any positive drug findings in the meconium sample could potentially reflect drugs administered to the mother during labor and delivery or to the infant post-delivery but prior to sample collection [37,69]. There are numerous reports of the specificity and sensitivity for different analytes using different testing methodologies [39,41,70–72]. It should be understood that the values for sensitivity and specificity differ depending on the prevalence of drug use/abuse, or pre-test probability.

View all citing articles on Scopus
*

Supported in part by National Institute on Drug Abuse grant 1 RO1 DA06821-01A1.

**

Reprint requests: Enrique M. Ostrea, Jr, MD, The Department of Pediatrics, Hutzel Hospital, 4707 St Antoine Blvd, Detroit, MI 48201.

View full text