Article Text

Download PDFPDF
In utero drugs of abuse exposure testing for newborn twins
  1. Ping Wang,
  2. Claudia P Molina,
  3. Joyce E Maldonado,
  4. David W Bernard
  1. Department of Pathology, The Methodist Hospital System, Houston, Texas, USA
  1. Correspondence to Ping Wang, The Methodist Hospital, 6565 Fannin St. MS 205, Houston, TX 77030, USA; pwang{at}


This report describes testing of a case of in utero drugs of abuse exposure in which discordant results were seen between urine and meconium, and between twin meconium samples. The discordance between urine and meconium could be explained by the differences in detection window, threshold concentration and screening technology, and the discordance between dizygotic twin meconium samples could be explained by the differences in drug diffusion and placental and fetal biotransformation of drugs. The meconium sample of one twin screened negative for benzodiazepines was reported positive in the confirmation assay with higher sensitivity and a lower cut-off concentration. Negative screening results of drugs of abuse should be interpreted with caution, taking into account matrix type, reactivity of drugs in the assay and cut-off concentration. If screening results are inconsistent with each other or with the clinical scenario, confirmation testing using more sensitive and specific methods with lower cut-offs is warranted.

  • In utero
  • drugs of abuse
  • twin
  • urine
  • meconium
  • drug abuse
  • toxicology

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Case report

A patient, gravida 3, para 0, aborta 1 was admitted in preterm labour with approximately 26 and 30 weeks' twin gestation as assessed by ultrasound.

The patient had had an abortion 1½ years previously. She also had a history of seizure disorder but was non-compliant with medication. A history of polyintoxication with alprazolam, marijuana, cocaine, nicotine and opiate was noted and it was unclear if the seizure disorder was related to the substance abuse. She denied any history of alcohol abuse. Family history was remarkable for leukaemia and alcoholic liver disease in the patient's mother.

On admission the patient was afebrile. Blood pressure was 141/89 mm Hg, pulse 69/min, and respirations 20/min. Initial laboratory data showed leucocytosis with neutrophilia. Urinalysis was unremarkable. Maternal urine toxicology testing was performed. Positive screening results were reported for benzodiazepines, THC and cocaine (see table 1).

Table 1

Urine drug testing results with cut-off levels*

The patient was taken to the operation room for caesarean section. Twin A in vertex presentation was delivered first and weighed 2 pounds 10 ounces, with Apgar score 8 and 9. Twin B in frank breech presentation was delivered 1 min later and weighed 3 pounds 7 ounces, with Apgar score 8 and 9. Two separate diamniotic dichorionic placentas were delivered uneventfully.

Urine samples from twins A and B were collected 14 and 11 h after delivery, respectively. Meconium samples from both twins were collected 4 days 13 h and 2 days 11 h after delivery, respectively. Both urine and meconium were tested for evidence of intrauterine drug exposure. Urine toxicology screening for both twins was reported as negative (see table 1). However, meconium testing showed the presence of cocaine and cannabinoid metabolites in both twins (see table 2). Benzodiazepines were reported in meconium of twin B but not twin A. The clinician caring for the patient enlisted help from the pathologists to interpret the discordant results between the twins and between the urine and meconium samples.

Table 2

Meconium drug testing results with cut-off levels

As part of the investigation process, twin A meconium was tested for benzodiazepines using the confirmatory liquid chromatography/mass spectrometry/mass spectrometry (LC-MS/MS) assay. The sample was reported to contain 70 ng/g nordiazepam, 76 ng/g oxazepam and 93 ng/g temazepam.

The urine drug screening for the mother and both twins was performed in the Clinical Chemistry Laboratory at San Jacinto Methodist Hospital using the V-Twin Drug Testing System (Siemens Healthcare Diagnostics, Newark, Delaware, USA). The analyser runs liquid ready-to-use reagents (EMI II Plus assays) through an enzyme immunoassay for the analysis of each specific compound in urine. Table 1 lists the cut-off concentrations. The coefficients of variation (CVs) at the cut-off concentrations were 1–3%. The sensitivity levels for benzodiazepines, cannabinoids and cocaine were 15 ng/ml, 15 ng/ml and 35 ng/ml, respectively. All meconium screening and confirmation tests were sent out to ARUP Laboratories (Salt Lake City, Utah, USA). Screening was done using ELISA with cut-off concentrations listed in table 2 under ‘initial test’. The CVs of screening assays at the cut-off concentrations were 6–10%. Benzodiazepines and cannabinoids were confirmed using LC-MS/MS and cocaine was confirmed using gas chromatography/mass spectrometry with cut-off concentrations listed in table 2 under ‘confirmation’. The sensitivity levels of the confirmation assays for benzodiazepines, cannabinoids and cocaine were 10 ng/ml, 2.5 ng/ml and 10 ng/ml, respectively. The CVs at the cut-off concentrations were 2–4%.


Matrices and assays for in utero drugs of abuse exposure testing

In utero drugs of abuse exposure may adversely affect not only the development of the fetus but also the long-term physiology and neurocognitive function of the child in later stages of life. There are increased risks of obstetrical complications such as placental abruption and premature birth among drug abusing mothers. Withdrawal symptoms that require immediate medical attention may also occur in the mother and the newborn infant. Detecting in utero drugs of abuse exposure is important in preparing medical and social resources to care for the mother and the infant.

The extent of fetal exposure to drugs of abuse is dependent on the chemical nature of the drugs and the physiology of the maternal–fetal complex. Drugs cross the placenta primarily by passive diffusion and less by active transport and pinocytosis.1 Drugs that are small in molecular size, lipophilic and unionised in blood can diffuse across the membranes and gain access to fetal blood. Placental blood flow and permeability also affect drug concentrations in the uterus. There is evidence that drugs undergo biotransformation in the placenta.2 It has been shown that human placenta has butyrylcholinesterase that metabolises cocaine.3 Placenta also expresses other important metabolising enzymes such as cytochrome P450s.4 The fetus may also metabolise some drugs especially during the later stages of gestation.

Different maternal and fetal biological matrices have been employed to detect in utero drug exposure.5 For the mother, urine is the most frequently used material. The detection window of urine testing is usually 1–3 days, except in chronic cannabis users, for whom the detection window is longer because of chronic drug deposition within the body. In the case presented, maternal urine testing results indicated ingestion of benzodiazepines, cannabinoids and cocaine within several days before delivery. Maternal blood and oral fluid are of limited use as they only reflect acute drug use. Sweat also has a detection window of days but the ease of collection varies among individuals. Hair is a matrix that provides drug use history up to months or years before collection but requires caution for environmental contamination,6 colour bias in the deposition of drugs7 and the use of chemicals on hair that may interfere with drug analysis.

Neonatal urine was traditionally the specimen of choice for newborn drug testing, but only provides maternal drug use history up to a few days prior to delivery, if it is collected soon after delivery. Another disadvantage is the difficulty of collection. Meconium, on the other hand, is an easy-to-collect material that starts to form around the 12th week of gestation. It is the first stool that is usually passed in the first 1–3 days of life. Meconium is composed of epithelial cells, lipids, mucopolysaccharides, bile acids, bile salts and water. Drugs of abuse testing results using meconium reflect intrauterine drug exposure over several months before birth, and therefore provide a longer history than neonatal urine. The timing of collection is also less of an issue for meconium, as compared to urine.

Meconium analysis requires additional weighing and extraction steps compared to urine, and cannot be accomplished by simply adopting urine assays without further validation. Some studies suggest that meconium testing is more sensitive than urine testing. However, this discrepancy can usually be eliminated by using more sensitive urine analysis methods and lower urine cut-offs.8 In the case presented, the urine testing results were negative while meconium testing was positive for the twins. This is likely due to the higher cut-offs used in the urine screening assay and different screening technologies (see cut-off concentrations and footnotes in tables 1 and 2).

Other matrices that are mostly used in research settings are neonatal hair, amniotic fluid, umbilical cord blood, cord tissue, neonatal fingernails, toenails and vernix. These are used infrequently in clinical setting because of difficulty or invasiveness of collection, limited detection window or lack of well developed clinical protocols.

Screening of urine or meconium is usually carried out using enzyme multiplied immunoassay, fluorescence polarisation immunoassay, radioimmunoassay or ELISA. These assays are usually performed on automatic platforms, and use antibodies that have reactivity for multiple drugs in the same class. For example, amphetamine immunoassays can detect both amphetamine and methylamphetamine; opiate immunoassays react with codeine, morphine, hydrocodone and hydromorphone. Therefore screening assays produce quick turnaround of results without sophisticated technical requirements. This can be especially helpful in emergency settings. However, screening assays usually have the common limitations of relatively low sensitivity, low specificity and low reactivity with certain drugs in a drug class compared to confirmation methods. Confirmation assays have higher sensitivity and specificity and are based on a different analytical principle, such as gas chromatography/mass spectrometry, liquid chromatography/mass spectrometry or tandem mass spectrometry. The practice of confirming positive screening tests is helpful in eliminating false positives, but does not address false negatives caused by low sensitivity and low reactivity with certain drugs. In the case presented, twin A meconium was negative for benzodiazepines in the screening assay with a 75 ng/g cut-off. However, when tested in the confirmation LC-MS/MS assay with a 20 ng/g cut-off, the same sample was shown to contain 70 ng/g nordiazepam, 76 ng/g oxazepam and 93 ng/g temazepam. The concentration of the benzodiazepines in this specimen was not high enough to trigger a positive response in the screening assay, likely due to the lower reactivity of oxazepam and temazepam with the screening assay reagents (oxazepam and temazepam have reactivity of 67% and 55%, respectively, compared to nordiazepam9). Clinicians should be aware of the limitations of screening assays and treat negative screening results with caution. If the screening results are inconsistent with other testing results and/or clinical scenario (in this case the positive results in maternal urine and twin B meconium), confirmation testing with higher sensitivity and lower cut-off should be conducted despite the negative screening results.

Discrepancy between in utero drugs of abuse testing results of twins

Twin pregnancies constitute a unique paradigm in intrauterine drug exposure assessment. Monozygotic twins share the same placenta and have identical genetic makeup. In utero drug exposure of monozygotic twins should therefore be identical or very similar. In contrast, dizygotic twins have separate placentas and different genetic makeup. The diffusion, biotransformation of drugs in placenta and fetal drug metabolism can be quite different between the twins, which may lead to the varying concentrations of drugs of abuse in dizygotic twins. Indeed, it was found in one report that out of seven pairs of twins who were born to cocaine or cannabinoids abusing mothers, only one monozygotic twin pair had identical levels of cocaine in hair. The other six pairs of dizygotic twins all showed large disparities in either hair or meconium drug concentrations. In three dizygotic pairs, levels of cocaine or cannabinoids were undetectable in one twin while positive in the other.10 Timing of sample collection should also be considered when interpreting results from twins. Urine testing results are more sensitive to the timing of collection, and therefore samples of twins should be collected as closely together as possible and soon after delivery.

In the case presented here, the dizygotic twins showed different concentrations of cannabinoids and cocaine in the meconium. Benzodiazepines were undetectable by the screening assay in twin A but were positive in twin B. The negative screening result in twin A did not indicate absence of in utero drug exposure, which became apparent when the sample was tested by the confirmation assay. The discrepancy between the testing results of the twins is consistent with the dizygotic nature.

Case follow-up

The mother was discharged on the third hospital day. The twins were discharged 2 days later after an uneventful course and under child protection services.

Take-home messages

  • In utero drugs of abuse exposure testing has important medical and public health implications.

  • Various maternal and neonatal biological matrices can be used for in utero drugs of abuse testing. These matrices provide different windows of detection.

  • Drugs of abuse confirmation testing methods have higher sensitivity and specificity, and lower cut-off concentration compared to screening assays. Negative screening results may not exclude the presence of the drugs of abuse; therefore confirmation testing is warranted for samples with negative screening results which are inconsistent with other testing results and/or clinical scenario.

  • The placenta and the fetus have the capacity to metabolise drugs. Drug exposure in dizygotic twins can be different because of differences in drug diffusion and placental and fetal biotransformation of drugs.



  • Competing interests None.

  • Patient consent Not received, as the patient has been lost to follow-up.

  • Provenance and peer review Not commissioned; externally peer reviewed.