Perinatal Effects of Combined Use of Heroin, Methadone, and Amphetamine during Pregnancy and Quantitative Measurement of Metabolites in Hair

Objective:There has been very limited research on the clinical features of newborns exposedto combined use of heroin, methadone, and amphetamine in the uterus. We describe a technique for the quantification of drug metabolites in neonatal hair samples.Methods:In a tertiary neonatal care center in Taiwan, three neonates whose mothers selfreported heroin abuse with methadone treatment during pregnancy were studied. Involuntaryexposure to amphetamine was not suspected before the births. To assess long-term illicit drugexposure during pregnancy, a quantifying technique of gas chromatography/mass spectrometry (GC/MS) for hair samples from neonates was developed to replace current methods forurine and blood specimens.

Results:All three mothers were addicted to heroin and prescribed oral methadone treatmentduring pregnancy. Two males and one female were born and then admitted to the neonatalintensive care unit because of apparent neonatal abstinence syndrome (NAS) after birth. Additional hypertonicity and cerebral dysfunction were also diagnosed by electroencephalographyin one case. Supportive care was given to the neonates, unless special treatments were neededin responding to tachypnea, fetal distress, or withdrawal symptoms. During follow-up periodsfrom 10 months to 15 months, the signs of NAS remained and delays in milestones of development were observed. Further follow-up on the infants’ neurobehavioral developmentis necessary. Measurement results of neonates’ hair samples revealed high levels of metabolites of heroin, methadone, and amphetamine, reflecting the amount of illicit drug exposure2e3 months before delivery.

Conclusion:The current study suggested the possibility of polydrug exposure, which was previously unknown in pregnant women in Taiwan. Measurement of neonatal hair samples couldprovide a basis for clinical evaluation and potential corresponding treatment.

Copyright©2012, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. All rightsreserved.

1. Introduction

As suggested in the 2002/2003 US National Survey on Drug Use and Health, the prevalence of suffering from at leastone substance abuse in a year was approximately 11% among adults. In addition, 4.3% of pregnant women aged 15-44 years used illicit drugs, resulting in 172,934 birthscomplicated by maternal use of illicit drugs in the USA.Approximately seven per 1000 adults aged 15-54 years were heroin dependent in Australia, the UK, and Europe.Among them, around 80-90% females were of reproductiveage.The babies of mothers with opiate dependence duringpregnancy have increased risks of bloodborne virus infections, including hepatitis B, hepatitis C, and HIV.Thus,substance abuse has been listed as one of the 10 highpriority public health issues for the US population inHealthy People 2010.

Fetuses exposed to heroin in the uterus have a fluctuating cycle of intoxication and withdrawal after birth.Among the adverse effects of heroin, NAS is the mostdevastating clinical finding, including the symptoms oftremor, fever, seizure, inability to self-quiet, elevatedstartle reflex, feeding difficulty, abnormal sleep patterns,and diarrhea. Among neonates who are exposed to opiatesin the uterus, 60-90% experience some degree of NAS.Pregnancy complications following substance use decreasedevelopment indicators, elevate central nervous systemand autonomic nervous system signs, and increase developmental and behavioral problems when the affectedbabies are growing up. Polydrug use, defined by the use ofmore than one non-prescribed licit or illicit substanceeither concurrently or simultaneously, worsens the situation in clinical management.

Methadone, an artificial opiate derivative, has beenclinically recommended for the treatment of heroindependency to pregnant women since the 1970s.5Methadone maintenance treatment gives the benefits of betterbirth outcomes of neonates, reduction of heroin use duringpregnancy, and improvement of overall maternal health,although the side effects of reduced fetal activities, lowered birth weight, and prolonged NAS symptoms exist.Physical changes during pregnancy complicate the clinicalmanagement protocol, and make it more challenging tomaintain effective plasma concentration but, in the meantime, minimize the chance of NAS occurring.10However,debates on how to minimize added maternal methadonedosage in order to sustain plasma concentration whilerestricting its impact on mothers and neonates areongoing.

Since 2000, opiates (including heroin) have become thetop illicit type of drug abused in Taiwan, contributing 70.3%of positive results in urine screening programs conducted bymental health service providers in 2006.Nonetheless,there is scarce research on the combined use of amphetamine and heroin during pregnancy to the health ofneonates.Most of the related studies were performedseparately in western countries.Consequently, forbetter insight into the issue of polydrug use of amphetamine and heroin during pregnancy, there is a need forresearch in various study settings. We thus report theclinical courses of three neonatal cases whose mothersconfessed that they continued using heroin with methadonereplacement therapy during pregnancy, with no idea thatamphetamine was foisted until it was identified in theneonate’s hair and urine after birth. The methodology ofquantifying these illicit drugs in infants’ urine and hair wascorrespondingly developed and reported.

2. Methods

2.1. Study settings

At our tertiary neonatal care center, three pregnant womenunder suspicion of illicit drug usage were referred from July1, 2009 to June 30, 2010. They were found addicted toheroin and probably amphetamine during outpatient visitswith in-depth interview and routine examinations duringpregnancy. Neonatal urine and hair samples were regularlycollected for toxicological exams at 1 day old and 14 daysafter admission. For Case 2, additional urine and hairsamples were collected on the 7thday and 16thday afteradmission. For Case 3, hair was additionally collectedafter the first day of use of morphine sulfate, after 48hours, and then 14 days after morphine sulfate was discontinued. The research protocol was approved by theEthics Committee of Chung Shan Medical UniversityHospital before the recruitment of cases.

2.2. Measurement methods

Hair samples were bundled and cut as close as possible tothe scalp because this is the region of least variation ingrowth rate, at sites behind the ears, and on the occipitalbone within an area of approximately 5 cm2. If it cannot becollected by the standard protocol, the alternate site forhair sampling area must be described. The proximal end ofthe sample must be specified if it was longer than 4 cm. In addition, sufficient amount of hair (50 mg for about 100hairs) must be collected to allow primary testing, followedby confirmation of the sample if necessary. The samplesshould be put in separate paper bags for shipment at roomtemperature.

The commonly abused drugs in Taiwan suggested by Centers of Disease Control, Taiwan were then measuredby gas chromatography/mass spectrometry (GC/MS),according to this modified method for which a simultaneous GC/MS assay was developed and validated.Utilization ofthis new technique made it possible to simultaneouslydetermine four classes of illegal drugs in human hair,including amphetamines (amphetamine, AP; methamphetamine, MA; 3,4-methylene dioxyamphetamine, MDA;3,4-methylene dioxymethamphetamine, MDMA; 3,4-methylene dioxyethylamphetamine, MDEA), ketamine(ketamine, K; norketamine, NK), methadone (methadone,MTD; 2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine,EDDP), and opiates (morphine, MOR; codeine, COD; 6-acetylmorphine, 6-AM). In brief, hair samples (10 mg)were washed, cut, and incubated overnight at 25℃ in methanol, then extracted by solid-phase extraction (SPE),and derivatized using heptafluorobutyric acid anhydride(HFBA) at 70℃ for 30 minutes. Finally, the derivatives wereanalyzed by GC/MS in full-scan or selected-ion monitoring(SIM) mode. GC/MS analyses were performed with an Agilent 6890 series automatic injector and GC interfaced to anAgilent 5973 MS operated in electron ionization (EI) mode(Agilent Technologies, Wilmington, DE, USA). Separationwas achieved with a 30-m HP-5 MS capillary column (5% phenyl methyl siloxane with 0.25 mm I.D and 0.25μm film thickness (Agilent Technologies, Palo Alto, CA, USA) and thehelium carrier gas flow rate was set at 1 mL/min.

3. Results

3.1. Clinical courses and treatments

All three case mothers self-reported as current heroinusers, receiving methadone therapy for 1 week to 4months, at the respective ages of 38, 31, and 32 years old atdelivery for gravida 3 para 3, gravida 5 para 2 SA 1 AA2, andgravida 5 para 1 AA4, separately. Details of their illicit druguse history, clinical features, symptoms/signs, and treatment outcomes are shown inTable 1. One subject selfreported combined use of amphetamine (Case 2). Twocases were heavy smokers of two to three packs a day(Cases 1 and 2), but neither was an alcohol user. Alloutcomes of prenatal laboratory tests for the cases werewithin normal range. There were no incidences of toxoplasmosis, other infections (syphilis, varicella, mumps,parvovirus, and HIV), rubella, cytomegalovirus, and herpessimplex (TORCH), hepatitis B, or any other sexuallytransmitted infections, but one subject had chronic hepatitis C with symptoms of hyperbilirubinemia, elevatedglutamyl oxaloacetic transaminase and glutamyl pyruvictransaminase levels, and an episode of fulminant hepatitisduring pregnancy (Case 2). Mild to medium cholecystitiswas also noted before delivery. In addition, mitral valveprolapse and a depressive episode of bipolar affectivedisorder were experienced, with a suicide attempt by jumping from a high building resulting in multiple fractures.Case 1 reported that throughout the whole pregnancyshe was on inhaled corticosteroid therapy for asthma.Particularly, besides a 4-year history of intravenous heroinabuse, Case 3 had been diagnosed as having depression withhallucinations about 9 years before, according to heranamnesis. However, she did not take antidepressantsregularly. She claimed that she had received electivetermination of pregnancy four times during her firstmarriage because of the concern of adverse effects fromantipsychotics. She had started oral methadone treatments3 years before and had been on oral methadone 1 mL/dayfor about 4 months preceding the delivery, but intravenousheroin injection was still used 1 day prior to delivery.

Two males and one female were born at the gestationalages of 35, 34, and 38 weeks, respectively, with labor painplus preterm premature ruptured of membrane (PPROM),labor pain plus uterine contraction by vaginal delivery(Cases 1 and 2), and via cesarean section due to breechpresentation (Case 3). Their Apgar scores at 1 min were 9,6, and 6, respectively, and 10, 7, and 9 at 5 minutes,respectively. Intrauterine growth restriction (IUGR) withlow birth weight was diagnosed in two of them. Ponderalindexes21of 2.02%, 2.72%, and 2.26%, respectively, wereidentified. On the first day, all neonates were admitted intothe neonatal intensive care unit (NICU) for significant NAS,including moderate tremors when undisturbed, highpitched crying, irritability, tachypnea, and poor feedingwith uncoordinated sucking. Only Case 3 had hypertonicity.Serial brain ultrasound exams, electroencephalography,and brainstem auditory evoked potentials were unremarkable for all neonates, except that electroencephalographyof Case 3 showed cerebral dysfunction on the 3rdday, butimproved after treatment.

Following our protocol for treatment of neonates withmaternal heroin exposure in the NICU, each infant wasplaced in a darkened incubator and the modified Finneganneonatal abstinence scoring system was applied every 4hours. In Case 1, a condition of tachypnea with 70 respirations/min developed on admittance, therefore nasalcontinuous positive airway pressure was used for 2 days. Inspite of the long duration of maternal heroin addiction, noconsecutive scores higher than eight within three timeswere obtained during the 2 weeks’ duration of observation.Thus, the baby received supportive care without particularmedication intervention. For Case 2, umbilical cord twicearound the neck with moderate to heavy meconium stainedand cyanosis was found when labored. Fetal distress andheroin withdrawal symptoms were noted. An oxygen hoodwas used for the first 2 days. Also, no continuous Finneganneonatal abstinence scores>8 for three times wereobserved within 2 weeks. Accordingly, only supportive carewas given. For Case 3, following the practical guidelines,14drug treatment with phenobarbital proceeded in a loadingdose of 15 mg/kg and the remainder administered 6 mg/kgfrom the first day of admission. Repeated scoring every4 hours using the modified Finnegan neonatal abstinencescoring system revealed 6e12, after treatment for 1 week.Thus, morphine sulfate was combined with phenobarbital ina dosage of 40mg/kg every 4 hours. After 5 days ofcombination treatment with phenobarbital and morphinesulfate, his condition was stabilized. The score decreased (≤7) and phenobarbital was gradually tapered till the 14^th day of life. Morphine sulfate was gradually stopped,reduced by 5μg/kg per day during a period of 8 days andwithheld on the 22^nd day after birth.

Case 1 has been followed for a period of 15 months. Thesigns of high-pitched cry, frequent irritability, and poorfeeding leading to failure to thrive and developmentalretardation were still observed at 6 months. Currently, thecase has delays in developmental milestones, sitting aloneat 1 year old, and not able to walk alone yet. Case 2 hasbeen followed for 10 months. Her signs of high-pitchedcry, frequent irritability, and poor feeding leading to failureto thrive continued in the follow-up period. She alsoexperiences delay in the developmental milestone forrolling over alone at 9 months old. At discharge, Case 3 wasfound with excessive sucking and loose stool. At present, hehas been followed up for 15 months, and delays in developmental milestones have been observed (sitting alone at10 months old and not walking alone yet). Also, he manifests inappropriate laughing. Prolonged follow-up was recommended for further assessments on his neurobehavioraldevelopment.

3.2. Measurement outcomes

Urine specimens of the three neonates were first examinedby an enzyme immunoassay to identify opiates, amphetamines, MDMA, marijuana, and ketamine, on the 1^st day after birth. The positive specimens were tested by GC/MSto determine the presence of the targeted analytes abovethreshold concentrations. The results showed that none ofthe samples revealed a positive finding in urine, except Case 3, who was positive for morphine at a concentration of 308 ng/mL.

All quantitative analyses of hair specimens were performed with GC/MS in selected ion-monitoring mode with20 ms of dwelling time for each ion (Figure 1). The GC/MSSIM chromatogram of the investigated drugs (HFBA, MAHFBA, MOR-HFBA, MTD, EDDP, MOR-HFBA, COD-HFBA, and6-AM-HFBA) after derivatization depicts pikes at 2.56 min,3.06 min, 6.50 min, 7.13 min, 8.24 min, 8.42 min, and8.83 min, respectively. Ions monitored in SIM mode were:

m/z 254, 210, 118 for MA; m/z 258, 123 for MA-d5; m/z 223,294 for MTD; m/z 224, 297 for MTD-d3; m/z 262, 277 forEDDP; m/z 265, 280 for EDDP-d3; m/z 464, 677 for MOR;m/z 467, 680 for MOR-d3; m/z 464, 523, 411 for 6-AM; m/z467, 526, for 6-AM-d3.

Hair samples of neonates 2-2.5 cm in length reflect thelevel of drug abuse of the mothers for 2-3 months beforedelivery. The results revealed positive readings for MA,MTD, EDDP, MOR, and COD at 865, 4145, 960, 180, and180 pg/mg, respectively, for Case 1 (Figure 2A), and 1035,15850, 1165, 1520, and 320 pg/mg, respectively, for Case 2(Figure 2B).Figure 2C shows Case 3’s AP, MA, MTD, MOR,COD, and 6-AM levels at 1115, 11396, 448, 3671, 1886, and165 pg/mg, respectively.

4. Discussion

In the current study, we present neonatal case series whosemothers had heroin addiction with methadone replacementtherapies during pregnancy for their clinical courses andtreatment outcomes. To our knowledge, this is the first report on this topic in East Asia.9Two of the cases’ mothershad no idea about using polydrug with amphetamine foisted. Mild to moderate NAS was identified among thesecases, including moderate tremors when undisturbed, highpitched cry, irritability, tachypnea, poor feeding withuncoordinated sucking for all of them, or even hypertonicity and cerebral dysfunction by electroencephalographyfor one case, which is similar to previous clinical reports.To a certain extent, findings on the scores of modifiedFinnegan neonatal abstinence scoring system might bea sign of the exposure levels and durations of heroinaddiction reported by subjects. Fortunately, other thanhepatitis C for one mother, the babies were born free ofbloodborne virus infections and sexually transmitteddiseases.Although all the cases were nearly full term(34-38 weeks), IUGR, defined by low birth weight, wasobserved in two cases (Table 1). Psychiatric comorbidity ofthe mothers and unconscious polydrug use make the clinicalmanagement even more challenging.

Quantitative hair specimens can be used to drawa gestational drug exposure profile and to study its potential link with the appearance of NAS. Nevertheless, theamount of neonatal hair is often insufficient because casemothers do not consent to infant hair collection forcosmetic or specific cultural reasons. Although the neonatehair sample is often small in amount, our modified assaymethod using 10 mg still provided enough sensitivitydetection of the abused drugs, and gave applicable information to trace the illegal drugs used during pregnancy.The limit of detection and limit of quantification obtainedwere 40 pg/mg for AP, MA, MDA, MDMA, and MTD and 80 pg/mg for K, NK, EDDP, MOR, COD, and 6-AM when only10 mg of hair sample was used.Calibration curves for 10analytes were established in the concentration range of 0.08-5 ng/mg with a correlation coefficient (r2) of 0.997,using 200 pg/mg deuterated analogs of each analyte asinternal standards for quantification.

In addition to patients’ self reports, drug testingprovides precise clinical information to assess individualexposures to illicit drugs. Samples of urine and blood arethe most commonly used specimens, but their detectionwindows are only the preceding few days. Only one case’surine sample (Case 3) was found with traceable opiatebecause his mother used it 1 day before delivery. Becauseillicit drugs and their metabolites can disappear from theurine and the bloodstream within days, a neonatal hairspecimen would be a better alternative because of itssubstantially wider detection window, enabling retrospective inspection of chronic exposure. In this study, wefound that the mothers of Cases 1 and 2 both had experiences of methadone treatment programs because highconcentrations of methadone and EDDP were found inneonatal hair. In addition, they also abused methamphetamine frequently. Moreover, the laboratory evidenceshowed that the mother of Case 2 abused heroin habitually,and Case 1 showed a low level of MOR and COD, indicatingthat the mother had abstained from heroin. Compared to Case 1, the higher levels of MTD and EDDP for Case 2 suggested that it could be the case that the illicit drugs sheused contained methadone because she was only treatedwith methadone for 1 week. The other minor but noticeable issue was this case’s weakened liver function due to chronic hepatitis C infection, resulting in a prolonged halflife period of methadone, which is 24-36 hours for healthypeople. Hair testing results of Case 3 showed that themother abused both heroin and methamphetamine, and shehad experienced fewer methadone treatments, shown bythe low amount of methadone in her hair. Althoughamphetamine and/or methamphetamine was identified inneonatal hair samples in the current study, we did not findthe related physical conditions or abnormal brain development suggested in the literature among these babies.However, concerns about neurobehavioral development forprenatal amphetamine exposure remain,25and furtherfollow-up with intensive assessment is recommended. Drugabuse in women of childbearing age is an emerging issueworldwide.A better understanding on the clinical courses of subjects with polydrug use and experiences oftreatment are in demand. Thus, the development ofintensive prenatal care strategies and enrollment in treatment programs before/after delivery are of interest forfuture studies.

Acknowledgments

The study was supported by a research grant from ChungShan Medical University Hospital (CSH-97-A-10), TaichungCity, Taiwan.