neonatal manifestations of maternal phencyclidine exposure

Golden et al, Maternal phencyclidine use and neonatal manifestations 185

J. Perinat. Med.15 (1987) 185

Neonatal manifestations of maternal phencyclidine exposure

Nancy L.Golden, Betty R. Kuhnert, Robert J. Sokol, Susan Martier, and ThomasWilliams

Department of Pediatrics and Reproductive Biology, Cleveland MetropolitanGeneral Hospital/Case Western Reserve University School of Medicine, Cleve-land, Ohio, U.S.A.

1 Introduction

Phencyclidine (angel dust) is a drug frequentlyabused by adolescents and young adults. Forexample, a New York State study of secondaryschool students indicated that 15.5% of stu-dents (276,000) had used phencyclidine [9]. Al-though originally developed as a short actinganesthetic, phencyclidine's sedative and halluci-nogenic effects have made it a popular drug ofabuse among young people of childbearing age[11]. The use of phencyclidine by pregnantwomen is of concern since phencyclidine hasbeen shown to cross the placenta and to reachthe fetus [13].There have been several reports in which phen-cyclidine exposure during gestation has beenassociated with adverse effects on the fetus.In one report, two neonates exposed in uterodemonstrated a classic withdrawal syndrome[13]. They were irritable, hypertonic, jittery, andhad diarrhea and vomiting for several dayspostpartum. A second report described neuro-logic and anatomic abnormalities in one infantwhose mother had used large amounts of phen-cyclidine throughout pregnancy [2]. This infanthad an abnormal neurologic examination andabnormal behavior in the neonatal period; hewas jittery, irritable, hypertonic, unable to track

Curriculum vitae

NANCY L. GOLDEN, M. D., was born in 1944 in Bridge-port, Connecticut, U. S. A. She received her M. D. fromThe Tufts University College of Medicine, Boston, in 1971and qualified as a pediatrician at the University of NorthCarolina School of Medicine in 1976. Since 1984, she hasworked at the Cape Cod Hospital Hyannis, Massachu-setts. Her main fields of interest are fetal alcohol syn-drome and drug effects on the fetus.

visually, and a poor feeder. Furthermore, hisappearance was unusual; he had a triangularlyshaped face, eyes with an antimongoloid slant,and a pointed chin. At the age of five he hadspastic quadriparesis and was severely mentallyretarded.Given the reported frequency of phencyclidineuse, the fact that phencyclidine crosses the pla-centa, and the possibility that intrauterine ex-posure to phencyclidine may be harmful to thefetus, we undertook a large prospective study todescribe the extent of phencyclidine use duringpregnancy and the characteristics of the userpopulation [4]. The results of consecutivescreening of more than 2,000 pregnant womenbetween 1981 and 1982 showed that 7.3 percent

1987 by Walter de Gruyter & Co. Berlin · New York

186 Golden et al, Maternal phencyclidine use and neonatal manifestations

gave a verbal history of phencyclidine use butonly 0.8 percent could be confirmed as currentusers by maternal urine screening [4], The pur-pose of the present study was to describe thephysical characteristics, neurologic findings, be-havior, and neonatal clinical course of the in-fants of these women, and to compare themwith a group of matched control infants.

2 Methods

All women attending the prenatal clinics affili-ated with Cleveland Metropolitan GeneralHospital from June, 1981 through October,1982 were questioned on their initial visit abouttheir drug, tobacco, and alcohol use. Trainedscreeners took substance abuse histories [4] andadministered Michigan Alcoholism ScreeningTests [10].Study patients were defined as those who gavea verbal history of phencyclidine use duringpregnancy and/or positive urine tests. Potentialcontrol patients were defined as those who didnot have a history of phencyclidine use duringpregnancy and whose urine tests were negativefor phencyclidine. For this study, controlpatients were matched to study patients onmaternal date of recruitment, age, ethnicity,weight, parity, weeks gestation at registration,and tobacco smoking behavior.All patients' urines were initially screened forphencyclidine using the Enzyme Mediated Im-munoassay Technique (EMIT) [1]. Specimenswere initially defined as "positive" if the EMITreading was ten units above the negative cali-brator. The presence of phencyclidine was sub-sequently verified by gas chromatography —mass spectrometry, using previously reportedmethodology [7]. Urines from study patientsand any controls thought to be at risk for drugabuse were retested at each visit. Most of thepatients had their urines tested 4 — 5 times priorto delivery. Furthermore, study patients andneonates had urine and blood analyses duringthe intrapartum and postpartum period. Allbloods were negative. The results of the phar-macological analyses have been reported separ-

ately [8]. Control patients' urines were also re-tested during the intrapartum period.

Table I. Neonatal outcome variables.

AnatomyWidely patent suturesSlanted eye shapeNarrow mandibular angleDecreased joint mobilityFlexed joints

Neurologic functionAbnormal movementHypertoniaHypotoniaTremorsDecreased visionDecreased hearingAsymmetrical facial movementAbnormal extra ocular movementsDecreased pupil reactivityMiotic pupil positionNystagmus-verticalNystagmus-horizontalIncreased ankle reflexIncreased knee reflexIncreased brachial reflexDepressed suck reflexDepressed root reflexDepressed grasp reflexAbnormal galant reflexAbnormal moro reflexAbnormal step reflex

BehaviorPoor trackingDecreased attentionMood (irritability)Poor feedingHigh pitched cry

Physiologic functionHeart rate - < 90 >Respiratory rate — <SepsisJaundiceHypoglycemiaHypocalcemiaAnemiaRespiratory distressOther

16020 > 60

GrowthHeight - < 10% > 90%Weight - < 10% > 90%Head circumference —

< 10% > 90%

J. Perinat. Med. 15 (1987)


Study and control neonates were evaluated midfeeding between 24 and 72 hours of age by asingle examiner blind to the maternal history.This examiner is a certified pediatric nurse prac-titioner who is also a certified Brazelton exam-iner with additional experience screening FetalAlcohol Syndrome infants [3]. Growth para-meters, physical characteristics, neurologicfindings, and behavior were documented ac-cording to a format based on previous descrip-tions of phencyclidine exposed infants (table I)[2, 13].The data were analyzed in the following ways:First, descriptive statistics and paired t testswere used to compare study and control popu-lations in regard to maternal age, race, parity,smoking behavior, and number of drugs used.Next, a hierarchial analysis was used to exam-ine the relationship between drug use and neon-atal outcome. First, paired t tests were used tocompare the total number of abnormal findingsbetween the study and control infants and thefrequency of abnormalities were plotted on his-tograms. Next, stepwise multiple regressiontechniques with nonforced variables were usedto determine relationships between neonataloutcome and the seven classes of maternallyabused substances (phencyclidine, narcotics,barbiturates, cocaine, marijuana, glue, and al-cohol). Then, chi square tests were used todetermine which of the dichotomous outcomevariables were significantly different in studyand control neonates. Finally, stepwise dis-criminant function analysis was used to relatespecific outcome variable to phencyclidine use.A probability level p < 0.05 was accepted assignificant.

3 Results

A total of 2,327 pregnant women were screenedfor phencyclidine use early in pregnancy. Averbal history of phencyclidine use during preg-nancy was reported by 149 (6.4%) and currentuse was reported and/or documented in 23(1%). Data collection was completed for 94study patients, 14 of whom were verified users,and 94 matched controls. Table II summarizes

Table II. Entry of patients into neonatal PCP study.

Parameter Number

Women screened 2347Positive responders 172

Historical use 149Confirmed use 23

Positive responders lost to follow-up 78

Remaining study patients 94Historical use 80Confirmed use 14

the entry of patients into the study. Mostpatients lost to follow-up were lost due to deliv-ery at another hospital, delivery at times whenexaminations could not be obtained or refusalparticipate. The demographic characteristics ofthis population has been described in depth [4]and previous analysis indicated that there wereno significant demographic differences betweenthe patients studied and those lost to follow-up.Table III shows that study and control groupsdiffered significantly only in the mean numberof drugs used. This was true both within thetotal study sample and the subset of confirmedusers. Both study and control patients mostfrequently abused marijuana, cocaine, barbitu-ates, alcohol and glue in that order. Matchingcriteria were effective as both groups werefound to be similar in maternal age, race, par-ity, and smoking behavior. Furthermore, othermaternal characteristics, for which patientswere not matched, were also similar in the studyand control groups. Maternal education andsocioeconomic status were not statistically dif-ferent between study and control groups.Patients were also similar in terms of medicalobstetric risk. Antenatal risk scores (Hobel)[12] were similar after adjustment for substanceabuse. Furthermore, the infants were all at termand similar in regards to intrapartum risk andapgar scores.A significantly increased number of abnormalfindings in neonates was associated with ma-ternal phencyclidine exposure (table IV). There



Table III. Comparison of study and control mothers.

Historical and confirmed (94 pairs) Confirmed users (14 pairs)

Studya Control

AgeMean ± SD 20.78 ± 3.06 21.11 ± 3.87

RaceWhite 70 70Appalachian 7 7Hispanic 2 2Black 14 14

ParityMean ± SD 0.91 ± 0.95 0.99 ± 1.04

Education (Years) 10.77 ± 1.69 9.13 ± 1.54

Socioeconomic status(Hollingshead) 4.91 ± 0.32 4.90 ± 0.39

ReligionCatholic 52 56Protestant 26 24Other 15 13

SmokingYes 77 77No 17 17

Number of classes ofdrugs used

Total number 133 58Mean + SD 1.41 ± 1.15 0.64 ± 0.80

Antenatal risk score0 20.94 ±11.88 15.89 ± 8.88

a 80 with history of use and 14 verified present usersb 14 verified usersc Modified Hobel

Table IV. Abnormal findings in neonates of phencycli-dine exposed women.

Study Control(N = 94) (N = 94)

Total number ofabnormal findings 477 392

Mean ± SD 5.07 ± 2.93 4.17 ± 2.65Significance* p < 0.01

* Paired t test

were significantly more abnormalities in boththe sample which included all users, and in thesubset of confirmed users. In the latter group,

Signi- Studyb Control Signi-ficance ficance

NS 20.42 ± 2.38 19.86 ± 2.77 NS

NS 7 7 NSNS 2 2 NSNS - - NSNS 5 5 NS

NS 0.86 ± 0.86 1.00 ± 0.88 NS

NS 10.57 ± 1.40 10.14 + 1.51 NS

NS 4.93 ± 0.27 4.93 ± 0.36 NS

NS 11 10 NSNS 3 4 NSNS 0 0 NS

NS 12 12 NSNS 2 2 NS

p < 0.001 37 4 p < 0.00l2.64 ± 1.39 0.31 ± 0.63

p < 0.001 21.71 ± 17.08 13.93 ± 11.33 NS

the study infants averaged 5.8 + 3.3 abnormal-ities and the control group averaged 3.5 + 2.4abnormalities (p < 0.02). While the means weredifferent, the histograms (figure 1) of study andcontrol patients, show that the distributions ofabnormalities were similar in each group.Many of the study patients were multiple drugusers. However, stepwise multiple regressiontechniques showed that the number of abnor-malities was related to phencyclidine use andnot other drugs. Phencyclidine was the onlysubstance of the seven classes of substancesabused by the patients which accounted for asignificant portion of the variance related to



COLLJH

—I^DCΟω^u.0ocUJffi23

02

4

68

101214

16

18

-XXX

-·····«·«

···················•X• XX

···X

•

1 1 1 1 I I I 1

••••XXXXXXXXX»•••xxxxxxxxxxxxxxxxxx»••••••••••••xxxxxxxxx xxx xxxxxxx••••••••••••xxxxxxxxx······· xxxxxxxxxxxx••••XX•••xxxx•XXX

• phencyclidineX controls

1 1 1 1 I I I 1

10 15 20 25 30 35

NUMBER OF PATIENTS

40

Figure 1. Number of abnormalities seen in infants ofphencyclidine exposed women (·) and infants of con-trols (χ) .

the number of abnormal findings (F = 4.38;df = 1,138; ρ < 0.05). Nevertheless, drug inter-actions cannot be ruled out. There were noopiate users in the study sample.Abnormal neurological and behavioral findingswere associated with maternal phencyclidineexposure (table V). Those found significantlymore often in the study group included poorattention, hypertonic ankle reflexes, and de-pressed neonatal reflexes — grasp and rooting.The abnormal grasp reflex was the only reflexsignificantly related to phencyclidine use withstepwise discriminant function analysis(F = 8.38; df = 1,174; ρ < 0.01). No signifi-cant relationships were found between abnor-mal anatomic findings and maternal phencycli-dine use.

Table V. Specific abnormal findings more frequent instudy infants.

Abnormal findings Signi-ficance*

Decreased attention 0.01Depressed neonatal reflexes

Grasp 0.01Root 0.03

Hyperactive deep tendon reflexes ankle 0.05

* Pearson's χ2

4 Discussion

The results of this study of 188 infants suggestthat infants born to women with a history ofphencyclidine use during pregnancy have moreabnormal neurological and behavioral findingsat birth than control infants. These abnormali-ties are related to phencyclidine use and not toother drug use. However, interactions withother drugs cannot be ruled out. No anatomicalabnormalities were detected, but the samplesize may need to be much larger to see theseabnormalities.

Women with a verbal history of phencyclidineuse, as well as those with chemically confirmeduse during pregnancy, were included in thestudy group for two reasons: First, phencycli-dine use could be confirmed in only a smallpercentage of the population. This was prob-ably due to infrequent urine testing [4]. There-fore, the frequency of phencyclidine use (whichexceeds 15% in some populations [9]) wouldalmost certainly be underestimated and manyusers would have been missed if only the con-firmed users had been studied. The decision toinclude historical and confirmed users as studypatients appears justified since the numbers ofabnormal findings were similar in these twogroups.

The results of this study do not completelyagree with what has previously been reported.These findings are similar to those of a reportwhich described an infant exposed to phencycli-dine in utero with lethargy, inability to trackvisually, hypertonicity, and poor feeding [2].However, the results of the present study aredifferent in that structural abnormalities werenot found. Thus, this study suggests that phen-cyclidine is not an anatomic teratogen but doesnot confirm this. A larger sample size may beneeded to see anatomical abnormalities. Fur-thermore, the possibility exists that study in-fants were not exposed to phencyclidine at thetime of organ formation, or that exposure wasof insufficient amount or duration to causestructural defects such as those previously re-ported [2]. Moreover, the earlier report involvedan infant also exposed simultaneously to very



high amounts of marijuana throughout gesta-tion.Even though phencyclidine users are almostalways multiple drug abusers [4], this studydemonstrated statistically significant effects onthe fetus only from phencyclidine exposure.However, previous studies have suggested arelationship between gestational exposure tobarbiturates [6], marijuana [2, 5], and alcohol[13] and an adverse fetal outcome. In addition,past studies have shown that narcotic exposurecan have severe effects on the neonates [14].Since none of the patients in this study werenarcotics users, this study cannot address theissue of neonatal effects due to combined phen-cyclidine and narcotic use.

The results of this study suggest that long-termfollow-up of these infants would be worthwhile.While it is unlikely that the neonatal findingswould directly correlate with long-term find-ings, one could speculate that these childrenmay be at increased risk of minor neurologicabnormalities and/or learning difficulties.Our previous study indicated that phencyclidineuse during pregnancy was widespread [4]. Thisstudy suggests that exposure of the fetus tophencyclidine can result in abnormal neonatalneurologic findings and behavior. Future stud-ies must determine the long-term effects ofexposing the fetus to phencyclidine.

Summary

The purpose of this study was to determine the effectsof maternal phencyclidine use on the fetus. Ninety-fourneonates with maternal phencyclidine exposure werecompared with 94 controls. Maternal phencyclidine usewas assessed by questionnaire and repeated urine testing.Mothers of study and control patients were matchedfor demographic characteristics. Infants were assessedbetween 24—72 hours postnatally by a single examinerblind to the maternal history. The results showed thatstudy infants had a mean of 5.02 + 2.93 abnormalities

while controls had a mean of 4.13 ± 2.65 abnormalities(p < 0.01). Furthermore, study infants were more likelythan controls to have poor attention, hypertonia, anddepressed neonatal reflexes (p < 0.05). The contributionof seven drug classes to the total number of abnormali-ties was assessed using stepwise multiple regression. Onlyphencyclidine accounted for a significant proportion ofthe variance (f = 4.38; p < 0.05). The results of thisstudy suggest that maternal phencyclidine use may leadto abnormal neonatal neurologic findings and behavior.

Keywords: Fetus, newborn infant, pregnancy, prenatal exposure delayed effects, phencyclidine, phencyclidineabuse.

Zusammenfassung

Manifestationen einer mütterlichen Phenzyklidineinnahmebeim NeugeborenenIn der vorliegenden Studie wurden die Auswirkungeneiner mütterlichen Phenzyklidineinnahme auf den Fetenuntersucht. 94 Neugeborene von Müttern, die Phenzy-klidin eingenommen hatten, wurden verglichen mit 94Kontrollkindern. Der Phenzyklidinabusus wurde überFragebögen und wiederholte Urintests erfaßt. Frauenaus der Phenzyklidingruppe und Frauen der Kontroll-gruppe wurden nach demographischen Parametern ein-ander zugeordnet. Die Kinder wurden von einer einzigenPerson, die die mütterliche Vorgeschichte nicht kannte,innerhalb von 24 — 72 Stunden post partum untersucht.Kinder aus der Phenzyklidingruppe zeigten im Mittel

5.02 + 2.93 auffallige Befunde, Kinder aus der Kontroll-gruppe dagegen 4.13 ± 2.65 (p < 0.01). Darüber hinausfielen bei den Kindern der Phenzyklidingruppe im Ver-gleich zur Kontrollgruppe eine verminderte Reaktions-bereitschaft, ein erhöhter Tonus sowie herabgesetzteneonatale Reflexe auf (p < 0.05). Durch eine multipleRegressionsanalyse wurde für 7 Medikamentengruppenermittelt, wie häufig auffallige Befunde bei den Kindernauftraten. Nur für Phenzyklidin konnte ein signifikanterAnteil an der Varianz berechnet werden (f = 4.38;p < 0.05). Die Ergebnisse zeigen, daß eine mütterlichePhenzyklidineinnahme zu auffalligen neurologischen Be-funden und Verhaltensmustern beim Neugeborenen füh-ren kann.

Schlüsselwörter: Fetus, Neugeborenes, Phenzyklidin, Phenzyklidinabusus, pränatale Exposition und Manifestationbeim Kind, Schwangerschaft.



Resume

Manifestations neo-natales de la prise maternelle de phen-cyclidineLe but de cette etude a etc de determiner les effets de laprise maternelle de phencyclidine sur le foetus. On acompare quatre vingt quatorze nouveaux-nes dont lamere avait pris de la phencyclidine avec 94 controles.On a determine la prise maternelle de phencyclidine al'aide de questionnaires et par des analyses urinairesrepetees. Les meres de I'etude et les patientes temoinsont etc appariees selon les caracteres demographiques.Les enfants ont ete examines entre 24 et 72 heuresde vie par un seul examinateur qui ne connaissait pasFhistoire maternelle. Les resultats ont montre que lesenfants de I'etude presentent en moyenne 5,02 ± 2,93

anomalies tandis que les controles ont en moyenne4,13 + 2,65 anomalies (p < 0,01).En outre, les enfants etudies presentent plus souvent queles temoins une attention faible, une hypertonie et desreflexes neonataux deprimes (p < 0,05). A Taide d'unecourbe de regression multiple, on a estime le role de 7types de medicaments face au nombre total d'anomalies.Seule la phencyclidine entrame une variance significative(f = 4,38; p < 0,05).Les resultats de cette etude suggerent que la prise mater-nelle de phencyclidine peut entrainer des anomalies deFexamen neurologique neonatal et des troubles du com-portement.

Mots-cles: Abus de phencyclidine, effet ä long terme de Fexposition prenatale, foetus, grossesse, nouveau-ne,phencyclidine.

Acknowledgements: Research supported by the National Institute on Drug Abuse — Grant number ROI DA02903and US PHS Grant number 5M01RR-00210.

References

[1] BASTIANA RJ, RC PHILLIPS, RS SCHNEIDER, EFULLMAN: Homogeneous immunochemical drug as-says. Am J Med Technol 39 (1973) 211

[2] GOLDEN NL, RJ SOKOL, IL RUBIN: Angel dust:possible effects on the fetus. Pediatrics 65 (1980) 18

[3] GOLDEN NL, RJ SOKOL, BR KUHNERT, SF BOT-TOMS: Maternal alcohol use and infant develop-ment. Pediatrics 70 (1982) 931

[4] GOLDEN NL, BR KUHNERT, RJ SOKOL, S MARTIER,BS BAGBY: Phencyclidine use during pregnancy.Am J Obstet Gynecol 148 (1984) 254

[5] GREENLAND S, KJ STAISCH, N BROWN, SJ GROSS:The effects of marijuana use during pregnancy. I:A preliminary epidemiologic study. Am J ObstetGynecol 143 (1982) 408

[6] HILL RM, WM VERNIAUD, MG HORNING, LBMcCuLLEY, NF MORGAN: Infants exposed in uteroto anti-epileptic drugs. A prospective study. Am JDis Child 127 (1974) 645

[7] KUHNERT BR, BS BAGBY, NL GOLDEN: Measure-ment of phencyclidine and two hydroxylatedmetabolites by selected ion monitoring. J Chroma-togr 276 (1983) 433

[8] KUHNERT BR, NL GOLDEN, CD SYRACUSE, BSBAGBY, PM KUHNERT: Phencyclidine disposition inmother and neonate. Res Comm Sub Abuse 5(1984) 301

[9] MCADAMS MA, RL LINDER, SE LERNER, RS BURNS(eds): Phencyclidine abuse manual. University ofCalifornia, Los Angeles, Extension 13 (1980)

[10] SELZER ML: The Michigan alcoholism screeningtest; the quest for a new diagnostic instrument. AmJ Psych 127 (1971) 89

[11] SHOWALTER CV, WE THORNTON: The increasingabuse of phencyclidine. Ill Med J 151 (1977) 387

[12] SOKOL RJ, MG ROSEN, J STOJKOV, L CHIK: Clinicalapplications of high risk scoring on an obstetricservice. Am J Obstet Gynecol 128 (1977) 652

[13] STRAUSS AA, HD MODANLOU, SK Bosu: Neonatalmanifestations of maternal phencyclidine (PCP)abuse. Pediatrics 68 (1981) 550

[14] VARGAS GC, RS PILDES, D VIDYASAGAR, LG KIETH:Effect of maternal heroin addiction on 67 livebornneonates. Clin Pediatr 14 (1975) 751

Received October 20,1985. Revised November 18,1985.Accepted February 10, 1986.

Betty R. Kuhnert, Ph. D.Department of Reproductive BiologyCleveland Metropolitan General Hospital3395 Scranton RoadCleveland, Ohio 44109, U.S.A.


A Modern Approachto thePerimenopausal YearsEditor: Robert B. Greenblatt1986.17x24 cm. VI, 256 pages. 62 illustrations. 44 tables.Softcover DM 98,- ISBN 311010937 9(New Developments in Biosciences 2)With advancing knowledge of the physiology of the hypothalamic-pituitary-ovarian axis and how gonodal steroids influence cerebral neurohumors, boneand skin metabolism, serum lipoproteins, a new approach to the management ofthe perimenopausal years is now possible. The dimensions of the menopausalsyndrome may now be extended to include not only vasomotor symptoms, butalso metabolic disorders, bone loss, decreases in skin collagen, reversal of highand low density lipoproteins, and psychogenic disturbances (menopausalanxiety and depression, migrainoid headaches, insomnia and sexual dysfunc-tion) that are initiated or aggravated by loss of ovarian hormones.The text furnishes much information that is not generally known - to wit:• estrogens and androgens diminish collagen loss of the skin,• estrogens and progestogens not only prevent further bone decay but may

increase bone density,• cyclic progestogens added to an estrogen regimen actually lessen the natu-

ral incidence of endometrial cancer and may add some degree of protectionagainst development of breast malignancy.

• non-bacterial urinary incontinence and nocturia may be partially or com-pletely alleviated by estrogen therapy.

• androgens added to an estrogen regimen increase sexual fantasies anddesire, and facilitate the orgasmic response.

Φ the danger of myocardial infarction may be reduced greatly by estrogensbecause of their beneficial effect on high density lipoproteins.

wDE

Gde Gruyter · Berlin · New York

neonatal manifestations of maternal phencyclidine exposure

Documents