• 
• 

Three Part Question

In [neonates], what are the [acute symptoms of withdrawal] following [antenatal maternal exposure to cocaine and amphetamines]?

Clinical Scenario

The SHO on the post-natal ward is performing routine newborn examinations and notices a baby who is being monitored using a neonatal abstinence score chart due to a history of maternal illicit drug use.
Urine toxicology results are positive for amphetamines, but negative for opioids.
The SHO has seen the effects of opioid withdrawal on babies before, but asks the SCBU registrar whether babies act in a similar manner when withdrawing from stimulatory drugs such as cocaine and amphetamines.

Search Strategy

Medline via PubMed was the primary source of articles.
Search terms were: [amphetamin* OR metamphetamin* OR cocaine] with the following limits activated: Humans, Clinical Trial, Meta-Analysis, Randomised Controlled Trial, Comparative Study, English, All Infant: birth-23 months.
A secondary search on the Cochrane database was performed using the same search terms.

Search Outcome

A total of 181 papers were found via PubMed, but only 8 were both relevant and of sufficient quality.
The Cochrane Library yielded no further relevant results.
Reasons for excluding papers included poor study design and lack of detailed descriptions of withdrawal symptoms.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Oei 2010	200 amphetamine exposed infants (AMP) vs. 669 infants exposed to other illicit drugs (non-AMP).	Retrospective cohort study (level 2b)	Finnegan score ; Medication requirement for withdrawal symptoms	Median maximum score lower in AMP infants ( 8 vs. 10). AMP infants less likely to require pharmacological treatment (0.9% vs 23.8%)	Finnegan scoring only happened in approx. half of infants. Drug exposure status not routinely confirmed by urine toxicology. Polydrug exposure in both groups.
Bauer et al. 2005	717 cocaine exposed infants vs. 7442 nonexposed infants	Prospective comparative cohort study (level 1b)	Results of physical examination and events during hospitalization	Symptoms more frequent in exposed group: tremors (OR 2.17), high-pitched cry (2.44), irritability (1.81), excessive suck (3.58), hyperalertness (7.78).	‘Non-exposure’ status required negative meconium screen. ‘Cocaine exposed’ did not use opiates, but may use alcohol/marijuana.
Ogunyemi et al. 2004	200 cocaine-exposed (CE) infants Vs. 200 non exposed (NE) infants	Retrospective case-control study level 2b)	Record of events during hospitalization	CE more likely to suffer RDS (14% vs. 4%, p=0.001).37.5% of CE diagnosed as suffering NAS	No allowance made for polydrug use. Only 65.5% of CE tested positive for cocaine. No clear definition for ‘NAS’
Chomchai 2004	47 metamphetamine exposed (ME) vs. 49 non-exposed (NE) infants	Prospective comparative cohort study (level 1b)	Observation for withdrawal symptoms	In ME group, increased frequency of vomiting (11/47, p≤0.001), tremor (5/47, p≤0.001),excoriation(6/47, p≤0.001) temp. instability (7/47, p≤0.001)	No figures provided for symptoms of NE group. Non-blinded study: maternal admission of metamphetamine use triggered automatic NICU admission. No toxicology screening performed in ‘NE’ group.
Myers et al. 2003	Treatment group (n=76): infants of mothers who initially used cocaine and other drugs during present pregnancy but ceased drug use and entered drug treatment programme at least 30 days prior to delivery Vs. Rejecter group (n=18): infants exposed to cocaine and other drugs during pregnancy whose mothers refused treatment Vs. Nonuser group (n=43): infants of mothers who used no drugs during the pregnancy.	Prospective comparative cohort study (level 1b)	APGAR score, Brazelton Neonatal Behavioural Assessment Scale (NBAS) at 2 days and 2 weeks	No statistically significant difference between APGAR or NBAS scores in any of 3 groups.	Avoidance of maternal cocaine/illicit drug/alcohol use in Treatment group and Nonuser confirmed by random urine screening.
Smith et al. 2003	134 metamphetamine exposed infants (ME) vs. 160 unexposed infants (UE)	Retrospective cohort study (level 2b)	APGAR score, Neonatal abstinence score (not Finnegan), Requirement for pharmacological intervention	No significant difference in APGAR score; 49% of ME infants had evidence of withdrawal (scores of ≥5); 4% required pharmacological treatment for withdrawal symptoms (score ≥8 for 3 consecutive scores).	No detailed break-down of NAS scores. Toxicology not performed on ‘control’ infants-possibly drug exposed infants in this group. Infants excluded from ME group if urine toxicology positive for heroin/cocaine. Authors unable to adjust for large differences in alcohol and marijuana use between two groups.
Bada et al. 2002	7442 non-exposed infants(NE) vs. 1185 durg-exposed (EXP)infants : 717 Cocaine-only exposed(CO) vs. 100 Opiate only (OP) exposed vs. 92 both cocaine and opiate exposed (OP+CO) infants	Prospective cohort study (level 1b)	Physical examination	Poor suck, nasal stuffiness and sneezing not significantly associated with either CO or OP. In OP, prevalence of most of CNS/ANS signs higher than in CO,e g, jitteriness/tremors 22% vs 15%, irritability 24% vs. 12%, high pitched cry 11% vs. 3%, hypertonia 8% vs. 2%, excessive suck 5% vs. 2%.	Wide variety in timing of examination (28% 24hrs, 52% 24-48hrs, 12% 48-72hrs). Symptoms occurring later would not have been identified.
Held et al. 1999	5 Studies comparing cocaine-exposed(CE)to non-exposed(NE) infants	Meta-analysis (level 1a)	NBAS cluster scales at birth and 3-4 weeks of age	Only small cocaine effect on neonatal behaviour at both time periods. Birth: CE had poorer performance in habituation(d=0.247), orientation(d=0.282), autonomic regulation(d=0.194) and motor performance(d=0.368).	Variability of times at which 2nd NBAS administered. High incidence of polydrug use in studies.

Comment(s)

Studies examining the effects of antenatal illicit stimulatory drug exposure on the neonate are hampered by difficulties in obtaining reliable information on quantities and timings of maternal drug use, the high incidence of polydrug use, and difficulties adjusting for the large number of confounders (eg, poor nutrition and living conditions) that may exist in this population. A meta-analysis was identified examining the effects of in utero cocaine exposure on neurobehavioural outcome. The five studies included in the meta-analysis used the Neonatal Behavioural Assessment Scale (NBAS) administered at birth and again at 3-4 weeks of age as the outcome measure. This involves scoring several behavioural items and elicited responses before grouping the results into seven ‘cluster scales’ representing distinct aspects of neonatal behaviour (habituation, orientation, motor performance, range of state, regulation of state, autonomic regulation, and abnormal reflexes). Although ‘nonexposed’ infants had significantly superior performance than cocaine-exposed infants in five of the seven clusters at birth, the magnitude of the effect was small in all cases. The largest reliable difference appeared in motor performance (d=0.37) and abnormal reflexes (d=0.30) clusters (where d=0.20 is small size effect ; d=0.50 is a medium size effect ; d=0.80 is a large size effect). Results of the meta-analysis comparing the NBAS scores of the two infant groups at 3-4 weeks of age showed significant superiority of nonexposed infants in four of the seven clusters. Again, the largest effect size was found in motor performance (d=0.43). Although motor performance and abnormal reflexes showed a trend for increased standard differences from birth to 3-4 weeks of age, the reverse was true for orientation and autonomic regulation. The largest problem with the meta-analysis was the inability to isolate the effects of cocine exposure from that of exposure to other substances due to the high incidence of polydrug use. A well-designed study by Myers looked at three groups of infants to examine acute cocaine-withdrawal effects (Treatment group, Rejecter group, Nonuser group: see table). Infants were evaluated at 2 days of age and 2 weeks of age using the NBAS. After adjusting for potential confounding variables, no significant differences on the NBAS scores were found between the three groups at either time period. No high quality study was identified comparing a (met-) amphetmine-exposed infant group to a non-exposed population. A retrospective cohort study examined a metamphetamine exposed group (polydrug use including alcohol, nicotine, majijuana, but not heroin/cocaine) to a ‘non-exposed’ group. However, toxicology screening was not performed on the latter group, leading to the possibility that neonates in the ‘control’ group may have been drug-exposed. Results showed that 49% of infants in the metamphetamine-exposed group had evidence of withdrawal symptoms, and 4% required pharmacologic intervention. However, no detailed breakdown of how scores on the neonatal abstinence scoring system were allocated was provided. Oei examined a polydrug-exposed infant population and found only 1 out of 111 infants exposed exclusively to amphetamines required pharmacological intervention for treatment of withdrawal symptoms (compared to 23.8% of infants exposed to a variety of other drugs including opiates and benzodiazepines). Research into symptoms of cocaine and amphetamine withdrawal in neonates is characterised by inconsistent findings and methodological design flaws. Evidence regarding the effect of amphetamine exposure on the neonate is especially poor. Isolating the effect of in utero exposure to a particular illicit substance is problematic due to the high incidence of polydrug exposure among the target maternal population. The few good quality papers available suggest prenatal exposure to illicit stimulatory drugs produce very limited acute withdrawal symptoms in the neonate.

Clinical Bottom Line

► Prenatal exposure to cocaine produces acute withdrawal symptoms that are, at worst, mild and transitory (Grade A). ►Infants exposed to amphetamines antenatally are less likely to require pharmacological intervention for withdrawal symptoms than infants exposed to opioids (Grade B).

References

1. Oei J, Abdel-Latif ME, Clark R, et al. Short-term outcomes of mothers and infants exposed to antenatal amphetamines. Arch Dis Child Fetal Neonatal Ed 2010; 95: F36-41.
2. Bauer CR, Langer JC, Shankaran S, et al. Acute neonatal effects of cocaine exposure during pregnancy Arch Pediatr Adolesc Med 2005; 159:824-34
3. Ogunyemi D, Hernández-Loera GE The impact of antenatal cocaine use on maternal characteristics and neonatal outcomes. J Matern Fetal Neonatal Med 2004; 15: 253-9
4. Chomchai C, Na Manorom N, Watanarungsan P, et al. Methamphetamine abuse during pregnancy and its health impact on neonates born at Siriraj Hospital, Bangkok, Thailand. Southeast Asian J Trop Med Public Health 2004; 35:228-31
5. Myers BJ, Dawson KS, Britt GC, et al. Prenatal cocaine exposure and infant performance on the Brazelton Neonatal Behavioral Assessment Scale. Subst Use Misuse 2003; 38:2065-96
6. Smith L, Yonekura ML, Wallace T, et al. Effects of prenatal methamphetamine exposure on fetal growth and drug withdrawal symptoms in infants born at term J Dev Behav Pediatr 2003; 24:17-23
7. Bada HS, Bauer CR, Shankaran S, et al. Central and autonomic system signs with in utero drug exposure. Arch Dis Child Fetal Neonatal Ed 2002; 87: F106-12
8. Held JR, Riggs ML, Dorman C The effect of prenatal cocaine exposure on neurobehavioral outcome: a meta-analysis Neurotoxicol Teratol. 1999; 21: 619-25