Neonatal abstinence syndrome due to prolonged administration of fentanyl in pregnancy


Correspondence: Dr J. Regan, Our Lady's Hospice, Harold's Cross, Dublin 6W, Ireland.

Case report

A 31 year old woman was admitted at 38 weeks' gestation for an elective caesarean section. Four years previously she had sustained soft tissue injuries to her cervical and lumbar spine following a road traffic accident. Nonsteroidal anti-inflammatory agents, anti-spasmodics and physiotherapy failed to control her pain, and she was started on transdermal fentanyl patches. Pre-pregnancy counselling addressed the problems of neonatal opioid dependence and the unknown teratogenic effects of fentanyl. Prior to conception, her sole medication was transdermal fentanyl at 125 μg/h.

Her pregnancy proceeded uneventfully. She continued to use the transdermal patch (125 μg/h) up to delivery. At 38 weeks' gestation, a planned elective lower segment caesarean section was performed. General anaesthesia was induced using thiopentone, and muscle relaxation was achieved by suxamethonium and atracurium. Anaesthesia was maintained with isoflurane in a mixture of oxygen and nitrous oxide. No opioids were administered prior to delivery. Surgery proceeded uneventfully and a healthy female infant weighing 3.13 kg was delivered. Apgar scores at one minute and five minutes were both recorded as nine. The baby's head circumference was 35.4 cm, which is on the 50th centile. The baby was observed closely in the neonatal special care unit for five days. She was bottle-fed formula milk and therefore was not exposed to fentanyl in breast milk. At 24 hours, she was noted to be ‘jittery, fisting and irritable, with a high-pitched cry’.

Her mean Finnegan scores at day one (day of birth) and day two averaged 4.5, with a single peak at 72 hours (score = 11). Thereafter, these scores gradually diminished, with scores of < 4 consistently recorded on day four. Table 1 shows the maternal fentanyl levels predelivery, and the neonatal levels in the first 48 hours.

Table 1.  Maternal and neonatal fentanyl levels pre- and post-delivery.
SampleTimeBlood concentration (ng/mL)
BabyAt birth1.23
BabyDay 10.31
BabyDay 2≤0.25

The baby was treated conservatively and did not require pharmacological intervention. Initially she was slow to feed but by day five was taking formula milk well. By 96 hours, the baby demonstrated no signs of opioid withdrawal, and by day ten both mother and child were discharged home. At her six week check-up she was developmentally normal and was reported to be sleeping and eating well. Subsequent follow up has confirmed that to date the baby has reached normal developmental milestones.


Neonatal abstinence syndrome is a well documented condition resulting from prenatal exposure both to legal and illegal drugs1. With increasing use of opioids for the treatment of chronic benign pain2,3, an increase in the number of infants being treated for complications of intrauterine opioid exposure can be anticipated. Fentanyl is a synthetic opioid agonist with a potency 50–100 times that of morphine4. As a transdermal formulation, it may be administered in incremental doses of 25 μg/h. The transdermal route of delivery offers the potential benefits of simplicity, efficacy and acceptance by the patient5.

Previous case reports have highlighted the problems of acute exposure of the fetus to fentanyl in the uterus when administered via an epidural catheter. To date, no reports are available regarding prolonged gestational exposure to fentanyl. Placental transfer of highly lipophilic molecules, such as fentanyl, from the mother to the fetus is efficient because the placenta has a very thin lipophilic membrane, with a large surface area for exchange and high maternal and fetal blood flow6. Transplacental equilibration of fentanyl is thought to be achieved in a single circulation7. Recent work by Fernando et al.8 on the placental transfer of fentanyl during combined spinal and epidural analgesia for labour confirmed that similar concentrations of fentanyl exist in both maternal and placental circulation. In our case, use of fentanyl by the mother was constant and consistent throughout pregnancy at 125 μg/h, and plasma levels within the range 3.56–3.73 ng/mL were recorded prior to delivery. We can assume, based on fetal fentanyl levels at delivery, that the fetus was exposed to similarly high concentrations of fentanyl throughout the pregnancy.

Neonatal abstinence syndrome describes the constellation of symptoms exhibited by infants born to mothers who are dependent on opioids during pregnancy9. It has been well described in infants exposed to heroin and methadone1. Salient features include: signs of neurological dysfunction (e.g. poor feeding, uncoordinated and constant sucking, vomiting, diarrhoea and dehydration); autonomic signs (e.g. increased sweating, nasal stuffiness, fever, mottling); poor weight gain; and skin excoriation due to excessive rubbing9. Symptoms from mild to severe are manifest in at least 70%–90% of infants passively exposed to opioids through maternal ingestion during pregnancy1. Onset depends on various factors, including the time of the last dose before giving birth, the quantity of drug taken and the capacity of the newborn infant to metabolise and excrete the drug. Usually the symptoms of heroin withdrawal start 24–48 hours after birth and of methadone withdrawal within the first four days; with methadone withdrawal the symptoms last longer10. Neonatal seizures are more frequent following withdrawal of methadone1. A dose of 125 μg/h of fentanyl is the equivalent of 450 mg of methadone per day.

The relationship between maternal methadone levels, neonatal methadone levels and neonatal withdrawal has not been defined consistently11,13. Doberczak et al.11 studied 21 women who were dependent on methadone and demonstrated a positive correlation between maternal plasma and neonatal plasma levels of methadone. Neonatal levels were consistently lower than maternal levels, possibly due to rapid sequestration of placentally transferred methadone to fetal tissues. Higher neonatal methadone levels were associated with more rapid declines of drug levels and the rate of decrease in serum methadone levels correlated significantly with the severity of central nervous system withdrawal signs. Peak withdrawal scores were reached on the third and fourth day. In our patient fentanyl levels decreased exponentially with time and were last recorded as < 0.25 ng/mL 48 hours after birth. The clinical features of opioid withdrawal were most prominent on day 3, which is consistent with low plasma levels seen with fentanyl withdrawal. Katz and Kelly12 observed signs of withdrawal in infants after discontinuing fentanyl infusions. Some infants exhibited signs of withdrawal as the fentanyl infusion was decreased slowly, while others had their highest neonatal abstinence scores more than 48 hours after fentanyl was discontinued.

Fentanyl is a highly lipid-soluble compound that is cleared by N-dealkylation and hydroxylation in the liver with approximately six percent excreted as unchanged fentanyl by the kidneys. It is the free nonionised form of fentanyl that has pharmacological activity. However, the protein bound component may act as a store for the drug, thus potentially prolonging its effect4. The clearance of fentanyl in newborn infants is a function of two factors: blood flow to the liver and the ability of the liver to extract fentanyl, which depends on the maturity of the cytochrome P450 system. Neonatal liver blood flow may be compromised by patency or reopening of the ductus venosus, which results in diversion of portal blood flow away from the liver13. In addition, neonatal pharmacokinetics are influenced by their relatively larger total body water content, extracellular fluid compartment, brain and liver when compared with adults. Koehntop et al.14 demonstrated mean values of terminal β half-life, total body clearance, central compartment volume and volume of distribution a steady state to be greater for neonates than similar values for adults with fentanyl.

Finnegan's scoring method for assessment of acute opioid withdrawal in newborn infants is based on nursing observations, with each sign and symptom assigned a numerical scale (scores: 0–7 mild, 8–11 abstinence, 12–15 severe abstinence syndrome). Drug treatment is usually given for Finnegan abstinence scores of ≥ 8 (for an average of three scores) or for two scores > 1215,16. In our case, a peak value of 11 was recorded on day three, with an average score of 6, and the infant did not require drug therapy. By day four, the Finnegan scores averaged 3, and thereafter the infant showed no signs of opioid withdrawal.

The prescription of fentanyl has increased for chronic, benign pain, despite the lack of controlled trials and data on its long term effects. Transdermal administration of fentanyl has the advantage of maintaining a relatively constant serum level for a prolonged period17.

Despite a high maternal dose of fentanyl, the infant developed a mild withdrawal syndrome with no apparent long term effects at 31 months of age. It is important that similar cases of neonatal syndrome due to chronic exposure to fentanyl in pregnancy are reported.