Neonatal Abstinence Syndrome

Etiology

The clinical features of NAS result from chronic in utero exposure to certain drugs or substances of abuse followed by subsequent abrupt cessation of exposure at birth. Research cites several drugs as possible etiologies, but traditionally, opioids have been the most common agent identified. Among opioids, heroin or morphine has been the most commonly abused drug since the 1800s. However, with improved drug manufacturing over the decades, neonatal withdrawal from prescription opioids like hydrocodone, oxycodone, and opioid-containing combination products has also been well documented. Likewise, NAS can also result from synthetic and semi-synthetic opioids such as methadone and buprenorphine, which have been the standard of care for opioid addiction (medication-assisted therapy or MAT) during pregnancy.[5][3] This is important to identify during antepartum care since patients are encouraged to continue MAT during the pregnancy. It has been shown that maternal opioid use during pregnancy poses a higher risk to the fetus than MAT. A 2023 study showed shorter hospital stays in infants born to mothers in the MAT group when they didn't require pharmacologic management themselves for neonatal withdrawal syndrome.[6] In comparing the 2 traditionally used medications for MAT- buprenorphine-naloxone versus methadone during maternal antepartum care with the postpartum presence and treatment of NAS, several important findings were noted. The neonates born to mothers using buprenorphine-naloxone had a lower incidence of neonatal opioid withdrawal syndrome (NOWS). At the same time, methadone-exposed infants were also 6 times more likely to require treatment with >1 medication for NOWS.[7] 

During pregnancy, psychiatric medication usage like antidepressants such as selective serotonin reuptake inhibitor (SSRI)/serotonin-norepinephrine reuptake inhibitor (SNRI), and antipsychotics, both typical and atypical, have been implicated in signs and symptoms similar to neonatal withdrawal syndrome. However, it is important to exercise caution when interpreting these symptoms as these studies have used Finnegan's scoring system, which is only designed for the evaluation of neonatal opioid withdrawal. Several other substances like benzodiazepines, nicotine, alcohol, methamphetamine, and inhalants may also produce signs and symptoms consistent with NAS but are less studied and have fewer guidelines in terms of treatment.[8] There is also a significant association between psychiatric diagnoses and substance use disorders. 

In addition to the possibility of being the sole etiology for NAS, substances like gabapentin, benzodiazepines, cocaine, and nicotine also alter the severity of NAS symptoms. Polysubstance abuse during pregnancy further potentiates the severity. Marijuana is the most commonly abused substance during pregnancy, and its effect on modifying the severity of NAS is unclear.[3][9] It is well known that Δ9-tetrahydrocannabinol (THC), the psychoactive compound in marijuana, agonizes endogenous cannabinoid receptors that are present in fetal brains at an early stage of development. These receptors play a pivotal role in the development of the fetal CNS. THC readily crosses the placenta, and in the more recent changing psychosocial culture and legalization of marijuana, neonatal exposure has increased without any clear, separately identifiable withdrawal syndrome. However, hyperactivity, memory, and learning difficulties persist into childhood for children exposed in utero.[10]

Epidemiology

The incidence and prevalence of NAS require an understanding of maternal substance use epidemiology during pregnancy. Globally, as of 2016, the highest prevalence of any substance use disorder was in Australia and Asia, while opioid dependence disorder was highest in North America.[11] Throughout the world, the prevalence of OUD increased by 47% from 1990 to 2016, representing the highest increase among any substance abuse disorder. Even though the increase was more among men, approximately one-third of all SUDs involved women. Estimates are that 10 million women are affected by opioid dependence as of 2016.[11] 

In the US, the prevalence of opioid use during pregnancy increased by 333% from 1999 to 2014 and continues to rise.[12] Among women in the US who had Medicaid, approximately 1 in 4 women (21.6%) were prescribed opioids during pregnancy. The increase in the prevalence of substance abuse in pregnancy, especially opioids, has correlated with an increase in NAS incidence. In the years 2004 to 2014, the incidence of NAS among babies insured by Medicaid dramatically increased by 5-fold (1.3 in 1000 to 5.8 in 1000 births) as per hospital data from the National Inpatient Sample, which has data from 97% of the population in the United States.[13] This data is equivalent to diagnosing 1 newborn with NAS every 25 minutes. More recent evidence from a study of 24 tertiary care hospitals has reported an incidence of NAS as high as 23 per 1000 births.[14] A survey of 28 states found regional differences in NAS incidence; the highest rates were West Virginia, Vermont, and Kentucky.[15][16] Even in the states where NAS reporting is mandatory, variations in case definitions and required data result in difficulty in surveillance and quantifying NAS's precise incidence.[17] Hence, there must be standardization of NAS definition across states to address the opioid crisis further.[18]

The rising incidence of NAS has resulted in increasing admission rates to NICU, from 7 per 1000 to 27 per 1000 cases.[19] As a result, the hospital costs for managing NAS have also increased almost 7 times, costing $462 million in 2014, just among the population covered by public insurance.[13] Tennessee had a 5-fold increase in hospital admissions for NAS over a decade.[20] The incidence of NAS rose from 21,732 in 2012 to 32,128 in 2016, while hospital charges have tripled to $2.5 billion.[21][22] The incidence of NAS also rose from 4.6 to 6.7 per 1000 in-hospital births between 2012 and 2016.[23][22] A recent study from Wisconsin showed an association of maternal OUD with infant mortality independent of the presence of NAS and maternal opioid treatment during pregnancy.[24] Given the long-term consequences of NAS, both medical and social, lawmakers must address the opioid epidemic and the resulting NAS epidemic.

Pathophysiology

NAS is influenced by many factors, including maternal-fetal-placental pharmacokinetics, neurotransmitter dysregulation, and genetic and epigenetic factors. The pathophysiology underlying NAS is not entirely understood. However, rat models show differences between neonatal and adult withdrawal processes. Although several mechanisms have been proposed as potential pathways, there appears to be a complex interaction between the various neurotransmitters, such as dopamine, serotonin, and glutamate. The lack of adequate expression of some opioid receptors (kappa and delta receptors) and the differential maturation of neuronal circuits all seem to play a role.

Genetic mechanisms for differences in NAS expression have additionally been suggested. Single nucleotide polymorphisms (SNP) in the genes for opioid receptors (mu - OPRM1, delta - OPRD1, and kappa - OPRK1) and dopamine metabolism pathway (catechol-O-methyltransferase [COMT] gene) are associated with shorter hospital stay and duration of treatment. Studies evaluating SNPs in the genes for opioid metabolism, like the ABCB1 gene and cytochrome P450 (CYP) genes like CYP2B6, CYP2D6, and CYP34A, have been inconclusive. Besides the genetic mechanisms, epigenetics, such as DNA methylation, resulting in decreased expression of OPRM1 genes, has also correlated with increased NAS expression severity.[25]

Several neurotransmitters play a role in NAS, but most signs and symptoms of NAS are secondary to increased noradrenergic output from the locus ceruleus, which is the primary site for noradrenaline production in the brain. Chronic opioid stimulation in utero and sudden cessation post-birth lead to increased norepinephrine levels through upregulation and super-activation of intracellular cyclic adenosine monophosphate (cAMP). Through the same pathway, other neurotransmitters are also altered and believed to cause NAS's clinical manifestations. For example, decreased serotonin causes sleep deprivation, decreased dopamine causes hyperirritability, and increased acetylcholine causes cholinergic symptoms such as sneezing and diarrhea in NAS babies. Increased corticotropin release from activation of the hypothalamic-pituitary-adrenocortical axis has also been noted to cause stress and hyperphagia during the opioid withdrawal process in neonates.

The mechanism for signs and symptoms of withdrawal from other drugs seems to be almost similar to opioid withdrawal, although some distinction exists. Withdrawal symptoms from in utero exposure to SSRI/SNRI may be secondary to excessive serotonin and norepinephrine. Neonatal TCA withdrawal may be due to excess cholinergic transmitters. Withdrawal symptoms from in utero exposure to benzodiazepines are likely due to increased gamma-aminobutyric acid (GABA).[26][3]

A recent study demonstrated that more than 30 days of opioid exposure in utero resulted in NOWS regardless of when in the pregnancy the infant was exposed. It also concluded that any third-trimester exposure resulted in NOWS regardless of duration of exposure.[27] However, opioid use in pregnancy rarely occurs in isolation. The use of additional substances during pregnancy makes it difficult to establish the long-term effects of in utero opioid exposure and neonatal withdrawal. For example, alcohol use during pregnancy leads to a well-known disorder - fetal alcohol syndrome (FAS). Ethanol is a known teratogen, and FAS is the leading cause of preventable intellectual disorders in the United States.[28] Especially in this situation, it becomes increasingly difficult for clinicians to disentangle the short- and long-term effects of opioids versus other substances during pregnancy.[29]

Toxicokinetics

The placenta generally acts as a shield for the fetus, providing some degree of protection from maternal exposures. The extent of protection the placenta provides is based on various factors, including volume of distribution, time to peak onset of substances, clearance, molecular weight, lipid solubility, binding to maternal and fetal proteins, and transporter-mediated efflux. Prolonged exposure to opioids is known to downregulate neurotransmission in the limbic system. Opioid-dependent changes to the NMDA receptors contribute to excitatory changes in the limbic system, which is thought to contribute to the addictive properties of these drugs.[30] Studies examining in vivo pharmacokinetics of opioid use in pregnancy are scarce. In 1 study regarding morphine administration in pregnancy from 1990, authors found faster clearance and a shorter half-life in pregnant women when compared to non-pregnant women. In addition, the authors found that morphine transfer across the placenta occurred rapidly.[31] A recent study utilizing a chicken model for neonatal exposure to morphine and methadone documented the brain distribution and binding of μ-opioid receptors.[32] A 2024 human study showed that the concentrations of methadone and EDDP (2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine), a methadone metabolite, in maternal and cord plasma, do not correlate with the development of NOWS raising the concern that even at low doses, neonates are at risk for this potential complication.[33] 

Low molecular weight and lipophilicity allow opioid drugs to transfer across the placenta to the fetus easily. These properties also allow these drugs to accumulate in the developing fetal brain by crossing the blood-brain barrier. The severity of NOWS is influenced by various extrinsic and intrinsic factors and varies greatly among neonates. However, the extent of neonatal exposure likely plays the largest role. There appears to be a direct relationship between the gestational age and the amount of opioids transferred across the placenta. Variations in the ability to cross the placenta also exist among the opioids; semisynthetic opioids are transported across the placenta in a lower amount than synthetic opioids. Slow metabolism and longer half-life of opioids in the fetus may be a reason for severe withdrawal symptoms.[3] Infants exposed to methadone in utero can have a delayed presentation for NAS. Shorter-acting opioids cause a withdrawal syndrome in infants sooner after birth, but methadone-exposed infants may not develop symptoms until 72 to 96 hours after delivery.[34]

All opioids are present to some extent in breast milk but are not likely to be present in clinically significant amounts. Codeine should not be routinely used in pregnant and breastfeeding women. Neonatal toxicity has been documented in the form of drowsiness, apnea, and bradycardia in neonates exposed to codeine metabolites in breast milk. This is secondary to the portion of the population that is considered CYP2D6 ultra-rapid metabolizers. These women have the genetic predisposition to metabolize codeine to morphine at a faster rate, resulting in subsequent neonatal toxicity.[34] Even therapeutic use of opioids during labor and delivery has been shown to decrease breastfeeding success rates in neonates born to these mothers.[35] Doses as low as 80 mcg to 150 mcg may result in these undesirable breastfeeding patterns in newborns.[36][37] The toxicokinetics involved in in utero exposure or breastfeeding exposure are difficult to study, and the full extent of all the mechanisms involved is not completely understood. Future research is needed to better elucidate this unique area of interest.

History and Physical

NAS is a constellation of signs and symptoms with multisystem involvement. The clinical features are categorized into CNS, autonomic, respiratory, and GI system disturbances. Initial clinical presentation varies widely, but the most specific presentation for withdrawal is a triad of increased muscle tone, tremors at rest, and exaggerated Moro reflex. Other initial presenting symptoms include excessive high-pitched crying, irritability, sneezing, and diarrhea, all of which are due to an increased noradrenergic state.[38][3] The extent of symptoms, as well as time of onset, varies depending on the specific opioid used during pregnancy. Duration of use, frequency of use, and gestational age at the time of fetal exposure also play a role in the development of neonatal withdrawal. Maternal metabolism, placental metabolism, placental transfer, and excretion ultimately will determine the degree of exposure and, therefore, the propensity for developing neonatal withdrawal.[38]

CNS symptoms typically include hyperirritability, high-pitched crying, jitteriness, and tremors. Hyperirritability leads to sleep disturbances and difficulty maintaining a calm state.[39] Myoclonic jerks and seizures are also possible. The occurrence of seizures secondary to withdrawal has been debatable; in the past, the frequency of seizures in NAS has been between 2% and 11%. Interictal EEGs have found it to be myoclonic jerks rather than true seizures, and many recent studies corroborate the rare occurrence of seizures secondary to withdrawal.[40][41] Autonomic instability leads to changes in vital signs such as tachycardia, tachypnea, and hyper- or hypothermia. Skin perfusion becomes altered, leading to mottling and sweating. Frequent yawning, nasal stuffiness, excessive sneezing, and nasal flaring may also be present. Some of these may persist for months, especially in cases of withdrawal from buprenorphine.[38]

GI symptoms observed in NAS include poor feeding, regurgitation, vomiting, and diarrhea. Heroin withdrawal is more notable for severe GI symptoms and may lead to dehydration and electrolyte disturbances in neonates with an already delicate metabolic balance. Diarrhea may also lead to skin breakdown around the perianal/buttock region and rash. Withdrawal is a state of excessive caloric demand. Combined with GI losses and excessive autonomic stimulation, this can lead to weight loss and failure to thrive. Therefore, increasing caloric intake with feeding in neonates with clinical signs of withdrawal may be necessary.[3]

NAS's clinical features are also believed to be due to neurobehavioral dysregulation. An imbalance exists between the 4 behavioral components of infants- ie, state control/attention, autonomic control, sensory processing, and motor and tone control. When the infant spends more time in 1 subsystem, such as increased muscle tone, the energy consumed in another subsystem, such as attention, is minimal. This imbalance is also considered to be a hallmark feature. The neonate's self-regulatory capacity is lost, leading to hyperirritability and difficulty being consoled by the caregiver. This may also alter the infant's developmental trajectory.[4]

NAS has significant variability in the presenting symptoms and severity of the presentation. Inter- and intra-subject differences also exist.[8] Multiple factors can affect the timing and severity of presentation, like the type of drug exposure, the type of opioids exposed in utero, the timing of the last dose, cumulative dosage, and simultaneous exposure to other drugs of abuse. For example, withdrawal from short-acting heroin presents on the first day of life (24 to 48 hours), while withdrawal from long-acting buprenorphine (36 to 60 hours) and methadone (48 to 72 hours) typically presents in the first 3 days of life.[38][42] However, the withdrawal syndrome can continue as late as 4 weeks after birth. While antenatal buprenorphine exposure affects the timing of onset, it also correlates with less severe features and shorter duration of hospital stay.[43] The cumulative dosage of methadone is not directly related to the severity of NAS symptoms, even though doses <30 mg have correlated with a lesser degree of symptoms. Sedative hypnotic withdrawal has a wide range of onset, from 7 days for barbiturates, 12 days for diazepam, and 21 days for chlordiazepoxide. The delayed presentation is likely secondary to the active metabolites of these substances, which continue to exert their effects long after the last exposure to the parent compound. Polydrug exposure, especially the combination with benzodiazepines, simultaneous cigarette exposure, and male gender, has been associated with increased severity. Premature infants are less likely to express severe symptoms and typically have a milder course. Research has not yet determined why. This may be related to the shorter exposure time overall or lack of exposure at a gestational age closer to full term. It is also possible that it is related to the incomplete fetal development of the necessary neurologic pathways affected by substance exposure. More research is needed in this area. Breastfeeding correlates with less severe symptoms, shorter hospital stays, and less use of pharmacological treatment, with only trace quantities of opioids identified in breast milk.[44] Genetic factors are also likely to influence the severity of symptoms.[3][16]

Evaluation

NAS is a clinical diagnosis based on pertinent signs and symptoms in the setting of a high index of suspicion. A maternal history of drug use or abuse, taken in a nonjudgmental, nonthreatening, and caring manner, helps identify the majority of NAS cases. However, this history may not be readily available, and testing the infant may be necessary.

Several methods are available to identify and/or confirm the in utero drug exposure; the infant's urine, meconium, cord blood, or hair can be useful specimens to test. Testing the infant's urine or meconium for drugs is most commonly practiced because of its ease of collection and availability of rapid results. Urine testing can identify drug exposure only a few days before delivery. Depending on the type, the timing of the last dose, and the cumulative amount of drug exposure in utero, the maternal drugs may be identified in the infant's urine between 2 and 4 days. If there is a delay in collecting neonatal urine, there is a high chance of false-negative results. Hence, collecting the urine as soon as possible is important, preferably the first void urine. Meconium testing can identify drug exposure dating back to 20 weeks of gestation and is generally more sensitive than the urine test. Internal metabolism and clearance of drugs in both the mother and the fetus can also affect the result. Either of these tests can identify opioids, but there is a high rate of false positive and false negative results with synthetic and semisynthetic opiates. These xenobiotics often require a specialized test to accurately detect. Although these immunoassay screening tests are easy to perform in most hospitals, confirmatory mass spectroscopy tests are expensive and require advanced expertise. Hair and umbilical cord testing for drugs have also been evaluated but found to be less sensitive, have practical difficulties, and have limited clinical utility. The above tests have false positive and false negative results and require cautious interpretation. False-positive results for amphetamines are common when soap or alcohol is used before collecting the specimens or when meconium has urine contamination. Maternal usage of analgesics during the peripartum period can result in the infant's test being positive for opioids and, hence, also require cautious interpretation. False negatives can occur after marijuana exposure or improper storage of meconium. The best yield for confirmation is a combination of maternal urine and infant meconium testing followed by confirmatory gas chromatography-mass spectrometry.[3][26] It is important to note that the positive or negative results of these tests should not guide the clinical management of these infants. They should be assessed, scored appropriately, and treated if necessary. 

The assessment for babies with known or suspected fetal exposure must start at birth. Several tools have been developed over decades to assess the severity and the need to start, adjust, and wean interventions, specifically pharmacological agents. The Finnegan Neonatal Abstinence Scoring System (FNASS) and its modified versions have long been considered the "gold standard" for NAS assessment.[45] It continues to be widely used among the pediatric community even though in 1998, the American Academy of Pediatrics (AAP) recommended the 11-item Lipitz scoring tool due to its ease of use. FNASS, consisting of 21 clinical signs and symptoms divided into 3 categories, was originally developed for opioid-exposed infants and full-term babies. Scoring occurs approximately every 3 to 4 hours, and depending on the score, clinicians can decide on further management. A cumulative score of ≥8 is suggestive of significant withdrawal, which should prompt providers to consider starting treatment. However, the treatment threshold is based on scores in non-opioid exposed neonates, and there is no consensus based on well-designed studies to use these designated scores for starting treatment in exposed infants.[46][47] The FNASS and its modified versions have been adopted widely but have several limitations. Accuracy depends on the training of the rater, has high inter-rater variability due to the subjective nature of the items, includes items of low clinical significance like sneezing or yawning, and requires disturbing the infant for accurate assessment. In addition, it is a score developed to assess opioid withdrawal, which limits the utility when guiding clinical decisions for withdrawal from other substances. The lack of generalizability to preterm infants and its usage in withdrawal from drugs other than opioids limits its use.[48] Several studies have noted a longer hospital stay and pharmacological treatment using an FNASS-based protocol.[16]

Multiple scoring tools have been developed to overcome the limitations of FNASS.[49] A modified FNASS called MOTHER NAS removed the overlapping items and added irritability and failure to thrive, tallying to 19 items. This scoring has been the objective measure in some studies but still lacks validation.[50][43] Other scoring tools like Neonatal Withdrawal Inventory (NWI), Neonatal Narcotic Withdrawal Index, and Finnegan Neonatal Abstinence Syndrome Tool – Short Form, have been studied and noted to have high inter-rater reliability and almost 100% correlation with FNASS.[16] However, they are not widely used for reasons that remain unclear.[45]

Alternative assessment approaches have also been attempted. A new approach called the "Eat, Sleep, Console" model was more recently developed based on the basic functioning of infants, the family's involvement in the infant's care, and maximizing the nonpharmacological treatments before starting medications.[41] It was started as a quality improvement project to evaluate the ESC approach against the FNASS approach and found that only 12% of the infants received morphine using the ESC approach. This is 5 times less than the FNASS approach. Studies were conducted to validate the functional ESC model further and found a significant reduction in length of stay (LOS) and the need for pharmacological treatment.[51][43] It remains a relatively new approach, and more well-designed randomized clinical trial  (RCT) studies are required to corroborate the findings, including the long-term outcomes. Besides the ESC approach, scoring tools with objective parameters are also available. For example, short-scoring scales based on skin excoriations, muscle tone and tremors, salivary cortisol level, pupillary size, serum levels of brain-derived neurotrophic factor (BDNF), and skin conductance exist. Such tools have some clinical significance, but their application in day-to-day management is subject to limitations by practical difficulties and limited data that validate their significance.[16]

Treatment / Management

Management options for NAS can be broadly divided into nonpharmacological, alternative, and pharmacological treatments.

Nonpharmacological Treatment

Treatment should always begin with nonpharmacological care while maintaining the mother-infant dyad and should continue even after discharge from the hospital. The goal of nonpharmacological treatment is to assist the self-organization of the neonate and support the neuro-maturation. The AAP has recommended it as the first line in the management of NAS.[38] These may include changes to the physical environment, like darkening the room and decreasing visual and auditory stimuli in the infant's surroundings. Techniques such as gentle vertical rocking, side-lying C-position, containment with hands held, swaddling, and swaying can be soothing and may help reduce irritability and hypertonicity.[42] Avoiding unnecessary tactile stimuli by clustering care and providing skin-to-skin is effective. Skincare using topical barrier cream for diaper dermatitis is necessary to reduce irritability. Pacifier use can decrease oral hypersensitivity. These infants also have difficulty maintaining the 3 hours of scheduled feeding and may become irritable secondary to hunger. Hence, it is imperative to provide frequent, small-volume, on-demand feeding to make the infant more comfortable.[52] Calorie-dense, thickened feeds are required to meet the energy requirements of 150 Kcal/kg/day in these infants to prevent growth failure and weight loss.[38] The use of partially hydrolyzed formula has not been shown to be beneficial in infants with NAS.[53] It is essential to understand that these infants have neurobehavioral dysfunction with disorganized behavior rather than adaptation problems. Individualizing the nonpharmacological care specific to the infant may be beneficial. Healthcare professionals should involve the mother and help her identify the interventions that alleviate the dysfunctional behaviors specific to her baby. Having the care for mother and infant in the same location, ie, parental rooming-in, promotes more time spent on skin-to-skin and breastfeeding, resulting in huge benefits such as a reduced LOS and decreased number of days on pharmacological treatment.[54][41](A1)

Since opioids are present in minute quantities in breast milk, breastfeeding should be strongly encouraged unless the mother has other absolute contraindications such as HIV, Hepatitis B or C, or current drug abuse. Support for breastfeeding has also received endorsement by various organizations, including the AAP. However, several barriers exist to breastfeeding among women in MAT programs. Efforts must be made to overcome those challenges and ensure the benefits of breastfeeding in the infant's neurodevelopmental outcomes.[4]

There may be a debilitation of maternal well-being in the perinatal period. Hence, multilevel professional care must be offered to the mother to maintain the mother-infant dyad. Despite many studies showing the benefits of preserving the dyad, significant barriers exist. For example, mothers report feeling guilty or having mistrust and judgmental behavior from the staff caring for their baby. Provider biases preventing the mother's involvement in the infant's individualized care also exist. To reap the benefits of this strong relationship, physicians must support and teach the parent nonjudgmentally to optimize the nonpharmacological care for their infant. This can be easily provided at the bedside by the parent and continued even after discharge from the hospital.[55][41] A multidisciplinary approach is also important to address the possibility of postpartum maternal psychiatric decompensation as well. (B2)

Alternative Treatment

Many alternative therapies have undergone evaluation for their use in NAS. Massage therapy, foot, auricular acupressure therapy, and Reiki have correlated with soothing effects and positive changes in the baby's vitals. A recent study on laser acupuncture has shown a reduction in LOS, although the LOS is comparatively longer than the national average. Besides these, ongoing studies are evaluating the effects of aromatherapy and music therapy.[56][55]

Pharmacological Treatment

Even though nonpharmacological treatment remains the first line, AAP endorses pharmacotherapy when indicated. While it is important to note that pharmacologic management is associated with a longer overall LOS in the hospital, treatment should not be withheld when needed in these infants.[38] There are reports that 27% to 91% of infants with NAS require pharmacological intervention.[3] The goal of pharmacological treatment is to improve the withdrawal signs and reduce the scores of FNASS or an equivalent.(B3)

Multiple studies, including RCTs, have been conducted to find the optimal medication. Still, the results have varied widely in LOS in hospitals and treatment duration. Despite these variations, opioids remain the preferred mainstay pharmacological agent in the treatment of NAS, although there is no clear consensus on the choice of opioids.[43][45] Paregoric (a drug with anhydrous morphine) and dilute tincture of opium were very popular in the past. However, due to its high alcohol content, toxic contents with multiple adverse effects, and very long LOS (even up to 79 days), these medications have been discontinued and are no longer used.[38] Morphine, a short-acting opioid, remains the most commonly used and preferred opioid based on the Pediatric Health Information System (PHIS) database.[14] It is dosed every 3 to 4 hours, which allows for easy adjustment in management, but it also means disturbing the infant more frequently. The dose is based on the severity of the symptoms and the scores, but generally, it starts at 0.04 mg/kg/dose with a maximum dose of 0.2 mg/kg/dose. Morphine treatment is associated with a longer LOS ranging from 5.9 to 42 days. This may be due to slower weaning. Current studies are exploring the use of other opioids.[56](B3)

Methadone, a long-acting synthetic opioid, has been evaluated as an alternative. It has the advantage of being dosed twice daily due to its long half-life, but that also limits the dose's frequent titration. The LOS had ranged from 16 to 44 days when using methadone. Recent evidence shows only a modest reduction of LOS with methadone compared to morphine.[57][58] The drug undergoes extensive metabolism by cytochrome 450 enzymes and interacts with multiple medications. Therefore, caution is needed, especially when used in combination with phenobarbital or antiviral medications.(A1)

Buprenorphine, a partial opioid agonist, has recently drawn attention because it correlates with a shorter LOS than morphine and methadone and its ease of use through sublingual administration. Even though the LOS range was between 12.4 days and 32 days, a recent RCT showed that the median LOS was significantly shorter by as much as 12 days compared to morphine.[50] It also correlated with a significant reduction in the number of pharmacologically treated days, which may be due to its partial agonism, functional antagonism, and relatively long half-life. However, caution is necessary when interpreting the results, as those infants predominantly suffered exposure to methadone in utero. Additional studies are needed to assess the utility of buprenorphine in infants exposed to opioids other than methadone. While 1 RCT showed promising results, more well-designed studies are necessary. The current drug formulation of buprenorphine with 30% ethanol might be a deterrent to its widespread use in infants.[55][59][50] Most recently, a study validated the previous findings and demonstrated that buprenorphine-exposed infants had decreased incidence of neonatal abstinence when compared to methadone, but once pharmacologic treatment was required for withdrawal, there was no statistical difference in the LOS between the 2 groups.[60] (A1)

Second-line pharmacotherapy for infants whose symptoms remain uncontrolled with maxed opioid treatment includes phenobarbital and clonidine. Again, there is no clear consensus that among the 2, 1 is more effective, but phenobarbital is used more frequently in the US than clonidine.[61] Both drugs can be used as an adjuvant to first-line opioids in severe NAS. Phenobarbital, an anti-seizure medication, may be more effective in cases of polydrug exposure and opioid and benzodiazepine combination exposures. However, its adverse effects, such as oversedation, uncertain long-term neurodevelopmental outcomes, and its high alcohol content, must merit consideration. Clonidine, an alpha-2 agonist, has been effective in reducing the duration of pharmacotherapy. Even though there is a risk of hypotension and bradycardia, a few studies showed a low occurrence rate of cardiovascular adverse effects, and preliminary data supports its use in neonatal withdrawal. However, it is crucial to know that the safety profile for both drugs has not been adequately established.[43](B2)

Naloxone, an opioid antagonist, is not recommended in the management of NAS because it can precipitate and worsen withdrawal symptoms. Its efficacy, dosing, and safety profile in neonates have also not received adequate study.[62](A1)

Pharmacotherapy can be weaned and discontinued once the NAS signs and symptoms are well-controlled based on the scoring tool used. Many institutions have used a 10% to 20% reduction in the daily dosing of opioids as the weaning protocol. However, institutions have no consistency, and the necessity of longer weaning protocols is also questionable.[63][43] This factor may play a role in the longer LOS. Many studies have called for standardizing the weaning protocol and pharmacotherapy regimen.[64][38] One quality improvement study has used 'as needed' treatment protocols and shown a reduction in the amount and days on pharmacotherapy.[1](B2)

Location of the Treatment

Babies with NAS have long received treatment in the NICU, and this has been a traditional practice rather than evidence-based. According to the PHIS database, about 87% of neonates with NAS were managed in the NICU.[14] While the NICU admission may be necessary because of the severity of symptoms or comorbidities, the environment is not ideal because of multiple distractions and could be overstimulating to these infants. Rooming-in, which promotes mother-infant bonding, skin-to-skin, and breastfeeding, is currently the preferred model of care as it has correlated with reduced severity of NAS.[54][56] This activity can be best achieved in the postnatal ward, and 1 study showed a consequent reduction in the duration of pharmacotherapy and LOS.[43] Another quality improvement study proved the same results in a general inpatient pediatric ward.[1] Thus, it may be time to reconsider the age-old practice of admitting all infants to a critical care setting to manage NAS. Some studies have also evaluated the possibilities of outpatient only and combined inpatient-outpatient management and found a relative reduction in LOS and costs with outpatient management. However, outpatient management was associated with increased days on pharmacotherapy, and concerns exist among physicians in handing over opioid prescriptions to caregivers of those infants who are at risk of relapse.[16](A1)

Differential Diagnosis

It is vital to consider other diagnoses before attributing the clinical features to NAS, even if there is a clear history of prenatal drug exposure. In some cases where maternal history is unyielding or unclear, physicians can mistake some signs and symptoms of NAS for a different condition. For example, autonomic instability leading to mottling, sweating, tachypnea, and changes in temperature in NAS can be mistaken for sepsis.[38] Irritability may be due to brain injury or GE reflux. Fever can be secondary to hyperthyroidism or sepsis. Jitteriness may be secondary to hypoglycemia or hypocalcemia. Poor feeding may occur due to oromotor dysfunction, congenital anomalies, or prematurity. Seizures secondary to withdrawal, even though rare, must be differentiated from CNS infections, hypoxic-ischemic encephalopathy, and electrolyte disturbances. Myoclonic jerks, often mistaken for seizures in NAS, need to be identified accurately to avoid unnecessary confusion and testing. In addition, signs of early adaptation in a healthy neonate secondary to pain, hunger, improper handling, and suboptimal physical environment must also merit consideration.[4] When assessing an infant for NAS, care must be utilized to rule out other causes or rule in NAS's diagnosis. Maternal history, such as medication use in pregnancy, substance misuse, adequacy of prenatal care, unexplained fetal demise, and psychiatric diagnosis, require inclusion. Neonatal birth history, including gestational age, birth weight, presence of a perinatal insult, or other congenital anomalies, must be thoroughly reviewed. When all signs and symptoms are consistent with NAS, and there is a maternal history of drug use/misuse, confirmatory testing in the neonate is the suggested approach to corroborate the diagnosis.

Essential diagnoses to consider in evaluating a patient with NAS include the following:

• Sepsis
• Birth trauma
• Gastrointestinal reflux
• Hyperthyroidism
• Hypoglycemia
• Hypocalcemia
• Hypoxic-ischemic encephalopathy

Prognosis

There is an association between infants with prenatal opioid exposure and poor long-term outcomes. Research has documented evidence of changes in neurodevelopment, cognition, school performance, behavior, and vision as well as increased mortality in these infants. However, much of this data comes from observational studies and therefore has limitations. The presence of multiple prenatal confounders such as polysubstance exposure, prematurity, intrauterine growth retardation (IUGR), and postnatal confounders like NAS with or without pharmacotherapy for treatment, continued maternal drug abuse, maternal socioeconomic and environmental factors makes it challenging to attribute prenatal opioid exposure as the sole cause for the poor outcomes.

Studies have consistently shown a significant difference in neurodevelopmental outcomes in opioid-exposed infants versus their counterparts.[65] These findings are consistent from 12 months of age through adolescence, although most data does not support a significant difference prior to one year of age. Several evaluation tools, such as the McCarthy Motor Scale and Vineland Social Maturity Scales, have consistently identified these differences, but there are conflicting results with Bayley Scales of Infant Development.[66] Visual-motor problems like strabismus and nystagmus and visual acuity impairment secondary to refractive errors are common among infants with NAS. Significant cognitive deficits like lower aggregate IQ scores, poor verbal performance, impaired short-term memory, and impaired executive functioning have been found in children more than three years of age who had a history of prenatal opioid exposure when compared to controls. Deficiencies in school performance in NAS children have traditionally manifested with low mean academic test scores in every grade and domain, with the most significant difference noted in seventh grade. Experiments have sought to understand the mechanism behind poor neurodevelopmental and cognitive outcomes in this population. Preliminary data showed a regional decrease in brain volume, alteration in the maturation of connective neuronal tracks, and relative microcephaly among opioid-exposed infants that may underlie these adverse changes in NAS infants.

Prenatal opioid exposure has been linked to psycho-behavioral problems as well. These children are two times more likely to be diagnosed with attention-deficit/hyperactivity disorder, conduct disorders, and adjustment disorders. They are also more likely to display aggressive behavior and have anxiety disorders compared to their controls. Infants with NAS are reportedly at increased risk of rehospitalization during childhood due to maltreatment, trauma, and behavioral problems as well. The mortality rates have been consistently higher among opioid-exposed infants for many decades and across many developed countries, with rates as high as four times the normal rate in New York City. These indicate the need for continued support for this highly complex and vulnerable population. A recent study showed an association between perinatal opioid exposure and the development of asthma and eczema in neonates. It also showed an association between maternal opioid use and the risk of perinatal infection, preterm delivery, and low birth weight.[67]

Despite all these findings, a large randomized control trial called MOTHER (Maternal Opioid Treatment: Human Experimental Research) evaluated the neurodevelopmental and growth outcomes among NAS infants and found conflicting long-term outcomes.[68] The research studied infants exposed to buprenorphine or methadone in utero and followed them until three years of age. They were assessed at regular intervals to monitor their neurodevelopment using a battery of tests. The authors found no significant difference in neurodevelopmental and growth outcomes among opioid-exposed infants. They concluded that buprenorphine or methadone exposure did not have any harmful effect on physical, mental, and behavioral development in those children. One may argue that the close follow-ups with health professionals may have positively impacted the mother-infant dyad by offering consistent support to the mother, thus resulting in an enriched home environment that allowed the normal development of the infants.[3][16] In addition, conclusions regarding this population are limited by the mother's enrollment in medication-assisted treatment programs. It is possible that illicit opioid use and the accompanying psycho-social factors would yield different outcomes. 

Complications

Opioid-exposed babies are more likely to be born premature, have intrauterine growth restriction, and have low birth weight. This is most commonly documented in neonates exposed to heroin. As alluded to earlier, the hypermetabolic state, combined with difficulties in feeding and increased intestinal losses, put these infants at high risk for growth failure. NAS has been associated with life-threatening complications such as seizures, which are unique to opioid withdrawal in this particular population. Seizures are most common in methadone-exposed infants, although still a rare occurrence. Ultimately, NAS babies are at higher risk for death.[42][3]