Persistent Feeding Difficulties Among Infants with Fetal Opioid Exposure: Mechanisms and Clinical Reasoning

Abstract

AIMS:

Infants with fetal exposure to opioids have varying pattern of feeding difficulties mainly manifesting as difficulties with aerodigestive adaptation and disruptive feeding behavior. The reasons are unclear; in a pilot study, we determined basal and adaptive pharyngo-esophageal motility in a group of infants with fetal exposure to opioids and persistent feeding difficulties impeding their discharge.

METHODS:

Six infants with fetal opioid exposure compared to 12 controls who underwent basal and adaptive pharyngo-esophageal manometry to characterize the basis for their symptoms. Spontaneous swallows (N=180) and pharyngeal stimuli (N=113) induced swallowing responses were analyzed.

RESULTS:

Resting upper esophageal sphincter (UES) pressure was similar in both groups, but resting lower esophageal sphincter (LES) pressure was significantly high and it relaxed slowly and inadequately in opioid exposed infants (P <0.05). Upon pharyngeal provocation, opioid exposed infants had higher LES nadir pressure, increased duration of esophageal body contraction at proximal-, mid-, and distal- esophagus, as well as greater area under the curve with distal esophageal waveforms, compared to controls (all P <0.05).

CONCLUSIONS:

These pilot observations are suggestive of up-regulation of central vagal effects with heightened cholinergic excitatory responses and inadequate relaxation responses at the forgut, and may form the basis for persistent feeding difficulties in opioid exposed infants. Abnormality with both sensory and motor aspects of vagal reflexes may be contributory.

INTRODUCTION

Use and abuse of opioids among pregnant women continues to escalate in the United States with an associated increase in neonatal abstinence syndrome (NAS).¹ Over the last decade, the incidence of NAS in opioid exposed infants has increased over five-fold,^2,3 and the reported incidence of NAS in opioid exposed infants ranges from 55–94%.^4,5,6 Severity of NAS depends on a variety of factors, such as degree of immaturity, duration of exposure, and type of exposure.^4,7 In infants diagnosed with NAS, 18.1% had feeding difficulties compared with 3% in a non-opioid exposed population.³

Opioids activate μ-opioid receptors in the central nervous system as well as in the enteric nervous system. Opioids rapidly cross the placenta⁸, and fetal exposure decreases brain growth and alters the maturation of neural connective tracts, but exact effects on neural pathways and opioid receptors remain unknown.^9,10,11 Opioids affect gastrointestinal (GI) functions manifested as regurgitation, gastro esophageal reflux (GER), nausea, emesis and delayed evacuation, although mechanisms remain unclear in human infants.⁴ It has been observed that maternal opioid use during fetal development may lead to altered neuronal characteristics, swallowing and feeding difficulties,¹² manifesting as aerodigestive symptoms, poor feeding, uncoordinated and repetitive sucking, GER, vomiting, and poor weight gain, often leading to empiric acid- suppressive therapies or dietary changes.^4,5 Abnormal modulation of the central and enteric nervous systems has been implicated in esophageal dysmotility and related aero-digestive symptoms,¹³ however, the mechanistic basis of these symptoms or of dysfunctional pharyngo- esophageal motility in infants with NAS remains unclear.

In this pilot study, we determined basal and adaptive pharyngo-esophageal motility in a group of infants with fetal exposure to opioids and persistent feeding difficulties delaying their discharge. To generate the hypothesis that fetal opioid exposure alters infant pharyngo-esophageal neuromotor characteristics attributing to the feeding difficulty-phenotype, we compared the findings from the predominantly opioid exposed group with gestational age matched infants with no exposure to opioids.

METHODS

Subjects:

Infants with fetal opioid exposure manifesting severe feeding difficulties, thus delaying their hospital discharge that were referred for gastrostomy evaluation at the Neonatal and Infant Feeding Disorders Program, Nationwide Children’s Hospital, Columbus, Ohio, were included. Opioid exposure was determined by maternal history, obstetric data, and laboratory evaluation. We excluded infants receiving opioid or sedative medications, suffering from withdrawal due to iatrogenic opioid exposure, or diagnosed with congenital aerodigestive or gastrointestinal tract anomalies, or grade III/IV intraventricular hemorrhage. Healthy age matched controls are a difficult population to recruit for pragmatic and ethical reasons. Therefore, convalescing premature infants with no prior opioid drug exposure who had good oral feeding ability, and have participated in our maturational study protocols served as the comparison group. The studies were performed after obtaining written, informed consent from parent(s) and approval by Nationwide Children’s Hospital Institutional Review Board.

Manometry Methods:

As previously published, pharyngo-esophageal basal and adaptive characteristics were collected using a pharyngo-esophageal manometry system (Medical Measurement Systems, Dover, NH, USA).^14,15 An esophageal manometry catheter assembly was connected to the system that consisted of sensors to record pressures from the pharynx, esophageal body (at proximal-, middle-, and distal- esophagus) and stomach, along with sleeves to record upper and lower esophageal sphincter pressures (UES and LES respectively), and an infusion port at the level of the pharynx for sterile water stimuli (Dentsleeve International, Mui Scientific, Ontario, Canada). Pharyngeal infusions of sterile water were administered in volume increments of 0.1 ml, 0.3 ml and 0.5 ml. We analyzed the adaptive responses induced by pharyngeal infusion including pharyngeal reflexive swallow or pharyngo-UES-contractile reflex.¹³ Subject safety was monitored throughout the study.

Basal and adaptive pharyngeal swallow characteristics between the two groups were evaluated and compared, based on the previously established criteria.^14,15,16 For each subject 10 basal swallows were analyzed for UES, LES, esophageal body characteristics (contraction amplitude, contraction duration at proximal-, middle-, and distal-esophagus, and propagation patterns as complete/incomplete/failed peristalsis). For adaptive pharyngo-esophageal responses, in addition to changes in UES, LES, and esophageal body characteristics, we also recorded the prevalence of response type to provocation stimuli (pharyngeal reflexive swallow or pharyngo- UES-contractile reflex or none).¹⁷

Statistical Analysis:

Discrete variables were compared between the groups using chi-square test or Fisher’s Exact Test, whereas two-sample t-tests were used to analyze the intergroup differences for continuous variables. Non-normally distributed group comparisons were made using nonparametric tests (Wilcoxon rank-sum test). Linear mixed effect models were applied for the comparisons between controls vs. exposed groups and basal vs. adaptive swallows when reporting UES, esophageal body, and LES characteristics. Generalized Linear models with generalized estimation equations were used to compare categorical outcomes between groups. Mean ± SD values are reported, unless stated otherwise. With non-normally distributed data, values are shown as median (IQR). A P-value < 0.05 was considered significant. SAS version 9.3 (SAS Institute, Cary, NC) was used for statistical analysis.

RESULTS

Six infants with fetal opioid exposure (4 infants exposed through maternal opioid medical support program with either methadone or suboxone, 2 infants born to mothers who received chronic prescription opioids during pregnancy) were compared to 12 control infants. Of those infants exposed to opioids, two were diagnosed and treated for NAS with opioids ranging from 13–21 days, but none were on any opioid or sedative medications at study. However, five of the six opioid exposed infants had varying fetal exposure to other drugs as well (benzodiazepines, marijuana, tobacco, zolpidem or SSRI). Demographic and outcome characteristics for both groups are listed in Table 1. There were no statistically significant anthropometric differences (weight, length and head circumference) found between opioid and control populations at the time of birth or at study, or at NICU discharge. Although PMA was identical between the groups at study, the feeding methods were distinct, i.e., gavage-fed: transitional fed: oral-fed were 1/3 each in opioid-exposed infants, compared to 90% oral-fed in controls (P=0.06) (Table 1). At discharge, feeding outcomes, i.e. oral/gastrostomy: oral among opioid exposed vs. controls respectively were 3:3 vs. 1:12 (P=0.08) (Table 1).

Table 1:

Demographic and outcome characteristics of opioid-exposed infants vs. controls

Characteristic	Opioid Exposed (N=6)	Control (N=12)	P value
At birth
Gestational Age, weeks	29.8 ± 3.9	30.6 ± 1.3	0.73
Weight, kg	1.36 ± 0.80	1.66 ± 0.90	0.49
APGAR at 1 min*	5(3–6)	6(6–8)	0.33
APGAR at 5 min*	7(3–7)	8(8–9)	0.24
Maternal smoking (%)	2(33.3)	2(16.7)	0.57
At study
Postmenstrual age, weeks	42.5 ± 4.9	42.0 ± 3.5	0.85
Weight, kg (Range)	3.2(2.0–3.8)	3.3(2.1– 4.0)	0.89
Feeding method (gavage: transitional: oral)	2:2:2	0:1:11	0.06
Airway maintenance (oxygen: room air)	1:5	2:10	1.00
At discharge
Postmenstrual age, weeks	45.6 ± 4.9	42.0 ± 4.5	0.13
Weight, kg	3.8 ± 0.49	3.6 ± 0.69	0.57
Gavage duration, days (median)	76.3 ± 36.7(79)	41.8 ± 39.8 (43)	0.10
Feeding method (oral: oral/gastrostomy)	3:3	11:1	0.08
Airway maintenance (oxygen: room air)	2:4	1:12	0.13
Length of hospital stay, days (median)	113.5 ± 32.3 (111)	81.3 ± 48.8(106)	0.12

Data shown as mean ± SD or as stated otherwise.

^*median (IQR)

Manometry Findings:

A total of 180 spontaneous swallows were analyzed to compare the basal esophageal motility characteristics between opioid-exposed infants and controls (Table 2, Figure 1). Opioid exposed infants have similar UES resting pressure compared to controls (13.3 ± 9.6 vs. 12.3 ± 9.6 mmHg, P=0.89). Higher basal LES pressures were noted in opioid-exposed infants (40.5 ± 12.7 vs. 21.0 ± 12.8 mmHg, P=0.01) whereas LES nadir pressure showed a higher trend in opioid exposed infants (3.6 ± 2.8 vs. 0.7 ± 6.9, P=0.42). In opioid-exposed infants vs. controls respectively, pharyngeal swallow onset to LES relaxation onset was significantly lower (−0.2 ± 0.9 vs. 1.0 ± 1.0 sec, p=0.04), and LES relaxation onset to LES nadir pressure onset was significantly longer (2.8 ± 0.7 vs. 1.2 ± 0.7 sec, p=0.001) suggesting slow and inadequate LES relaxation. Characteristics of esophageal body peristaltic waveforms including contraction duration, amplitudes, and propagation were similar in both groups (Table 2, Figure 1).

Table 2:

Basal and adaptive esophageal motility characteristics in opioid-exposed infants vs. controls

Characteristics	Opioid Exposed (N=6)	Control (N=12)	P value
Basal esophageal motility characteristics
Proximal esophagus amplitude (mmHg)	36.6 ± 13.5	43.4 ± 13.8	0.33
Mid-esophagus amplitude (mmHg)	57.6 ± 21.8	56.3 ± 21.7	0.91
Distal esophagus amplitude (mmHg)	49.5 ± 19.6	55.7 ± 15.9	0.48
Proximal esophagus contraction duration (sec)	2.7 ± 0.9	3.4 ± 1.0	0.19
Mid-esophagus contraction duration (sec)	3.3 ± 0.7	3.5 ± 0.7	0.67
Distal esophagus contraction duration (sec)	3.5 ± 0.9	3.7 ± 1.0	0.78
Swallow propagation (complete: incomplete: failed), %	73.3:23.3:3.3	81.5:12.6:5.9	0.42
Adaptive esophageal motility characteristics
Pharyngeal reflexive swallow: pharyngo-UES-contractile reflex: no response), %	77.5:5.0:17.5	67.1:11.4:21.4	0.86
Proximal esophagus amplitude (mmHg)	45.2 ± 16.4	43.8 ± 17.3	0.88
Mid-esophagus amplitude (mmHg)	73.9 ±30.3	68.8 ± 33.9	0.75
Distal esophagus amplitude (mmHg)	83.8 ±28.0	62.1 ± 23.2	0.12
Proximal esophagus contraction duration (sec)	5.3 ± 1.5	3.3 ± 1.4	0.01
Mid-esophagus contraction duration (sec)	5.3± 1.7	3.4 ± 1.7	0.04
Distal esophagus contraction duration (sec)	6.5± 2.2	3.9 ± 1.7	0.03
Distal esophagus area under the curve, mmHg.sec	258.1± 113.4	141.2 ± 95.5	0.04
Swallow propagation (complete: incomplete: failed, %)	81:9.4:9.4	91:05:04	0.16

Data shown as mean ± SD or as stated otherwise. UES=Upper Esophageal Sphincter

Figure 1:

Comparing the characteristics of basal esophageal motility between controls vs. opioid- exposed infants: Effects on skeletal muscle (UES and proximal esophagus), mixed muscle (mid- esophagus) and smooth muscle (distal esophagus and LES) parts of the esophagus are explained. Note the high LES resting pressure in the opioid exposed infants.

A. Upper Esophageal Sphincter (UES) Characteristics. B. Esophageal Contraction Duration–Proximal-, Middle-, and Distal-Esophagus. C. Lower Esophageal Sphincter (LES) Characteristics.

Data from a total of 113 responses to incremental pharyngeal stimuli were analyzed to compare the characteristics of adaptive esophageal motility between the two groups (Table 2, Figure 2). In both groups, pharyngeal reflexive swallow was the primary response, and administering volume in increments of 0.1 ml, 0.3 ml and 0.5 ml lead to an increase in the prevalence of pharyngeal reflexive swallow and a decrease in the prevalence of pharyngo-UES- contractile reflex. As was also seen with the basal swallows, pharyngeal infusions in opioid exposed infants showed similar UES resting pressure, trend for higher LES resting pressures and significantly higher LES nadir pressure (Figure 2). Pharyngeal stimulus in the order of 0.1 to 0.5 ml (1–5 drops of sterile water) has resulted in prolonged esophageal body contraction durations in opioid-exposed infants at proximal-, middle-, and distal- esophagus, compared to controls (Table 2). In addition, the contractile area under the curve (mmHg.sec) at the distal esophageal waveform was greater in the opioid-exposed infants. Within the esophageal body, amplitudes of peristaltic waveforms and propagation characteristics were similar in both groups (Table 2).

Figure 2:

Comparing the characteristics of adaptive esophageal motility between controls vs. opioid- exposed infants: Effects on skeletal muscle (UES and proximal esophagus), mixed muscle (mid- esophagus) and smooth muscle (distal esophagus and LES) parts of the esophagus are explained. Note the prolonged contractile duration at all esophageal loci and positive LES nadir pressure.

Upper Esophageal Sphincter (UES) Characteristics. B. Esophageal Contraction Duration–Proximal-, Middle-, and Distal-Esophagus. C. Lower Esophageal Sphincter (LES) Characteristics.

DISCUSSION

In this pilot study, we provide clinical reasoning and proof of concept that sensory-motor mechanisms of pharyngeal provocation induced swallowing and esophageal motility reflexes in infants with prenatal exposure to opioids are impaired and may form the basis for their persistent feeding difficulties. Using an experimental design and provocative interrogation of sphincteric tone and esophageal motility, we noted that, esophageal body coordination and clearance, and LES characteristics were aberrant in predominantly opioid exposed infants.

Fetal exposure to addictive drugs is increasing; neonates present with a constellation of symptoms of withdrawal involving the gastrointestinal, central and autonomic nervous systems which are indeed the basis of the Finnegan scoring system.⁵ Excessive sucking, poor feeding, regurgitation and emesis combined with respiratory symptoms, increased muscle tone, arching and irritability (all of these are components of Finnegan’s scoring) can often be mistaken for GER with the consequences of being prescribed with acid-suppressive therapies.¹⁸ Since both groups in our study had similar proportion of infants requiring oxygen (a surrogate for chronic lung disease) at study; this would indicate that lung disease alone did not play a major role in persistent feeding difficulties. While prematurity is in itself a risk factor for protracted feeding difficulties, substance abuse is an important confounder modifying enteric-neural, autonomic-neural and central-neural circuit interactions.

In this exploratory study, the cardinal findings revealed are: a) infants with fetal exposure to opioids have up- regulation of central vagal effects with heightened cholinergic tone as evidenced by heightened LES basal tone and increased effort and prolonged clearance time at distal esophagus; b) persistence of feeding and swallowing abnormalities suggest modification of sensory-neural circuitry and interactive effects even after the cessation of therapeutic agents; c) the application of novel experimental design to examine the vagally-mediated aero-digestive-neural interactions in vulnerable neonates provides the basis for with unexplained symptoms.

Specifically, the following significant mechanisms among the opioid exposed infants were noted: a) During spontaneous swallowing, the higher resting LES pressures suggests a balance favoring cholinergic excitatory pathway activity rather than decreased LES inhibition via the nitrergic pathway (via inhibitory neurotransmitter nitric oxide); b) With pharyngeal provocation swallowing, the increased area under the curve at distal esophagus is suggestive of prolonged distal clearance amidst increase in LES nadir pressure c) Longer durations of esophageal body waveforms in the proximal, distal and mid- esophagus are suggestive of a longer esophageal contractile duration; d) Nadir pressure at LES remained higher, again tilting a balance towards increased cholinergic excitatory pathway activity and decreased LES inhibition.

While spontaneous swallowing is mediated by central effects (supra-nuclear pathways),¹⁹ the adaptive swallowing evoked upon pharyngeal provocation is mediated by brain stem nuclei (glossopharyngeal and vagus).^15,16,20,21 With fetal exposure to opioids, differences were apparent in both spontaneous and adaptive swallowing responses. Although exposures to opioids were 4–9 weeks prior to manometry studies, and 2–4 weeks after treatment for NAS, altered parasympathetic tone persisted for a longer period of time. In adult studies, esophageal motor dysfunction including impaired LES relaxation and distal esophageal spasm was found to be common in chronic opioid users whether studied on opioids or off opioids.^22,23 In the human gut, peripheral opioid receptors are mainly localized in the submucosal and myenteric plexus neurons.²⁰ Opioids are reported to evoke tonic spasm of the smooth muscle by inhibiting release of nitric oxide from inhibitory enteric neurons, thus leading to unopposed excitatory input that can result in spastic esophageal motor activity.^23,24 Symptoms and signs in opioid exposed infants such as arching, irritability and frequent gulping may mimic GER; however, it is likely that these may be due to inadequate esophageal motility (heightened contractile activity at distal esophagus) and heightened LES tone and inadequate LES relaxation. Hence these infants have repetitive swallowing which permits LES relaxation, although incompletely.

Limitations of this pilot study are addressed below: 1) All the infants studied were preterm infants. NAS is much more common in term infants than preterm infants. Due to the longer duration of exposure to opioids in utero, term infants with NAS are likely to have more robust opioid induced physiological gastrointestinal findings. However, term infants with NAS are difficult to recruit for invasive physiological testing as minimal stimulation is the main strategy to manage withdrawal symptoms, and the physiological testing may not be considered to add any clinical benefit. 2) The opioid exposed group trended toward longer gavage feeding days, although this did not reach significance; this small study was underpowered to assess this metric. However, the chronic feeding difficulties in the opioid exposed group may be related to impaired adaptive mechanisms. 3) Sensory-motor difficulties related to feeding in opioid exposed infants may be contributing to longer hospitalization. In general, opioid exposed infants stayed 4 weeks longer than the controls. Although this metric is not statistically significant, it can contribute to significant economic and social problems. 4) Using a novel investigative approach, we studied a small population of opioid exposed infants with persistent feeding difficulties, and did not necessarily identify all those with variable multiple drug abuse periods during pregnancy. We investigated for potential mechanisms for feeding difficulties in this cohort rather than the addictive-dose or poly-drug abuse response relationships. Due to the exploratory nature of our study, findings are only associations and not considered as causation. Application of such novel study designs is needed in a carefully selected larger sample of perinatal-opioid exposed infants to clarify the mechanisms of malfunction or maladaptation of aerodigestive reflexes. 5) Concomitant exposure to other addictive substances and psychotropic medications is common in opioid exposed infants, and may increase the risk for neonatal drug withdrawal²⁵; however, the short- and long-term consequences of this exposure are not entirely clear. 6) Increasing prevalence of fetal drug exposures call for increased awareness and public health measures to reduce these antenatal exposures. Current treatment strategies for mothers or babies exposed to opioids do not fully avoid the neuronal consequences for the neonate.

SUMMARY AND CONCLUSIONS

Preterm-born infants in general, are at increased risk for airway and digestive difficulties; however, convalescing infants with fetal opioid exposure are at greater risk for longer hospital stay, oral feeding difficulties, and supplemental oxygen needs. The current study is the first one to investigate and report the sensory-motor mechanisms with pharyngeal provocation induced swallowing reflexes in premature fetuses exposed to opioids when studied ex utero for the presence of persistent feeding difficulties. Infants with fetal exposure to opioids have up-regulation of central vagal effects with heightened cholinergic excitatory responses; these neuromotor findings may form the basis for feeding difficulties in infants exposed to opioids in utero. Our data highlight for the first time, the lasting effects of fetal exposure to addictive agents on specific sensory-motor esophageal dysmotility mechanisms that may have a bearing on post-exposure consequences. Further longitudinal studies at defined maturational time points are needed to clarify the causal and restorative mechanisms so that feeding rehabilitation programs can be optimized to enhance functional capabilities.

Abbreviations:

NAS

neonatal abstinence syndrome

UES

upper esophageal sphincter

LES

lower esophageal sphincter

LOS

length of stay

GER

gastroesophageal reflux