Treatment with Gabapentin Associated with Resolution of Apnea in Two Infants with Neurologic Impairment

Introduction

Cardiopulmonary events, including apnea, are common problems in premature infants and in infants with neurologic impairment. Apnea may include apnea of prematurity, obstructive apnea, or a mix of obstructive and central apnea. We report two cases of infants with improvement in apnea after treatment with gabapentin for presumed central neuropathic pain and/or visceral hyperalgesia. We hypothesize that pain resulted in alterations of airway tone and respiratory muscle coordination. Infants are at greatest risk given their smaller airways, lower functional residual capacity, and the high rate of unrecognized pain at this age.

Case 1

A twin born at 31 weeks' gestation, now with a corrected gestational age of 2 months, suffered a devastating central nervous system (CNS) injury as a result of a twin–twin transfusion. Her head magnetic resonance imaging (MRI) was significant for marked cystic encephalomalacia and cerebral loss. Prior to discharge, a 12-hour event recording demonstrated that heart rate, respiratory pattern, and oxygen saturation were within normal limits. She was being fed thickened liquids, based on the results of a swallow study, and was on a proton pump inhibitor (PPI).

Since discharge, parents had observed changes in her respiratory pattern during sleep, including “prolonged pauses until a deep breath is taken.” In addition, episodes of irritability developed, with the abnormal breathing pattern noted when irritable. When comfortable she is described as “calm” when she sleeps. A sleep study identified severe obstructive apnea with 27 obstructive apneas, longest duration 22.8 seconds; 22 obstructive hypopneas, longest duration 69.4 seconds; and a lowest desaturation of 46%. She was assessed to either “need a tracheostomy” or that “parents should agree to do not resuscitate/do not intubate.” A lengthy discussion with the palliative care team allowed parents to identify the following goals: improved comfort, improved sleep, and a “meaningful life.” They were worried that a tracheostomy would prolong her life without providing benefit to a meaningful life. They also hoped that she would do better than anticipated.

Subsequent evaluation included normal upper gastrointestinal series, normal airway fluoroscopy, and swallow study negative for aspiration. An assessment by a pediatric otolaryngologist identified abnormal airway tone, with the palate and base of the tongue resulting in obstruction. The physical examination included minimal spasticity at baseline. Consideration was given to pain as a trigger for episodic hypertonia. Evaluation by pediatric neurology found no evidence for seizure activity and agreed with pain as a potential trigger. Features consistent with pain included moaning, inability to console, stiffening, and facial grimacing. Symptoms could occur “out of the blue” and were especially pronounced with intestinal gas and prior to bowel movements. An empiric gabapentin trial was initiated as treatment for possible central neuropathic pain, given the severity of her cerebral injury. Two weeks later, at a dose of 20 mg/kg per day divided into three doses, improvement in irritability and sleep were noted. Further increases resulted in ongoing improvement in comfort and resolution of observed pauses in breathing at night. She continued to do well 9 months later and was described as very happy and social.

Case 2

Patient 2 is a 5-month-old former 25-week premature infant with severe chronic lung disease and significant CNS injury, including periventricular leukomalacia. The palliative care team was consulted to assist the family with decision making. Significant problems included episodes of apnea with oxygen desaturation and bradycardia, which required stimulation and supplemental oxygen with positive pressure, multiple times each day. Multiple failed attempts to wean her off of ventilatory support led to tracheostomy and gastrostomy tube placement. Other interventions included a PPI since 6 weeks of age. Despite these interventions, the episodes persisted.

Upon review, episodes were typically preceded by irritability, restlessness, arching, increased tone including extension of legs, and grimacing with facial flushing. Some episodes were noted at a time of sleep, with a sudden onset of features followed by desaturation and bradycardia and resolution after passing of intestinal gas. Clinically, she appeared to have a “rigid chest” during the episodes. Evaluation by pediatric neurology included a normal EEG at 3 months of age.

Nurse and physician staffing were frustrated with the lack of progress and generally agreed the patient was suffering and at risk for sudden death if sent home. Given her severe episodes, the medical team thought it appropriate to discuss a do-not-resuscitate (DNR) order with the child's guardians. The PC team requested a cardiac echo, which demonstrated “flattening of the intraventricular septum…[and a] peak TR gradient of 60 mm Hg” consistent with significant pulmonary hypertension. This was considered by the neonatal intensive care unit (NICU) and palliative care teams to increase the risk for sudden death. The child's guardians requested a DNR order after a 2-week trial of sildenafil failed to improve estimated pulmonary pressures by echo. In addition, there was no improvement following treatment with metoclopramide and limited benefit from morphine sulfate.

The palliative care team recommended a gabapentin trial, given that the episodic irritability and hypertonia prior to apneic episodes as well as the association with intestinal gas suggested central neuropathic pain and/or visceral hyperalgesia as triggers for these events. Starting with a nighttime dose, the number of apnea and bradycardia events decreased, with the benefit most notable at night. Further improvement, with a decrease from several times an hour to several times a day, was noted following a dose increase. Metoclopromide was discontinued without any regression of clinical progress or return of apnea. The gabapentin dose was increased to 22 mg/kg per day divided into three doses, without any notable side effects. Irritability decreased significantly along with the decrease in severe cardiorespiratory episodes. Staff and family alike described her as “a totally different child” including no further apneic episodes, more alert during the day, and sleeping better at night. A follow up echocardiogram several months later, following discharge from the hospital, showed “less TR…[and] normal RV free wall and septal motion.”

Discussion

Apnea can result in life-threatening events in infants, as was a concern in both cases. Causes of apnea considered included seizures, gastroesophageal reflux disease (GERD), hypertonia, and pain. Neither infant was identified to have seizures as an explanation for the observed events. Though GERD is frequently considered a cause of apnea in infants, as treated in both cases, literature questions this common assumption. Recent studies indicate a lack of correlation between apneic episodes and episodes of GERD or after feeding.^1–3 Although the gastric pH, frequency and height of GERD episodes was more significant following feeding; apnea, bradycardia, and oxygen desaturation was not more prevalent following feeding. ¹ When apneic events and GERD episodes were reviewed in 71 preterm infants, GERD episodes were rarely correlated with episodes of apnea, bradycardia, or oxygen desaturation. ²

Pain and hypertonia were also considered as triggers for apneic events, by altering airway tone and chest wall movement. Pain was suggested in both cases by nonspecific irritability. Other features that indicate pain in nonverbal children with neurologic impairment (NI) and were present in both include: moaning, facial grimacing, difficult to console, withdrawn, appearing restless, increased tone, stiffening, and arching.^4–7 Pain is a frequent problem in children with NI, occurring daily in 40% of those with the most severe impairment.^4,8 In addition, it typically presents in the first year of life in such children.^8,9 Pain is also a known trigger for hypertonia and muscle spasms. Other potential contributing factors associated with pain that can alter airway position and chest wall movement include stiffening, arching and posturing.

Spasticity (involuntary, velocity-dependent increase in muscle tone that results in muscles that are resistant to movement) at baseline was minimal in both cases. Pain was considered a potential trigger for intermittent episodes of hypertonia/muscle spasms followed by apneic episodes. In the first case, obstructive apnea could be due to increased airway or pharyngeal tone and neck positioning during the pain (and hypertonic) episodes. In the second case, acute “restrictive” lung physiology was thought to result in apnea, with pain triggering increased tone in the abdomen and respiratory musculature in the chest wall. Infants would be particularly susceptible given their smaller airway and lower functional residual capacity, as would infants with underlying lung disease. Spasticity contributing to obstructive apnea and impaired respiratory muscle coordination is suggested by a case report of a young woman with cerebral palsy in which obstructive apnea resolved following use of an intrathecal baclofen pump. ¹⁰

Pain sources in children with NI include nociceptive, peripheral neuropathic pain, central neuropathic pain, and visceral hyperalgesia. ¹¹ Central neuropathic pain is best described in adults with insults to the CNS, such as following a cerebral vascular accident or from multiple sclerosis.^12–14 Symptoms can include generalized pain along with pain that can occur in sudden bursts and include visceral pain. This can include painful fullness from normal distention of the bladder and intestinal tract, described by one adult as “like my bowels will explode.” ¹⁵ The symptoms associated with the intestinal tract may be relevant to children with NI as is described further with visceral hyperalgesia.

Visceral hyperalgesia is a lowered activation threshold to pain generation in response to a stimulus, such as distention within the intestines. ¹⁶ Visceral hyperalgesia was identified as the source of symptoms in 12 of 14 medically fragile children with persistent symptoms despite treatment of GERD with medications and Nissen fundoplication. ¹⁷ Gastrointestinal injury and inflammation may contribute to sensitization of visceral afferents, such as GERD, gastrostomy tube placement and antireflux surgery.

These sources of pain are important to consider given that the gut is identified as one of the most common sources of pain in children with NI despite treatment of problems such as GERD and constipation.^4,18–20 These two cases included symptoms that suggest the gastrointestinal symptoms associated with central pain and visceral hyperalgesia, including pain associated with intestinal gas and relief of symptoms after passing of stool and gas.

Each patient was treated with a gabapentin trial given its demonstrated benefit for central pain and visceral hyperalgesia.^{12–14,16,17,21,22} Gabapentin has been studied to a lesser degree as treatment for spasticity, studied best in adults with multiple sclerosis, including one study identifying a significant reduction in the frequency of muscle spasms and degree of pain. ²³ In addition, gabapentin has a good safety profile in children, including no known drug-drug interactions, making it a reasonable first choice for an empiric trial. As an example of benefit, nine children with NI treated with gabapentin experienced a significant reduction in pain symptoms as well as improvement in feeding intolerance and sleep. ²⁴ In another case report, an infant with a chromosomal abnormality, associated NI, and hypotonia, was noted to have a significantly lower pain score following treatment with gabapentin, after failed benefit from morphine sulfate, lorazepam, and phenobarbital. ²⁵

In both cases, the initial gabapentin dose was 2.5 mg/kg starting with one dose in the evening, with dose titration occurring every 4–7 days initially, with the option to titrate more rapidly once tolerated. This lower initial dose and slower titration reflected caution, given the degree of apnea. Further information about pain management in children with NI can be found elsewhere. ¹¹

This report is limited to the observed improvement of these two infants. The hypothesis that children with NI and apnea benefit from treatment with gabapentin will need to be addressed in a prospective study. Such a study would benefit from a standardized medical evaluation, including assessment of potential pain sources for each patient, monitoring of pain frequency and severity, documentation of spasticity and frequency of spasms, and a standardized assessment of pain and tone post treatment.

In conclusion, infants with NI as a result of an insult to the CNS are also at increased increase for pain. Problems associated with pain in children with NI include increased tone. This secondary hypertonia may also contribute to apnea as a result of alterations in airway tone and chest wall restriction. These cases illustrate that treatment of pain can lead to resolution of apnea. Infants with NI and apnea should have a careful pain assessment when considering other causes and interventions for apnea. Gabapentin is suggested as a first-line trial, given its potential benefit for central neuropathic pain and hypertonia as a result of impairment of the CNS.