Infants with long neonatal intensive care hospitalizations may lose opportunities for learning to eat and may associate eating with pain or discomfort (1). Although the causes for prolonged hospitalizations often resolve with treatment, a tube-feeding requirement may persist for months to even years. As the use of gastrostomy tube feeding has risen, so have the challenges associated with weaning a child from tube to oral feeding.
There are several methods for achieving transition from tube to oral feeding, but no optimal solution. Many programs consist of intensive outpatient or even 2- to 3-week long inpatient stays (2–5). Such intensive programs can cause family disruption, distress, and financial burden. Other studies include several case reports of behavioral interventions with only moderate success (6–8).
Few studies (9) examining the effectiveness of outpatient feeding programs have been published, likely because of the relative dearth of such programs. One theory has suggested that some toddlers refuse to eat because they have chronic upper chest or abdominal discomfort, which is worsened by eating. This results from pain nerve sensitization with hyperalgesia and allodynia (1). Because many medically fragile toddlers have experienced gastrointestinal pain, approaching the tube to oral feeding transition from a pain management perspective may be advantageous. The present study used such an approach (1,10) to transition nonverbal children from tube to oral feeding. The objective of the present study is to report a multidisciplinary outpatient method for moving nonverbal children from tube feeding to exclusively oral feeding.
PATIENTS AND METHODS
We conducted a chart review for all nonverbal toddlers who underwent an attempt to move from tube to oral feedings by the University of Kansas Medical Center Multidisciplinary Feeding team from 2004 to 2007, resulting in identification of 9 subjects. Their mean age was 27.3 months (standard deviation [SD] 15.6 months), and ages ranged from 7 months to 52 months at the initial visit. Upon starting this protocol, all 9 subjects received 100% of their energy intake through a gastrostomy tube. The medical or developmental diagnoses of the toddlers are shown in Table 1.
Before initiating the protocol, 5 of 9 participants had undergone esophageal and antroduodenal manometry with normal results. Four sets of parents refused manometry, preferring a trial-and-error approach.
As recommended by many experts in the field (11), our feeding team comprised a multidisciplinary group, including a pediatric gastroenterologist, a pediatric psychologist, an occupational therapist, and a dietitian. The feeding team met as a group with each patient and caregiver(s) to complete a thorough assessment of the child's oral motor skills, behavioral strengths and weaknesses, nutritional status and growth pattern, and gastrointestinal symptoms, as well as to observe a typical meal. At the end of the initial session, recommendations were made to caregivers related to oral motor skills, behavior modification, nutrition, and medication for implementation at home.
Before beginning the medication, the team performed a thorough assessment to ensure that each child possessed the oral motor skills necessary for eating (ie, chewing, lateralized tongue movement; see Table 2) (12). Skills necessary for eating included oral skill development at or near the child's developmental age. These skills supported the ability of the mouth to coordinate and control food that is nutritionally appropriate and adequate for growth (13). If oral skills were below what would be expected, then the child may not have had the strength to support adequate eating. If the move to all-oral feedings was accomplished before the age of 2 years and there were no other concerns related to motor or sensory development, then there is a reasonable expectation in the child's ability to develop a more mature eating pattern. Head, neck, and trunk support were assessed because they are necessary not only to support the child's ability to control food in the mouth but also for control when swallowing and for respiratory stability (14). Finally, all sensory processing issues were evaluated and managed or treated before beginning the protocol, because untreated sensory processing issues may impair the ability to eat.
The team also provided guidance related to eating behaviors to the child and family and made certain that the toddler and family had the necessary mealtime behaviors (Table 3). Behavioral skills were assessed through interview and observation of a meal in the clinic. Behaviors necessary for mealtime included coming to the table willingly several times per day, remaining seated throughout the meal, consuming some food by mouth willingly, and maintaining parental control over timing of meal's end. If behavioral skills were not sufficient, then families were trained to meet minimal criteria before initiating the medication.
After the oral motor and behavioral conditions were met, the protocol moved to amitriptyline (1 mg · kg−1 · day−1) or gabapentin (10 mg/kg thrice daily), as well as continuous drip gastrojejunal tube feedings (20 hours/day). We began with postpyloric, continuous drip feedings and one drug. If the patient had a history of poor sleep or diarrhea, then we started with amitriptyline 0.25 to 0.3 mg · kg−1 · dose−1 at bedtime because amitriptyline not only relieves chronic pain but it also reduces stool water through its anticholinergic effects, and improves sleep through antihistaminic effects (15–17). We increased the amitriptyline dose by 0.25 to 0.3 mg · kg−1 · day−1 each week to attain a maximal dose of 1 mg · kg−1 · day−1. If the child had a history of seizures or heart disease with conduction abnormalities we avoided amitriptyline, but began with gabapentin 10 mg/kg twice daily, because gabapentin not only reduces chronic pain but it also raises the seizure threshold and does not provoke cardiac arrhythmias as overdoses with amitriptyline (18). After 1 week we increased the dose to 10 mg/kg 3 times daily. Patients returned for their second visit 8 weeks later. During this visit, the team answered questions, provided support, offered guidance to the child's parents/caregivers, and we prescribed appetite stimulant megestrol (3 mg/kg twice daily). We continued the pain medications amitriptyline or gabapentin. Five days after beginning the appetite stimulant, tube feedings were progressively reduced. This process lasted between 10 days and 2 weeks until tube feeding was discontinued. The appetite stimulant was discontinued after 6 weeks to avoid potentially serious side effects of longer treatment with this medication. Weight was monitored weekly at home and reported to the team. Families were encouraged to contact team members with questions or concerns. The other medications were stopped after 6 months, by weekly titration to smaller doses.
All of the subjects were able to complete the transition protocol without negative side effects. At the end of this 14-week protocol, all 9 patients received 100% of their energy intake orally. Eight of the 9 were able to maintain their weight over time and never required additional tube feeding. One patient with cerebral palsy had difficulty maintaining his weight due to oral motor stamina issues. At the time of writing, his intake was 50% oral and 50% gastrostomy tube. Therefore, all of the subjects had improvement in oral feeding during the 14-week outpatient program. Eighty-nine percent of subjects maintained exclusively oral feedings following the 14-week protocol.
This study described an outpatient transition program based on a pain rehabilitation approach. There are several methods used for moving children from tube to oral feeding that focus on behavioral, oral-motor, and medical issues (19). However, many involve extended inpatient stays or intensive day treatment. Other published outpatient studies have reported limited success, even when taking a multidisciplinary approach including medication and cognitive behavioral family therapy (1). Several case reports using behavioral interventions achieved only moderate success (6,7).
A recent case study used a method similar to the present study, but instead of using medication to treat physical pain, they conceptualized the oral feeding difficulties as emotional and psychological in nature. The case study reported the successful use of fluoxetine to treat 2-year-old twins who became distressed and agitated during oral feeding following several invasive gastrointestinal procedures (20). Researchers noted that their fear and anxiety decreased with treatment, which allowed the twins to discontinue tube feeding within several months.
To our knowledge, the present study is the first published protocol to describe an outpatient program for transitioning medically fragile toddlers from gastrostomy tube to oral feeding that uses a pain rehabilitation model. Although we did not specifically assess physical pain during the transition, we hypothesize that, at least for some children, a pain management approach may be effective for moving children from tube to oral feedings. In addition, these changes were accomplished in an outpatient setting, preventing children and their families from having to endure stressful, expensive, and unnecessary inpatient stays.
In the case of our subjects, we hypothesized that gastric distension from a routine meal caused discomfort, potentially stimulated retching against a fundoplication in an attempt to avoid the discomfort, and therefore maintained or further sensitized pain nerves. Continuous drip postpyloric feedings treated pain by imposing an 8-week period of no gastric distension and so no retching, thus allowing pain nerves to return to normal sensitivities. This hypothetical return to normal nerve sensitivity was presumably accelerated with drugs used to treat chronic neuropathic pain in children and adults. Amitriptyline and other tricyclic antidepressants have been used for the past 20 years to treat irritable bowel syndrome (15,16). The amitriptyline doses used for pain treatment are a magnitude less than the doses used for treating depression and pose no cardiac risk in individuals without preexisting cardiac conduction defects. The antiseizure medication gabapentin has been used for chronic neuropathic pain (21) and quiets irritability in children with neurological impairments (22).
Megestrol has the potential to increase negative behaviors including insomnia, hyperactivity, and irritability (23). However, patients in the present study experienced no problems related to megestrol.
There are several limitations to the present study. First, the small sample makes it difficult to generalize about the effectiveness of this approach. Second, this is a retrospective chart review rather than a randomized controlled trial, limiting the scientific rigor of the findings. Finally, the program was not fully successful in one case of a child with oral motor difficulties limiting his stamina and preventing him from maintaining his weight with exclusively oral feeding.
This study is the first to examine the effectiveness of a 14-week outpatient feeding protocol based on a pain rehabilitation model for the transition from tube to oral feeding in toddlers. The protocol was successful in 89% of the sample in maintaining exclusively oral intake. A randomized controlled trial comparing this pain rehabilitation protocol to standard care group with a larger sample should be conducted to better assess the value of the protocol.
The authors acknowledge the support of the National Institutes of Health (NIDDK K23068221 to A.M.D.), and thank the families for their strong contributions.
1. Zangen T, Ciarla C, Zangen S, et al
. Gastrointestinal motility and sensory abnormalities may contribute to food refusal in medically fragile toddlers. J Pediatr Gastroenterol Nutr 2003; 37:287–293.
2. Blackman JA, Nelson CL. Rapid introduction of oral feedings to tube-fed patients. J Dev Behav Pediatr 1987; 8:63–67.
3. Byars KC, Burklow KA, Ferguson K, et al
. A multicomponent behavioral program for oral aversion in children dependent on gastronomy feedings. J Pediatr Gastroenterol Nutr 2003; 37:473–480.
4. Clawson EP, Kuchinski KS, Bach R. Use of behavioral interventions and parent education to address feeding difficulties in young children with spastic diplegic cerebral palsy. NeuroRehabilitation 2007; 22:397–406.
5. Senez C, Guys JM, Mancini J, et al
. Weaning children from tube to oral feeding
. Childs Nervous Syst 1996; 12:590–594.
6. Dunbar SB, Jarvis AH, Breyer M. The transition from nonoral to oral feeding
in children. Am J Occupat Ther 1991; 45:402–408.
7. Gutentag S, Hammer D. Shaping oral feeding
in a gastronomy tube-dependent child in natural settings. Behav Modif 2000; 24:395–410.
8. Luiselli JK. Cueing, demand fading, and positive reinforcement to establish self-feeding and oral consumption in a child with chronic food refusal. Behav Modif 2000; 24:348–358.
9. Benoit D, Wang EEL, Zlotkin SH. Discontinuation of enterostomy tube feeding
by behavioral treatment in early childhood: a randomized control trial. J Pediatr 2000; 137:498–503.
10. Chitkara DK, Lorenzo CD. Pharmacotherapy for functional gastrointestinal disorders in children. Curr Opin Pharm 2006; 6:536–540.
11. Ayoob KT, Barresi I. Feeding disorders in children: taking an interdisciplinary approach. Pediatr Ann 2007; 36:478–483.
12. McIntosh DN, Miller LJ, Shyu V, et al. Overview of the short sensory profile (SSP). In: Dunn W, ed. The Sensory Profile: Examiner's Manual
. San Antonio, TX: Psychological Corporation; 1999. pp. 59–73.
13. Delaney AL, Arvedson JC. Development of swallowing and feeding: prenatal through first year of life. Dev Dis Res Rev 2008; 14:105–117.
14. Morris SE, Klein MD. Pre-Feeding Skills. 2nd ed San Antonio, TX: Therapy Skill Builders; 2000.
15. Crowell MD, Jones MP, Harris LA, et al
. Antidepressants in the treatment of irritable bowel syndrome and visceral pain syndromes. Curr Opin Invest Drugs 2004; 5:736–742.
16. Bahar RJ, Collins BS, Steinmetz B, et al
. Double-blind, placebo-controlled trial of amitriptyline for the treatment of irritable bowel syndrome in adolescents. J Pediatr 2008; 152:685–689.
17. Mayers AG, Baldwin DS. Antidepressants and their effect on sleep. Hum Psychopharmacol 2005; 20:533–559.
18. Thanacoody HK, Thomas SH. Tricyclic antidepressant poisoning: cardiovascular toxicity. Toxicol Rev 2005; 24:205–214.
19. Kindermann A, Kneepkens CMF, Stok A, et al
. Discontinuation of tube feeding
in young children by hunger provocation. J Pediatr Gastroenterol Nutr 2008; 47:87–91.
20. Celik G, Diler RS, Tahiroglu AY, et al
. Fluoxetine in post-traumatic eating disorder in two-year-old twins. J Child Adol Psychopharm 2007; 17:233–236.
21. Dworkin RH, O'Connor AB, Backonja M, et al
. Pharmacologic management of neuropathic pain: evidence based recommendations. Pain 2007; 132:237–251.
22. Hauer JM, Wical BS, Charnas L. Gabapentin successfully manages chronic unexplained irritability in children with severe neurologic impairment. Pediatrics 2007; 119:e519–e522.
23. Marchand V, Baker SS, Stark TJ, et al
. Randomized, double-blind, placebo-controlled trial of megestrol acetate in malnourished children with cystic fibrosis. J Pediatr Gastroenterol Nutr 2000; 31:264–269.