Functional Dysphagia (Food Phobia, Fear of Swallowing) in Children

Food Phobia of Childhood:  a “new” eating disorder

Introduction: Food phobia of  childhood, primarily seen in pre- or early pubertal children, was first described as such by Bryan Lask (pediatric psychiatrist) and Rachel Bryant-Waugh (psychologist) in the early 1990’s as a result of their work at Great Ormond Street Children’s Hospital in London. To my knowledge, with the exception of one article reporting work with a single young boy, food phobia has not been discussed as an entity in the American pediatric literature except under the general title “dysphagia” where it is likely to come to the attention primarily of pediatric gastroenterologists and otolaryngologists.  In the adult literature it is usually referred to as “choking phobia”. More recently Dr. Lask has chosen to refer to it again as “functional dysphagia,” although we at the Kartini Clinic for Disordered Eating prefer the more intuitive “food phobia”. In our experience, pediatricians report they are often at a loss about what to do with these challenging patients. We are grateful to Drs. Lask and Bryant-Waugh for calling attention to this condition.

Food phobia is the sudden onset of food refusal and may extend to refusal to swallow anything, even the patient’s own saliva.  These young patients easily become undernourished and dehydrated.  The “reason” these children give for their food refusal is often a fear of choking when they swallow or a fear of vomiting and/or germ contamination leading to illness; they may report food and water getting “stuck” in their throats or they may simply refuse to talk about why they won’t eat or drink.  In our experience, onset is usually abrupt; the child is young; there is a family history of anxiety disorders, and the child does not have anorexia or bulimia nervosa with the body dysmorphism and “fat fears” associated with those diagnoses. The onset of illness may be preceded by a choking episode, either experienced or witnessed, or in the case of Patient # 3 (see below), overheard.  This disorder may be preceded by an illness that involved vomiting or severe nausea.  Efforts on the part of the parents to bribe, cajole, threaten, plead, reward eating with “gold stars”, etc. are all to no avail.

As you will see from the case histories below, our treatment team at Kartini Clinic managed these patients on the general pediatric ward with some combination of 1) an NG tube and weight restoration; 2) the use of a neuroleptic (typically Olanzepine titrated up to about 7.5 mg q h.s., valued for its impact on cognitive distortions as well as its anxiolytic properties); 3) the eventual use of an SSRI to achieve more long-term anxiety control; and 4) behavioral incentives introduced towards the end of the stay.  The timing of the behavioral intervention is crucial.  If it is initiated too soon, before the child has learned to trust the team, it may be experienced as punitive and accompanied by increased anxiety. Our treatment team is multi-disciplinary and this seems to be an essential feature of its success as we need to contain parental anxiety, assess for any co-morbid psychiatric or medical problems, and plan for re-integration into home and school. During an in-patient stay our pediatricians lead the team and work with a child psychiatrist or psychiatric nurse practitioner and a family therapist. In our Day Treatment unit and outpatient clinic a child is cared for by milieu therapists, group therapists, pediatricians, family therapists and psychiatric providers.  Modeling ordered eating in a day treatment unit prior to going to an outpatient clinic has proven an essential ingredient to the re-integration of food phobic children to home and school.

A discussion of the treatment and results follows the case histories presented below. Of note is the fact that Patient #3 was treated with Risperidone (Risperdal) rather than Olanzepine (Zyprexa), something we no longer do because of concerns about galactorrhea (milk production).  Because of her underlying diagnosis of autism we were unable to use N.G. feeds, a practice we usually find indispensable.

Patient #1:

“Rex” was a 10 2/12 year old boy referred by his local pediatrician for a six pound weight loss (from 56 lbs to 50 lbs) and abrupt onset of refusal to eat enough to sustain himself as well as increasing restriction in the type of food he would eat.

Rex had been born at full term weighing 8lbs 4 oz to a 33-year old primigravida with no problems during pregnancy, labor or delivery except for maternal allergic reaction to an antibiotic during labor.  He was breastfed for 10 months and was described as easy to feed as a very young child, although he became a picky eater with age.  He suffered from some shyness and social anxiety, but his development was generally normal. He did well in school, had never seen a therapist and had never taken psychotropic medications.  Rex had not yet entered puberty and had some residual childhood fears about sleeping alone.  He attended the fifth grade and loved reading, Gameboy, Pokemon and especially soccer.  His parents described him as compassionate, creative and conscientious.

Rex lived with his biological parents, one brother and a sister. Mother was a stay-at-home Mom and his Dad was an engineer, they had been married for 12 years, were Roman Catholics and had no particular family issues bothering either them or Rex, although there was some difference in parenting styles with father being somewhat less patient.  Family history was positive for perfectionism in Rex’s father, a history of depression and a “sensitive stomach” in other family members; Rex’s brother had a throat-clearing tic, ADHD and asthma. Several family members were described as “high energy” and the paternal grandmother had a history of panic attacks, depression and anxiety.  The maternal grandmother had hypertension, glaucoma and breast cancer.

In July, one month before Rex’s presentation to us, he began drastically to decrease the amount of food he was willing to eat.  Mother recalled that Rex had been sick that January or February with “the flu” and vomiting, and sick again in March.  Some increased pickiness began around that time, but his refusal to eat had been recent.  Rex even refused foods he had recently accepted, such as milkshakes, and became very upset when offered more, stating emphatically and categorically, “I can’t.”  He talked incessantly about his fear of “becoming too full and vomiting”. He was taken to see his pediatrician who checked the usual blood work (CBC, UA, metabolic panel) which was normal; ova and parasite studies were normal; because of the weight loss and the increasingly firm refusal to eat enough to sustain himself, he was referred to Kartini Clinic.

On physical exam Rex was a tiny, emaciated boy, with an anxious look in his eyes who nonetheless answered questions willingly; he was 50 inches tall (127cm) and weighed 50.6 lbs (23.0 kg).  His respiratory rate was 16, blood pressure 106/80, heart rate 55 BPM.  HEENT was unremarkable except for wasting of the subcutaneous tissue, tongue was midline. Chest was clear to auscultation, heart was a regular, slow rate and rhythm without murmurs; abdomen was soft and non-tender; he was Tanner Stage I; his neurological exam was normal.  His extremities were thin but capillary fill was acceptable.

Extensive eating disorder interview did not elicit any body dysmorphism and no fear of fat, eating fat or becoming fat; Rex agreed he was too thin and wished he could eat.

Psychiatric evaluation agreed that no anorexia nervosa was present but that a generalized anxiety disorder (ICD9 300.2) probably was.

Rex was admitted to our eating disorder service in the Children’s Hospital medical ward and started on 1500 kcals/day of regular food.  He cried miserably, would not touch any but a few bites of  food and refused to drink his replacement Boost Plus.  When his parents saw that he would be unable to make any further progress, they allowed us to place a nasogastric tube and give him his calories via continuous drip (gentler on the stomach than bolus feeds).

The next day a conference was held with the parents about medication and Rex was started on 1.25 mg Olanzepine (Zyprexa) which was increased within 24 hours to 2.5 mg via the tube at bedtime.  That night his tube “accidentally came out”; it was matter-of-factly replaced.  A throat culture came back normal oropharyngeal growth, his ASO titer was normal 95 IU.ml (0-240); BUN had been slightly elevated on admission (19); creatinine = 0.4 urine drug screen was normal; TSH= 1,450; t4=1.08; ESR=3; Na=140;K=3.9; Cl=106; CO2=23; WBC= 4.0 with ANC 2.02 (normal); Hgb=13.6 and Hct=39.2.  Rex experienced some nocturnal bradycardia with his weight loss and was orthostatic with a change in pulse rate of 50 BPM (lying, daytime) to 109 BPM (standing). By the sixth day he was receiving 2100 kcals/day, still all via NG tube.  He was offered solid food at this point as he stated that he wanted to “get rid of the tube”, but by day was only able to eat 30 percent of his dinner.  By the seventh day he seemed to have “turned a corner” in mood and was able to eat all of his meals.  He was transferred to the Kartini Day Treatment unit weighing 25.2 kg.

Rex spent 14 days in the Day Treatment unit where he was followed by his family therapist, the team pediatrician, psychiatrist, massage therapist, group therapist, occupational therapists, movement therapists, yoga teacher and milieu therapists.  He participated in a ten hour per day program which included all his meals.  By the conclusion of day treatment he needed nearly 3000 kcals to continue to gain weight at 0.2 kg/day.  His Olanzepine (Zyprexa) was not raised and he had some difficulty finishing dinner.  He was able to eat things he had never eaten before; he did well with the predictable schedule of meals on the unit and the predictable menu and was discharged at his premorbid weight of 26.7 kg. He was followed at infrequent intervals at Kartini Clinic for the next eight months where he continued to do well with food, grew 3.7 cm (1.7 inches) and gained weight to 30.7 kg (67 3/4 lbs.) and had advanced to Tanner Stage II pubic hair.  Two months after discharge he was weaned from Olanzepine (Zyprexa), and was back playing soccer with the other boys in his class.

Patient #2:

“Margo” was a sweet nine-year old girl referred to us from a rural California town about six hours drive from Kartini Clinic, for dehydration, weight loss and refusal to eat.  Four weeks prior to her admission Margo had been out to eat at a fast food place where they had all ordered milkshakes with whipped cream on top.  Margo was aware of her history of severe anaphylactic reaction from cashews at 18 months of age and had therefore carefully avoided foods containing nuts.  One of the milkshakes on that day had peanuts sprinkled on top of the cream and Margo was accidentally given that one.  Although she was not allergic to peanuts, and despite her mother pointing this out, Margo panicked after a single bite; she began screaming and crying and spitting on the table. She was eventually able to calm down and recognize she was not experiencing an allergic reaction, however later that evening at the dinner table she only played with her food, stirring it around her plate.  She refused to eat anything even when her parents tried to force her; she refused to drink water as it had touched a straw she felt “was not sterile”.  Over the next few days, in addition to refusing to eat, she began to express escalating concerns about “germs”, opening all doorknobs with her sleeves pulled over her hands and refusing to touch things she could not “verify” were sterile.  She had expressed concerns about cleanliness in the past, but nothing to the extent of her current behaviors.  She insisted on color-coding her closet and arranging her clothes strictly in this way, she reported nightmares where her younger brother entered her room and messed up her closet.  She began to wash her hands after every contact and tried to convince her parents that she would eat things as long as she could agree that they were sterile and which were either “transparent” or “absolutely white”.  Unfortunately, as luck would have it, she was studying toxic chemicals in her science class at school and began to worry about being poisoned.

Margo’s previous medical history was unremarkable except for the anaphylactic reaction at 18 months of age and some “fearfulness” (afraid of the dark and some cleanliness concerns) and “neediness”.  She had been born at term weighing 8 lbs 13 ounces to a 31 year old gravida 2 para 1 Caucasian woman following an uneventful pregnancy, labor and delivery.  She was breastfed for 5 months and had been easy to feed as a young child; she had had a tonsillectomy and adenoidectomy; she took no medications and had no allergies.  She met developmental milestones although she spoke early; she related well to adults but had always had some difficulty fitting in with her own age group.  Margo had not yet achieved menarche and had never had sex voluntarily or involuntarily.  She had never seen a therapist or psychiatrist, never displayed any self-injurious behaviors and never taken any psychotropic medications.

Family history was interesting in that her mother (BMI=31) had a history of bulimia nervosa and panic attacks.  Father (BMI=30) reported being magically fixated on certain numbers, was afraid of germs, disliked having his food touching on his plate and was very perfectionistic.  He was unable to continue working in interior renovation because no project could be completed to his satisfaction within a reasonable amount of time.  Margo’s younger half brother had ADD, disliked having his food touching on his plate, was very concerned with his appearance and was afraid of germs.  There were additional family members who struggled with obesity and depression and a grandmother who suffered from extreme anxiety and perfectionism, sometimes scrubbing the sidewalk outside her house with a toothbrush to eradicate moss.

Margo lived at home with her parents, attended a Christian school in the fourth grade and had always gotten A’s and B’s.  She played the flute. Her mother prepared all the family meals and was home when Margo got home from school; the family ate dinner at the table five nights a week and ate out two nights, although recently this had become impossible with Margo’s escalating symptoms.

On physical exam in our clinic Margo was 57.7 inches tall and weighed 101.5 lbs BMI=21 (>97%tile); blood pressure 112/64, heart rate 78BPM Temperature 99.8  she was Tanner II breast and Tanner I pubic hair; the rest of her exam was unremarkable except for picked cuticles and cracked and peeling hands from washing.

Psychiatric evaluation yielded a diagnosis of obsessive compulsive disorder as well as food phobia.  Eating disorder evaluation revealed no food restrictions to lose weight, no bingeing, no purging, no weight or shape distortions, no inappropriate exercise.

Margo was admitted directly to the hospital after a conference with her parents, preparing them for a lengthy stay as well as for the use of a neuroleptic and placement of the nasogastric tube.  They were desperate for help with Margo’s disruptive food fears and readily agreed.

At the time her tube was placed (on Day One) Margo refused even sips of water.  Zyprexa (Olanzepine) 2.5 mg p.o. q h.s. was initiated.  She was fed1500 kcals/day continuous NG drip, and labs were drawn.  Her Sodium was 142, K=4.2, Cl=108, CO2=24, BUN=14, Creatinine= 0.7, glucose= 89, Hbg=14.8, Hct= 42, ESR= 1, DAU was negative, Bhcg negative, CBC normal, thyroid studies normal, hormones prepubertal.  She was bradycardic that night with a heart rate low of 42 BPM and orthostatic with a pulse from 83 BPM (lying) to 174 BPM (standing).  On Day Two her Zyprexa was increased to 5.0 mg.  By Day Three she said she was willing to try food and it was therefore offered but without success.  Her ASO titer came back a little high at 261 IU/ml (0-240) although her throat culture was negative.  Because of the dim possibility of PANDAS she was given 500 mg Amoxicillin p.o. three times a day for the next week. On Day Four she ate 10% of dinner, but none of anything else. On Day Five she ate bites of lunch and 40% of breakfast.  Her germ fears increased vocally. 

By Day Eight, with the parents consent and briefing, we began behavioral “incentivization”. Margo was allowed incremental increases in her activities if she was able to finish meals and her Zyprexa was increased to 7.5 mg/day.  When this increase in privileges did not seem to motivate her, we instituted a regimen of parental visitation (ordinarily never restricted) that was dependent on her completing her meals.  By the next day, although she was very vocal that she would “eat everything” because she “really wanted Mom to stay the night” she was unable to do so.  On hospital Day Ten because she still was “stuck” in her food refusal her Zyprexa dose was split into 2.5mg given 30 minutes before each meal as well as a bedtime dose (total 10 mg/day).  She was able to eat a few meals and have a few successful and very happy visits with her Mom.  Two days later she gagged up her NG tube when offered a dinner she “didn’t like”.  As she wouldn’t drink her Boost replacement, the NG tube was matter-of-factly replaced.  Thereafter Margo slowly did better with her meals, she seemed much less anxious.  She was interactive with the art therapists and occupational therapists and by Day Sixteen was able to eat 100 percent of 2150 kcals/day.  She was then stepped down to the Kartini Day Treatment unit.

Margo remained in the day treatment unit for ten days; Prozac was introduced and titrated up to 30 mg/day.  Her anxiety about food slowly extinguished, although she did better when her mother adhered to the clinic’s food plan with predictable meals and mealtimes. Her OCD symptoms improved, and several months later was still being followed by the team psychiatric nurse practitioner on a semi-monthly basis.

Patient #3:

This patient illustrated several things about food phobia for us: the failure of pure psychiatric hospitalization without re-feeding, and the failure of medical outpatient care for this condition even with frequent “check-ins”. It is also a cautionary tale about accepting psychiatric diagnoses made by others in a starved child.  It has been our experience working with starved children that, when malnourished, they can look severely depressed, cognitively delayed and even psychotic.  This patient, “Lenora”, had been referred to us as “autistic and psychotic” and because of the severity of these diagnoses, we almost did not accept her.  To our amazement, once she was re-fed, although her long standing diagnosis of autism remained, she was much more social and engaging and not at all psychotic, a real delight to work with.

Lenora was 13 years old and had carried a diagnosis of “high functioning autism” since before age two.  One day at school in early September she was told about a classmate who had died suddenly from a choking incident.  Three days later she was supplied with further graphic details of this choking death by other classmates and from that day forth refused to touch solid foods. She began to complain that “something was stuck in her throat”.  Over the next few weeks Lenora began to reduce the amounts of smooth foods that she would eat as well.  She reported fears that “metastasized” to other arenas of her life: she now feared car accidents, strangers, knives, bumps, bruises, etc.  She developed difficulty sleeping, constant diarrhea and refused to attend school.

Lenora was seen by her pediatrician several times and started on Ativan 0.5 mg three times a day with only minor abatement of her anxiety. She continued to lose weight while waiting for placement on the hospital psychiatric unit.  While waiting, she was placed by her pediatrician on 20 mg Zoloft a day (later increased to 60) and 0.5 mg Risperdal twice daily with no abatement of her eating symptoms and continued weight loss. She was then hospitalized on a pediatric psychiatric ward for 16 days without much improvement and subsequently discharged to a nutritionist, her pediatrician and “medication checks” with an outpatient psychiatrist, still not eating any solids. By early November her weight had dropped from 119 lbs. (in September) to 97 lbs., a total of a 22 lbs. loss (10 kg) and she was referred to Kartini Clinic.

Lenora was born full term to a gravida 2 para 1 Caucasian woman whose only complication of pregnancy, labor and delivery had been the operative removal of a ruptured ovarian cyst at 9.5 weeks of the pregnancy.  She was breast fed for eight months and had been easy to feed as a young child.  Prior to the onset of her food phobia she had eaten everything except oranges.  She had never no history of major illnesses or hospitalizations.  She was allergic to amoxicillin, erythromycin and oral nystatin.  Lenora’s development appeared normal at first, but at 18 months of age she exhibited persistent hand flapping as well as what were called “other autistic-like symptoms”.  She quit speaking and did not speak again until 4-1/2 years of age.  She became very shy and fearful of people, noise and lights and progressively more socially phobic.  She had not achieved menarche, but maternal menarche had been delayed until 15 years of age.  She no history of physical or sexual abuse or trauma.  She had never displayed any self-injurious behaviors.

Family history was negative for other members with autism or related problems. Her mother was tall and slim and suffered from anxiety and depression following her divorce.  Her father had a history of depression, ADHD, alcohol use and learning disability.  Her older sister had no health issues and the rest of the family suffered mostly from angina, arthritis, depression and smoking-related health issues.

Lenora lived at home with her divorced mother, who was devoted to her but experienced pressure from her employer to return to her job as a teacher.  There was some sibling rivalry as well as economic stressors typical for single parent households.

On exam Lenora weighed 100 lbs. and was 64.3 inches tall.  She was fearful and withdrawn, but appeared to know where she was, who she was and what her problem was.  She was Tanner Stage IV for both breasts and pubic region and except for a slight generalized weakness and stiffness to her gait the rest of her exam was unremarkable.

Because of complications with her insurance we were not able to admit Lenora for 10 days. Her mother did not feel she would tolerate N.G. feeds and she was given Ensure Plus orally until soft foods were introduced followed by the gradual introduction of solids. All labs, as outlined above in the previous two cases, were unremarkable.  Her total dose of Risperdal (Risperidone) was gradually increased and then divided into 1.0 mg before breakfast and 1.5 mg at bedtime to contain her fears of choking. Her Zoloft remained at 60 mg. After nine days she was discharged to the Kartini Day Treatment unit on a normal diet, weighing 102 lbs. (46.4 kg) and on discharge to outpatient weighed 48.2 kg (90% of her premorbid weight), was eating 2560 kcals/day solid foods and was much more outgoing and social. Because of the reported incidence of galactorrhea with Risperdal, our psychiatrist gradually shifted her to Seroquel (Quetiapine) 100 mg q hs. One week later her mother noticed an increase in paranoid thoughts which she attributed to the new medication.  At her mother’s request, Lenora was transitioned back to her split doses of Risperdal despite our misgivings about her borderline-high prolactin level of 27.1 ng/ml.  Her Zoloft had been gradually increased to 150mg/day.

One month later Lenora was eating well, sleeping well, and doing well in school.  She received individual therapy with the art therapist to address her panic attacks around her father’s visitation schedule and ongoing adjustment issues faced by a child with a PDD spectrum diagnosis going through adolescence.  Her medications were Risperdal 1.0 mg b.i.d. and Zoloft 200 mg/day. She reported occasional thoughts about choking when she was stressed but otherwise had gained weight to 137 lbs. (BMI=23) and was eating a variety of foods. 

One year after we had first treated her Lenora’s mother changed healthcare providers and Lenora was lost to follow-up.

Discussion:

The London experience, as described in Eating Disorders: A Parents Guide, is somewhat at variance to ours, in that we have given up treating this entity in an outpatient setting, believing that the much more prolonged course of outpatient treatment is inefficient in restoring normal growth and development at a critical stage in a young child’s life.  Other providers, principally adult psychologists or psychiatrists, have reported their experience with outpatient treatment as well.  Chorpita et al. reported a 13- year old girl whose “choking phobia” was treated  with 14 sessions of behavioral and exposure techniques.  Three case studies in adults ages 33, 20 and 28 years,  were reported by Ball, et al. who required 11-13 weekly sessions of between one and 1 1/2 hours to treat successfully.

In our view 12 weeks is too long a time to wait for results with a young child who is wasted, whose growth is “on hold” and who may be dehydrated.  It is furthermore not clear, that the cognitive development of many younger patients will yield the same response to CBT and aversion techniques that it does in adults. We prefer to treat children in the hospital, aggressively, who have lost enough weight to have fallen off their growth curve and are therefore failing to thrive, who are weak and tired, and/or who may be dehydrated. The advantage of initial NG feeds and intravenous fluids is the immediate control of fluid and electrolytes, the psychological relief of frightened and frustrated caretakers, and the focus on the illness as non-volitional (a brain imbalance) which relieves guilt and undermines fights with parents.

The multi-disciplinary nature of our mode of treatment cannot be over-emphasized. Psychological and social consolidation of these gains takes place in the day treatment unit where the milieu therapists make sure that both child and parents are on board with the food and eating plan that will establish good nutrition and growth potential without any fears of choking or vomiting.  This way we promptly restore weight and growth and return the child to full functioning at home and school in a fraction of the time that outpatient CBT/aversion/behavioral techniques would need. Once extinguished, our patients appear to resume normal life without further eating problems.  As with other childhood eating disorders, it is hoped that early and aggressive treatment will improve outcome.