Selective Serotonin Reuptake Inhibitors and Hydroxyzine in the Treatment of Avoidant/Restrictive Food Intake Disorder in Children and Adolescents: Rationale and Evidence

Abstract

Objective:

Despite lack of evidence, various pharmacological agents are judiciously used to manage anxiety in avoidant restrictive food intake disorder (ARFID). We aimed to explore the effectiveness of selective serotonin reuptake inhibitors (SSRIs), either alone or in combination with hydroxyzine, in a well-defined cohort of children and adolescents with ARFID receiving treatment in a partial hospitalization program for eating disorders.

Methods:

We conducted a retrospective chart review of 53 patients with ARFID who were prescribed an SSRI (n = 39) or SSRI with hydroxyzine (n = 14). We investigated changes from admission to discharge in these two medication groups on various outcome measures assessing weight, eating behaviors, mood, anxiety, and fears about food.

Results:

Participants in the SSRI+hydroxyzine group were significantly older than those in the SSRI only group. The majority of participants in both groups exhibited the fear presentation of ARFID. Repeated-measures analysis of variance yielded a significant main effect for treatment for all outcome measures, indicating that patients in both groups experienced improvements in weight, eating behaviors, mood, anxiety, and fears of food. A significant main effect for medication group emerged on the Children's Depression Inventory, suggesting that the group receiving SSRI+hydroxyzine experienced greater depressive symptomatology than the SSRI-only group. We did not find any significant interactions, indicating that participants in both medication groups experienced similar improvements over the course of treatment.

Conclusion:

These results provide preliminary evidence that SSRIs and hydroxyzine may be helpful in the treatment of children and adolescents with ARFID. Given that hydroxyzine was prescribed to patients who experienced high pre- and/or postmeal anxiety, it possibly contributed to similar decreases in anxiety and fear of food in a more challenging subset of patients. Randomized, placebo-controlled studies for children and adolescents with ARFID are warranted to better evaluate and understand the efficacy of SSRIs and hydroxyzine in this clinical population.

Introduction

Avoidant Restrictive Food Intake Disorder (ARFID) is an eating disorder (ED) that typically manifests at an early age and is associated with significant medical and psychosocial consequences (Norris et al. 2014). Individuals in this heterogeneous cohort avoid eating related to sensory aversion, low appetite, and/or feared consequences (e.g., choking and vomiting) independently or in some combination, but do not exhibit a drive for thinness, weight gain fears, or body image distortion.

Patients with ARFID frequently experience comorbid anxiety (Norris et al. 2014), but there are no universally accepted standardized treatment protocols for the management of anxiety in ARFID. Selective serotonin reuptake inhibitors (SSRIs) are commonly used to alleviate various anxiety disorders in children and adolescents (Wehry et al. 2015), and are also used off-label to reduce anxiety in patients with ARFID (Couturier et al. 2019). Disturbances in the serotonergic (5HT) neurotransmitter system have long been established in EDs, and psychotropic medication that modulate 5HT have shown some benefits for EDs (Bailer and Kaye 2011). Abnormality in 5HT also plays a key role in anxiety development and fear maintenance (Wehry et al. 2015). SSRIs are therefore judiciously used to manage anxiety in patients with ARFID despite lack of robust evidence in this population (Couturier et al. 2019).

Prior research suggests that SSRIs may not be helpful for underweight individuals and take weeks to achieve a therapeutic level (Kaye et al. 2011). Given that malnourishment can thwart the effectiveness of SSRIs, D2 5HT2 antagonists such as Olanzapine, have been used to assist patients with ARFID to improve eating behaviors and restore weight (Brewerton et al. 2017; Spettigue et al. 2018). Although there is some evidence for the use of antipsychotics in the initial stages of treatment for EDs, these medications also pose potential negative side effects such as sedation, dizziness, and extrapyramidal movements, which often concern patients and parents (Ballenger 2013). Alternative medications with fewer risks would likely be preferable and more acceptable to patients and parents. Hydroxyzine is an H1 antagonist that has anxiolytic, antihistaminergic, antiemetic, anxiolytic, and sedative benefits. It is used as premedication to reduce anxiety before surgery and to reduce postoperative agitation after administration of general anesthesia (Faytrouny et al. 2007; Koner et al. 2011). Major side effects include sedation, fatigue, and QTc interval prolongation. Sedation and fatigue are less likely with lower doses and slower titration. Drug–drug interactions must be monitored and caution is advised when using with other agents that may prolong QTc interval. Hydroxyzine should be discontinued in case of excessive sedation. Sedative effects quickly wear off after discontinuation secondary to its relatively short half-life. At present there are no specific guidelines for cardiac monitoring. Because premeal anxiety predicts lower food intake (Steinglass et al. 2010), we have used hydroxyzine to reduce pre- and postmeal anxiety to promote nutritional rehabilitation in children with ARFID.

The main aim of this study was to explore the effectiveness of SSRIs and hydroxyzine in a well-defined cohort of patients with ARFID and to provide a comparative review of the rationale and effectiveness of these pharmacological agents in the treatment of ARFID. To our knowledge, this is the first retrospective chart review to assess the effectiveness of either SSRIs or hydroxyzine for children with ARFID. Given our conjecture that children and adolescents with ARFID may have disturbances in the serotonergic (5HT) neurotransmitter system secondary to malnutrition that tends to deplete tryptophan, the precursor amino acid required for serotonin production leading to anxiety development, and that a subset of ARFID patients require adjunct anxiolytic medication to alleviate pre- and postmeal-related anxiety in addition to SSRIs, we hypothesize: (1) participants with ARFID in both groups taking either SSRI alone or an SSRI plus hydroxyzine will show significant improvements over time across outcome measures; (2) participants who are prescribed SSRIs plus hydroxyzine will have more pathology reflected on baseline anxiety measures than the SSRI only group; and (3) despite higher anxiety at baseline in the hydroxyzine group, both groups will make similar improvements over the course of treatment.

Methods

Participants and procedures

We conducted a retrospective chart review of patients with ARFID admitted to our partial hospitalization program (PHP) for children and adolescents with EDs between April 2009 and November 2019, after institutional review board approval. Patients received a diagnosis after interviews with a clinical psychologist and psychiatrist, and each patient's primary presentation of ARFID was verified by a psychologist by chart review. Our chart review selected patients who were: (1) diagnosed with ARFID, and (2) prescribed SSRIs and/or hydroxyzine. We identified 54 participants: 39 were prescribed SSRIs only, 14 were prescribed SSRIs and hydroxyzine, and 1 was prescribed hydroxyzine only (excluded from analyses). SSRIs prescribed in the final sample of 53 participants included: sertraline (32), fluoxetine (14), escitalopram (4), fluvoxamine (2), and citalopram (1).

Upon admission to the program, ARFID patients receive a structured behavior paradigm that included individual, family, group, and meal therapies involving support during two daily meals and two snacks using food exposures as a component of cognitive behavior therapy. Nutrition counseling and psychopharmacological management was provided weekly. Patients were expected to gain 0.5–1 lb/week when on a weight restoration meal plan. Other treatment goals for patients included expansion of food intake diversity, structured eating to avoid malnutrition owing to lack of appetite or lack of interest in food, and decrease in anxiety because of fear of aversive consequences secondary to food intake. Patients experiencing comorbid anxiety were offered SSRIs and those experiencing anxiety secondary to fear of aversive consequences before or after meals were additionally prescribed hydroxyzine on as needed basis to promote meal plan adherence and participation in food exposures. Mean dose for hydroxyzine was 25 mg before or after meals given in divided doses up to four times daily. SSRIs were initiated at a very low dose (fluoxetine 5 mg) and dose was titrated up on a slow schedule based on response. Weight and eating attitudes were monitored weekly to assess treatment response.

Measures

Medical staff measured participants' height and weight, which was used to calculate each participant's percent median body mass index (%MBMI). Psychometric measures evaluated at admission and discharge included the following: Total score and Oral Control subscale of the Children's Eating Attitudes Test (ChEAT) (Maloney et al. 1988); Children's Depression Inventory (CDI) (Kovacs 1992); Revised Children's Manifest Anxiety Scale (RCMAS) (Reynolds and Richmond 1985); and the Foods Eaten and Foods Feared subscales of the Food Acceptance/Fears Survey (FAFS) (Lane-Loney et al. 2020).

Data analysis

Data were analyzed using SPSS version 27.0 (IBM 2020). Chi-square and independent sample t-tests investigated differences between the two medication groups at admission. For our primary analyses, we used a mixed 2 × 2 analysis of variance (ANOVA) to assess: (1) main effects for treatment, or changes within participants from admission to discharge (i.e., repeated measures); (2) main effects for medication group, or differences between those prescribed SSRI only versus SSRI+hydroxyzine (i.e., between groups); and (3) interactions, or differential patterns of change between the two medication groups from admission to discharge. Dependent measures included the following: %MBMI; Total score and Oral Control subscale of the ChEAT; CDI; RCMAS; and the Foods Eaten and Foods Feared subscales of the FAFS.

Results

Demographics

Table 1 provides our sample's demographic information at admission. The SSRI group was significantly younger and had a greater proportion of females than the SSRI+hydroxyzine group. We did not find any other significant differences between groups. The majority of participants in both groups exhibited the fear presentation of ARFID.

Table 1.

Percentages, Mean Scores, and Standard Deviations on Measures Assessing Participant Demographic Variables

Variable	SSRI only (n = 39)	SSRI and atarax (n = 14)	Total (N = 53)	t or χ²
Gender
% Female	92.3	64.3	84.9	0.012^*
% Male	7.7	35.7	15.1	0.012^*
Age (in years)	11.23 (2.52)	13.23 (3.07)	11.76 (2.79)	0.042^*
BMI	15.43 (1.90)	16.74 (3.89)	15.77 (2.60)	0.246
%MBMI	87.96 (10.02)	89.37 (17.53)	88.33 (12.27)	0.779
Duration	8.15 (9.18)	17.57 (17.86)	10.64 (12.61)	0.445
Length of stay
Days	25.67 (11.34)	29.21 (10.06)	26.60 (11.03)	0.284
Weeks	7.15 (3.10)	7.21 (2.23)	7.17 (2.87)	0.938
DSM subtype
% Picky	2.6	7.1	3.8	0.812
% Appetite	15.4	21.4	17.0
% Coprimary	17.9	14.3	17.0

p < 0.05.

%MBMI, percent median body mass index; BMI, body mass index; DSM, Diagnostic and Statistical Manual of Mental Disorders; SSRIs, selective serotonin reuptake inhibitors.

Outcome measures

As noted in Table 2, a mixed 2 × 2 ANOVA yielded a significant main effect for treatment for all outcome measures, including: %MBMI; the Total score and Oral Control subscale of the ChEAT; the CDI; the RCMAS; and the Foods Eaten and Foods Feared subscales of the FAFS. In other words, patients in both groups improved on all measures from admission to discharge. A significant main effect for medication group emerged on the CDI, suggesting that participants in the SSRI+hydroxyzine group experienced greater symptoms of depression than the SSRI group. No other main effects for medication group were found. No significant group × treatment interactions were found, indicating that participants in both the SSRI and SSRI+hydroxyzine groups experienced similar improvements over the course of treatment.

Table 2.

Mean Scores, Standard Deviations, F Values, and η² _p Effect Sizes of Outcome Variables Exhibited in Patients

Outcome variable	SSRI (n = 39)		SSRI and atarax (n = 14)		F ( $η_{p}^{2}$ )
Outcome variable	Intake	Discharge	Intake	Discharge	Treatment	Med. Group	Treatment × med. group
%MBMI	88.29 (9.94)	96.51 (8.60)	89.37 (17.53)	98.91 (15.05)	115.28^*** (0.697)	0.246 (0.005)	0.647 (0.013)
ChEAT
Total	11.28 (6.37)	6.44 (7.59)	11.62 (10.77)	7.23 (6.66)	10.71^** (0.186)	0.079 (0.002)	0.025 (0.001)
Oral control	7.50 (4.19)	3.69 (4.19)	6.92 (4.63)	3.85 (3.24)	18.17^*** (0.279)	0.040 (0.001)	0.204 (0.004)
CDI	53.09 (9.61)	47.20 (10.79)	62.46 (16.51)	52.38 (13.01)	17.21^*** (0.272)	5.04^* (0.099)	1.19 (0.025)
RCMAS	51.50 (11.75)	45.19 (14.15)	54.46 (16.37)	50.08 (14.10)	8.13^** (0.148)	0.966 (0.020)	0.263 (0.006)
FAFS
Eaten	38.84 (21.49)	62.47 (13.35)	39.00 (24.76)	57.15 (17.38)	30.29^*** (0.413)	0.273 (0.006)	0.519 (0.012)
Feared	31.31 (25.93)	6.63 (9.21)	33.83 (24.48)	15.75 (17.31)	24.62^*** (0.354)	1.32 (0.028)	0.587 (0.013)

p < 0.05, ^** p < 0.01, ^*** p < 0.001.

CDI, Children's Depression Inventory—total; ChEAT, Children's Eating Attitudes Test—total and oral control subscale; FAFS, Food Acceptance/Fears Survey; RCMAS, Revised Children's Manifest Anxiety Scale—total.

Discussion

There is paucity of literature about evidence-based treatment options for ARFID. Several small studies have reported the beneficial role of family therapy, D2 5HT2 antagonists like olanzapine and risperidone, antihistaminergic agents like mirtazapine and cyproheptadine as well as benzodiazepines (Couturier et al. 2019).

This article provides a rationale for the use of SSRIs and hydroxyzine for ARFID, and describes results from the first systematic study reporting on the use of these medications in children with ARFID. Patients in both groups, who received these medications while receiving treatment at a PHP for EDs, experienced significant clinical improvements in weight, eating behaviors, mood, anxiety, and fears about food. These results provide preliminary evidence that SSRIs, alone and in combination with hydroxyzine, may be beneficial in the treatment of ARFID. Both medication options were well tolerated by patients. Mild fatigue or sedation was reported as side effect for hydroxyzine, whereas transient headaches were reported by patients taking SSRIs.

Pre- and postmeal anxiety leads to food avoidance and is often seen in the fear presentation of ARFID. We found that hydroxyzine facilitated meal plan compliance and participation in food exposures by decreasing the situation-specific anxiety in the group that struggled with pre- and postmeal anxiety in addition to a comorbid anxiety disorder.

In contrast to our expectations, patients with ARFID who were prescribed both an SSRI and hydroxyzine did not score higher on the RCMAS, a measure of anxiety, than those who were prescribed an SSRI alone. Instead, the SSRI+hydroxyzine group exhibited more depressive psychopathology on the CDI, and were significantly older, than the SSRI only group. Although we initially hypothesized that these eating difficulties were related to general symptoms of anxiety, it may instead be the case that a combination of anxiety and depressive symptomatology, paired with being slightly older, contributed to greater eating difficulties, particularly pertaining to pre- and postmeal anxiety that then resulted in an increased likelihood of being prescribed hydroxyzine to assist with anxiety during meals.

We did not find significant interaction effects, indicating that participants in the SSRI+hydroxyzine group experienced similar improvements as those in the SSRI only group. Hydroxyzine was prescribed as an adjunctive medication to patients who experienced particularly high pre- and/or postmeal anxiety and struggled to complete meals. Thus, it may be the case that hydroxyzine contributed to similar decreases in anxiety and fear of food in a more challenging subset of patients.

This retrospective study has several critical limitations. Of note, we cannot draw conclusions about the degree to which psychiatric medication versus other treatment components resulted in improved outcomes, and our small sample size prevented us from detecting smaller effects. Without a control condition, we do not know whether improvements were related to the medication, other aspects of treatment, or placebo effects. Although improvements in body weight, on the Oral Control subscale of the ChEAT, and on the subscales of the FAFS have been used as outcome measure in prior research for ARFID (Lane-Loney et al. 2020), we did not use measures that specifically assessed for ARFID symptomatology, as more relevant measures were yet to be developed during the time of this chart review. We also failed to systematically measure pre- and postmeal anxiety, which would be necessary to test our hypothesis that hydroxyzine assists individuals with ARFID by targeting eating-related anxiety. Randomized, placebo-controlled studies in children and adolescents with relevant ARFID measures, including pre- and postmeal anxiety, are warranted to better evaluate and understand the efficacy of SSRIs and hydroxyzine in this clinical population.

Conclusion

This retrospective chart review revealed that children with ARFID who were prescribed an SSRI, either alone or in combination with hydroxyzine, experienced improvements in weight, eating behaviors, mood, anxiety, and fears about food. Hydroxyzine was found to be effective in alleviating pre- and postmeal anxiety in a more challenging subset of patients. Given that SSRIs and hydroxyzine are relatively safe medications with fewer long-term side effects and no risk for dependence, they are appealing psychiatric medications for children with ARFID, and should be evaluated in rigorous randomized placebo-controlled research.

Clinical Significance

We retrospectively reviewed results of hydroxyzine prescribed as adjunct anti-anxiety medication for a more challenging subset of ARFID patients and found that patients reported subjective decrease in pre- and postmeal anxiety and had improved meal plan adherence. Hydroxyzine has been approved by FDA for anxiety in children. It is recommended as 0.5 mg/kg dose given every 4–6 hours as needed. Onset of action is within 15–30 minutes. Time to peak is ∼2 hours (Simons et al. 1984) and may therefore assist with anxiety before, during, and after completing anxiety-provoking meals. We have also found it to be effective in ARFID patients who report fear of vomiting after eating, possibly secondary to its antiemetic effects. In addition, it is a user-friendly choice for patients afraid of swallowing owing to availability in liquid form. Participants who received SSRI only experienced similar improvements over the course of treatment. SSRI dosing should start low (5 mg fluoxetine) to avoid paradoxical worsening of anxiety symptoms. Slow dose titration (e.g., increments of 5 mg) minimizes risk of precipitating side effects in malnourished ARFID patients. Level of anxiety and side effects should be closely monitored with each dose titration and after discontinuation.

Footnotes

Disclosures

No competing financial interests exist.

References

Bailer

, Kaye

: Serotonin: Imaging findings in eating disorders. Curr Top Behav Neurosci, 6:59–79, 2011.

Ballenger

: Adverse effects of second-generation antipsychotics in children and adolescents: A Bayesian meta-analysis. Yearb Psychiatry Appl Mental Health, 2013:354–355, 2013.

Brewerton

, D'Agostino

: Adjunctive use of olanzapine in the treatment of avoidant restrictive food intake disorder in children and adolescents in an eating disorders program. J Child Adolesc Psychopharmacol, 27:920–922, 2017.

Couturier

, Isserlin

, Spettigue

, Norris

: Psychotropic medication for children and adolescents with eating disorders. Child Adolesc Psychiatric Clin N Am, 28:583–592, 2019.

Faytrouny

, Okte

, Kucukyavuz

: Comparison of two different dosages of hydroxyzine for sedation in the pediatric dental patient. Int J Pediatr Dent, 17:378–382, 2007.

IBM Corp. IBM SPSS Statistics for Windows, Version 27.0. Armonk, NY, IBM Corp., 2020.

Kaye

, Wagner

, Fudge

, Paulus

: Neurocircuity of eating disorders. Curr Top Behav Neurosci, 6:37–57, 2011.

Köner

, Türe

, Mercan

, Menda

, Sözübir

: Effects of hydroxyzine-midazolam premedication on sevoflurane-induced pediatric emergence agitation: A prospective randomized clinical trial. Euro J Anesthesiol, 28:640–645, 2011.

Kovacs

: The Children's Depression Inventory (CDI). Multi-Health Systems, New York, 1992.

10.

Lane-Loney

, Zickgraf

, Ornstein

, Mahr

, Essayli

: A cognitive-behavioral family-based protocol for the primary presentations of avoidant/restrictive food intake disorder (ARFID): Case examples and clinical research findings. Cognit Behav Pract, 2020 (In press).

11.

Maloney

, McGuire

, Daniels

: Reliability testing of a children's version of the Eating Attitude Test. J Am Acad Child Adolesc Psychiatry, 27:541–543, 1988.

12.

Norris

, Robinson

, Obeid

, Harrison

, Spettigue

, Henderson

: Exploring avoidant/restrictive food intake disorder in eating disordered patients: A descriptive study. Int J Eating Disorders, 47:495–499, 2014.

13.

Reynolds

, Richmond

: Revised Children's Manifest Anxiety Scale. Los Angeles, CA, Western Psychological Services, 1985.

14.

Simons

, Simons

, Frith

: The pharmacokinetics and antihistaminic of the H1 receptor antagonist hydroxyzine. J Allergy Clin Immunol, 73:69–75, 1984.

15.

Spettigue

, Norris

, Santos

, Obeid

: Treatment of children and adolescents with avoidant/restrictive food intake disorder: A case series examining the feasibility of family therapy and adjunctive treatments. J Eat Disord, 6:20, 2018.

16.

Steinglass

, Sysko

, Mayer

, Berner

, Schebendach

, Wang

, Chen

, Albano

, Simpson

, Walsh

: Pre-meal anxiety and food intake in anorexia nervosa. Appetite, 55:214–218, 2010.

17.

Wehry

, Beesdo-Baum

, Hennelly

, Connolly

, Strawn

: Assessment and treatment of anxiety disorders in children and adolescents. Curr Psychiatry Rep, 17:52, 2015.