Invited Commentary Re: Prevalence and Characteristics of Avoidant/ Restrictive Food Intake Disorder in Pediatric Neurogastroenterology Patients

See “Prevalence and Characteristics of Avoidant/Restrictive Food Intake Disorder in Pediatric Neurogastroenterology Patients” by Murray et al on page 588.

The article by Murray et al describes a prevalence of Avoidant/ Restrictive Food Intake Disorder (ARFID) in up to 23% of pediatric patients who presented to a neurogastroenterology clinic. Electronic medical records were reviewed retrospectively and coders obtained the information from documentation by several providers and categorized them into “definite ARFID” (8.5%) if they met all Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria for the condition or “potential ARFID” (14.7%) if they met some criteria ⁽¹⁾. There are some concerns with the methodology of this study as well as interpretation of the results that should be addressed. Firstly, a retrospective review is likely not an ideal way to assess the prevalence of ARFID in this population, especially as ARFID is a relatively newly recognized condition and all providers’ knowledge about this condition may not be the same. The data collected would, therefore, be subject to whether a particular provider asked specific questions that would fulfill the criteria for diagnosing ARFID, a condition that most gastroenterology providers (GI) providers may not be aware of. Additionally, as the authors’ point out, there were no validated screening tools available for ARFID at the time, which could have been used by the providers. Regarding the prevalence of ARFID, it would be erroneous to state that the prevalence of ARFID in the patients attending the neurogastroenterology clinic was up to 23% as only 8.5% (11 of 129 patients) had “definite ARFID” and the rest did not meet all the criteria for ARFID. It should also be clarified that this clinic would typically see not just patients suspected of having Disorders of Brain Gut Interaction (DGBI) but also those with motility disorders. Hence the prevalence quoted for ARFID would be in this entire population and not restricted to DGBI patients alone.

A higher prevalence of psychiatric comorbidity, especially anxiety, in patients with DGBI is well established. More recently, disordered eating was reported to be highly prevalent (13%—55%) in adult patients with DGBI ⁽²⁾. Anorexia, a more recognized eating disorder based on extreme restriction of food intake and fear of weight gain, is often confused with the ARFID, a recent addition to the list of DSM-5 ⁽³⁾. Additionally, as the authors correctly mention, restricting oral food intake to prevent symptoms is a key feature shared by DBGI and ARFID. Despite some similar symptoms, DGBI and ARFID are, however, 2 clinically distinct conditions. DBGIs can be diagnosed based on the criteria described by the Rome Foundation and ARFID by the diagnostic criteria as enumerated in the DSM-5 ⁽⁴⁾. Even though these are 2 separate disorders and require distinct management approaches, their coexistence in some patients makes the management more challenging. There is an unmet need to identify patients with this overlap so that appropriate psychological intervention can be included in their management. Lastly, one of the criteria for ARFID is “requirement of artificial nutrition support is beyond of what would be expected for that condition”. In order to determine that, one should subjectively assess the presentation of each patient and apply clinical judgement—the appropriate assessment of this criteria cannot be made by the review of diagnostic codes ^(5,6).

Psychological stress has been shown to modulate the processing of sensory signals coming to the brain from the gut and contributing to DGBI. Regrettably there is no readily available test to evaluate sensory dysfunction in the gut. Hormones that play a role in sensory-motor functioning of the gut include CCK, GLP-1, PYY, and Ghrelin and these also act in the regulation of appetite. These hormones are released or suppressed in response to fasting or fed states. A disparity in these circulating hormones that lead to high levels of 'satiating’ gut hormones can result in “delayed gastric emptying” ^(7,8). In a patient with an eating disorder where the intake can be erratic and intercalated by prolonged periods of fasting, there is an imbalance in the levels of these hormones ⁽⁹⁾. During critical illnesses, studies have shown that levels of CCK and PYY correlate with feeding intolerance ⁽⁸⁾ and a similar pattern has been described in patients with anorexia nervosa ⁽⁹⁾.There is evidence that with nutritional rehabilitation, there is improved gut functioning and food tolerance associated with a decrease in symptoms. In a study by Heruc et al, patients with anorexia nervosa were shown to have delayed gastric emptying, which improved after refeeding ⁽¹⁰⁾.

It is often challenging to definitively determine the cause and effect, if any, in patients with concerns for gut dysmotility who have ARFID leading to poor nutrition. The temporal occurrence of one presenting before the other suggests that the former may be the trigger for developing the latter. Patients and families have a difficult time recognizing and accepting this potential relationship and desperately try to find a physical reason to explain the plethora of functional GI symptoms. This should be entertained and discussed at length early on in their care to avoid the risk of incurring expensive, invasive, and unnecessary tests.

Despite the caveats of retrospective data analysis and its interpretation, the findings of this study underscore the need for screening for ARFID in children with DGBI and vice versa. Early recognition and preventive intervention of ARFID in those with these co-existent conditions can result in improved outcomes. Clearly, care of this cohort of children would necessitate a patient-centered interdisciplinary approach by a team of gastro-enterologists, dieticians, eating disorders specialists, therapists, educators, and social workers. Interdisciplinary team members do not necessarily have to be located together at the same time and place to evaluate these patients. Communication, however, between members is key, and this can be done virtually or in-person.

A major barrier to providing this level of service is that health care providers lack familiarity and knowledge of this condition, resulting in delays in diagnosis and optimal treatment and outcomes. This explains some of the hesitancy in diagnosing and low confidence in managing this complex medical condition in children. Increasing awareness of the recently available structured tools to assess symptoms of ARFID, as well as the Rome IV questionnaire for diagnosing DGBI in children, will hopefully help medical providers diagnose children earlier and refer them to an appropriate center for management. As a future direction, it would be interesting to know what neurogastrointestinal diagnoses were ultimately assigned to patients classified as having ARFID. This information would underscore the overlap between neurogastrointestinal pathology and ARFID.