Mehr S, Kakakios A, Frith K, Kemp AS. Pediatrics. 2009;123(3). Available at: www.pediatrics.org/cgi/content/full/123/3/e459
PURPOSE OF THE STUDY. To investigate possible patterns in demographic features, causative foods, clinical features, treatments at presentation, and outcomes in children diagnosed with food protein-induced enterocolitis syndrome (FPIES).
STUDY POPULATION. Retrospective chart review of 35 children who presented with acute FPIES at a tertiary medical center in New South Wales, Australia, between 1992 and 2007.
METHODS. Diagnosis was made by pediatric allergists after referral to the allergy clinic (74%) or from the emergency department (ED) (26%), using previously published criteria. Cases were identified by codes signifying allergic and dietetic gastroenteritis and colitis or by searching for key words in letters written by allergists.
RESULTS. Thirty-two children fulfilled all criteria, and 3 presented with 1 typical episode and no other causal explanation. Sixty-six episodes were recorded, with a mean presenting age of 5.5 months and a median of 2.2 episodes before diagnosis (range: 1–4 episodes). Most children reacted to 1 food, and 6 children reacted to 2 foods. Causative foods included rice (n = 14), soy (n = 12), cow's milk (n = 7), oat (n = 2), sweet potato (n = 2), banana (n = 1), fish (n = 1), chicken (n = 1), and lamb (n = 1). The mean time from ingestion to reaction was 1.8 hours. Symptoms included vomiting (100%), lethargy (85%), pallor (67%), and diarrhea (24%). Information regarding evaluation of 64 episodes included admission from the ED (25 of 39 visits), abdominal imaging (34%), septic evaluation (28%), and surgical consultation (22%). A decreased body temperature of <36°C was noted in 6 (24%) of 25 episodes. Thrombocytosis not accounted for by hemoconcentration was noted in 15 (63%) of 24 blood counts performed. Only 2 of 19 initial cases presenting to the ED were correctly diagnosed. Other initial diagnoses included food allergy (26%), viral infection/sepsis (21%), gastroenteritis (21%), resolved intussusception (11%), or no diagnosis (11%). Treatments at presentation included intravenous fluid resuscitation (n = 19), antibiotics (n = 8), oxygen (n = 6), air or barium enema (n = 4), parenteral epinephrine treatment (n = 2), and laparotomy (n = 1). Tolerance was demonstrated by 3 years of age in 5 of 6 undergoing soy challenges and 4 of 5 undergoing rice challenges.
CONCLUSIONS. Delayed diagnosis and misdiagnosis is common in FPIES, leading to incorrect and/or invasive treatment. Thrombocytosis, in addition to previously recognized leukocytosis, may be a laboratory clue upon initial presentation. Diarrhea and body temperature of <36°C were associated with more-severe episodes. Foods commonly considered hypoallergenic (ie, rice) may cause FPIES. The prognosis of developing tolerance by age 3 years is favorable.
REVIEWERS COMMENTS. Currently, FPIES is a clinical diagnosis. The authors attempted to identify subjective criteria that may be used to diagnose FPIES; however, thrombocytosis and decreased body temperature are factors that will continue to lead to other common diagnoses, such as sepsis or gastroenteritis. The authors argue that, although cases increased in incidence during the 16 years, the number of episodes before diagnosis remained the same, indicating continued misdiagnosis. This, along with the inappropriate and sometimes risky treatments used in error, points out the need for greater awareness of the symptom pattern and triggers, especially rice, milk, and soy.
- Copyright © 2009 by the American Academy of Pediatrics