Advertisement

Clinical features and resolution of food protein–induced enterocolitis syndrome: 10-year experience

  • Author Footnotes
    ∗ These authors contributed equally to this work.
    Jean Christoph Caubet
    Footnotes
    ∗ These authors contributed equally to this work.
    Affiliations
    Jaffe Food Allergy Institute, Department of Pediatrics, Division of Allergy and Immunology, Kravis Children's Hospital, Icahn School of Medicine at Mount Sinai, New York, NY

    Division of Pediatric Allergy, University Hospital of Geneva, Geneva, Switzerland
    Search for articles by this author
  • Author Footnotes
    ∗ These authors contributed equally to this work.
    Lara Simone Ford
    Footnotes
    ∗ These authors contributed equally to this work.
    Affiliations
    Jaffe Food Allergy Institute, Department of Pediatrics, Division of Allergy and Immunology, Kravis Children's Hospital, Icahn School of Medicine at Mount Sinai, New York, NY

    Pediatrics & Child Health, Children's Hospital, Westmead, Australia
    Search for articles by this author
  • Laura Sickles
    Affiliations
    Jaffe Food Allergy Institute, Department of Pediatrics, Division of Allergy and Immunology, Kravis Children's Hospital, Icahn School of Medicine at Mount Sinai, New York, NY

    Jefferson Medical College, Philadelphia, Pa
    Search for articles by this author
  • Kirsi M. Järvinen
    Affiliations
    Jaffe Food Allergy Institute, Department of Pediatrics, Division of Allergy and Immunology, Kravis Children's Hospital, Icahn School of Medicine at Mount Sinai, New York, NY

    Division of Allergy and Immunology & Center for Immunology and Microbial Diseases, Albany Medical College, Albany, NY
    Search for articles by this author
  • Scott H. Sicherer
    Affiliations
    Jaffe Food Allergy Institute, Department of Pediatrics, Division of Allergy and Immunology, Kravis Children's Hospital, Icahn School of Medicine at Mount Sinai, New York, NY
    Search for articles by this author
  • Hugh A. Sampson
    Affiliations
    Jaffe Food Allergy Institute, Department of Pediatrics, Division of Allergy and Immunology, Kravis Children's Hospital, Icahn School of Medicine at Mount Sinai, New York, NY
    Search for articles by this author
  • Anna Nowak-Węgrzyn
    Correspondence
    Corresponding author: Anna Nowak-Węgrzyn, MD, Division of Allergy and Immunology, Pediatrics Mount Sinai School of Medicine, New York, NY 10029.
    Affiliations
    Jaffe Food Allergy Institute, Department of Pediatrics, Division of Allergy and Immunology, Kravis Children's Hospital, Icahn School of Medicine at Mount Sinai, New York, NY
    Search for articles by this author
  • Author Footnotes
    ∗ These authors contributed equally to this work.

      Background

      Food protein–induced enterocolitis syndrome (FPIES) is a non–IgE-mediated food allergy. FPIES diagnosis is frequently delayed because of the absence of classic allergic symptoms and lack of biomarkers.

      Objective

      We sought to characterize the clinical features and resolution of FPIES in patients evaluated in our practice.

      Methods

      Subjects 6 months to 45 years of age with FPIES were prospectively recruited for oral food challenges (OFCs). Medical records were searched to identify the subjects who did not participate in OFCs.

      Results

      Among 160 subjects, 54% were male; median age at diagnosis was 15 months. We performed 180 OFCs to 15 foods in 82 subjects; 30% of the study population had FPIES confirmed based on OFC results. The most common foods were cow's milk (44%), soy (41%), rice (22.5%), and oat (16%). The majority (65%) reacted to 1 food, 26% reacted to 2 foods, and 9% reacted to 3 or more foods. The majority were atopic, and 39% had IgE sensitization to another food. Thirty-nine (24%) subjects had positive specific IgE levels to the food inducing FPIES. Among children with specific IgE to cow's milk, 41% changed from a milk FPIES to an IgE-mediated phenotype over time. The median age when tolerance was established was 4.7 years for rice, 4 years for oat, and 6.7 years for soy. Median age when milk tolerance was established for subjects with undetectable milk-specific IgE levels was 5.1 years, whereas none of the subjects with detectable milk-specific IgE became tolerant to milk during the study (P = .003).

      Conclusion

      FPIES typically resolves by age 5 years. Milk FPIES, especially with detectable food-specific IgE, can have a protracted course and eventually transition to acute reactions.

      Key words

      Abbreviations used:

      ANC (Absolute neutrophil count), FPIES (Food protein–induced enterocolitis syndrome), IQR (Interquartile range), OFC (Oral food challenge), SPT (Skin prick test)
      Food protein–induced enterocolitis syndrome (FPIES) is a non–IgE-mediated food allergy that usually presents in young infants and manifests as profuse repetitive vomiting and lethargy, typically occurring 2 to 4 hours after ingestion of the offending allergen and occasionally followed by diarrhea 5 to 10 hours later.
      • Boyce J.A.
      • Assa'ad A.
      • Burks A.W.
      • Jones S.M.
      • Sampson H.A.
      • Wood R.A.
      • et al.
      Guidelines for the diagnosis and management of food allergy in the United States: report of the NIAID-sponsored expert panel.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome: case presentations and management lessons.
      • Järvinen K.M.
      • Nowak-Wegrzyn A.
      Food protein-induced enterocolitis syndrome (FPIES): current management strategies and review of the literature.
      Chronic exposure manifests as intermittent emesis and chronic diarrhea with blood, mucus, or both and can result in failure to thrive and hypoalbuminemia.
      • Sicherer S.H.
      • Eigenmann P.A.
      • Sampson H.A.
      Clinical features of food protein-induced enterocolitis syndrome.
      Cow's milk and soy, followed by rice and oat, are the most common causes of FPIES in the United States, although there are reports of reactions to many other foods, including egg, poultry, beans, vegetables, and seafood.
      • Nowak-Wegrzyn A.
      • Sampson H.A.
      • Wood R.A.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome caused by solid food proteins.
      • Ruffner M.A.
      • Ruymann K.
      • Barni S.
      • Cianferoni A.
      • Brown-Whitehorn T.
      • Spergel J.M.
      Food Protein-induced enterocolitis syndrome: insights from review of a large referral population.
      • Caubet J.C.
      • Nowak-Wegrzyn A.
      Food protein-induced enterocolitis to hen's egg.
      • Sopo S.M.
      • Giorgio V.
      • Dello Iacono I.
      • Novembre E.
      • Mori F.
      • Onesimo R.
      A multicentre retrospective study of 66 Italian children with food protein-induced enterocolitis syndrome: different management for different phenotypes.
      • Nowak-Wegrzyn A.
      • Muraro A.
      Food protein-induced enterocolitis syndrome.
      • Bruni F.
      • Peroni D.G.
      • Piacentini G.L.
      • De Luca G.
      • Boner A.L.
      Fruit proteins: another cause of food protein-induced enterocolitis syndrome.
      • Monti G.
      • Viola S.
      • Tarasco V.
      • Lupica M.M.
      • Cosentino V.
      • Castagno E.
      A case of severe allergic reaction to cooked potato.
      FPIES has been described in adults, usually related to shellfish ingestion.
      • Fernandes B.N.
      • Boyle R.J.
      • Gore C.
      • Simpson A.
      • Custovic A.
      Food protein-induced enterocolitis syndrome can occur in adults.
      • Leonard S.A.
      • Nowak-Wegrzyn A.
      Food protein-induced enterocolitis syndrome: an update on natural history and review of management.
      Prevalence estimates of FPIES are limited to a single report from Israel that found 0.34% of infants with FPIES attributable to cow's milk; the same population included 0.5% with IgE- mediated cow's milk allergy.
      • Katz Y.
      • Goldberg M.R.
      • Rajuan N.
      • Cohen A.
      • Leshno M.
      The prevalence and natural course of food protein-induced enterocolitis syndrome to cow's milk: a large-scale, prospective population-based study.
      FPIES is underrecognized, and the diagnosis is not straightforward. Skin prick test (SPT) responses or serum allergen-specific IgE antibody results are typically negative.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome: case presentations and management lessons.
      • Leonard S.A.
      • Nowak-Wegrzyn A.
      Food protein-induced enterocolitis syndrome: an update on natural history and review of management.
      Considering that FPIES pathophysiology is presumably T cell mediated,
      • Caubet J.C.
      • Nowak-Wegrzyn A.
      Current understanding of the immune mechanisms of food protein-induced enterocolitis syndrome.
      patch tests have been proposed for diagnosing FPIES.
      • Fogg M.I.
      • Brown-Whitehorn T.A.
      • Pawlowski N.A.
      • Spergel J.M.
      Atopy patch test for the diagnosis of food protein-induced enterocolitis syndrome.
      However, we and others have shown that atopy patch tests to common food allergens had poor utility in the follow-up management of FPIES.
      • Ruffner M.A.
      • Ruymann K.
      • Barni S.
      • Cianferoni A.
      • Brown-Whitehorn T.
      • Spergel J.M.
      Food Protein-induced enterocolitis syndrome: insights from review of a large referral population.
      • Fernandes B.N.
      • Boyle R.J.
      • Gore C.
      • Simpson A.
      • Custovic A.
      Food protein-induced enterocolitis syndrome can occur in adults.
      • Jarvinen K.M.
      • Caubet J.C.
      • Sickles L.
      • Ford L.S.
      • Sampson H.A.
      • Nowak-Wegrzyn A.
      Poor utility of atopy patch test in predicting tolerance development in food protein-induced enterocolitis syndrome.
      US food allergy guidelines
      • Boyce J.A.
      • Assa'ad A.
      • Burks A.W.
      • Jones S.M.
      • Sampson H.A.
      • Wood R.A.
      • et al.
      Guidelines for the diagnosis and management of food allergy in the United States: summary of the NIAID-Sponsored Expert Panel Report.
      recommend using the medical history and oral food challenge (OFC) results to establish a diagnosis of FPIES. A confirmatory OFC is considered unnecessary when the typical symptoms occur within 2 to 4 hours after food ingestion (particularly more than once), there is no alternative explanation for the symptoms, and the child remains well if the food is eliminated from the diet.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome: case presentations and management lessons.
      Data on the resolution of FPIES are lacking, particularly for solid foods.
      The aim of our study was to characterize the clinical features and resolution of FPIES to improve the management of subjects with this condition.

      Methods

       Study population

      The research protocol was approved by the Mount Sinai Institutional Review Board. Written informed consent was obtained before enrollment. This study includes a mixed prospective and retrospective method of patient ascertainment. Subjects aged from 6 months to 45 years with FPIES were prospectively recruited for OFCs under the natural history of the FPIES protocol from the Mount Sinai Allergy and Immunology Clinic patient population (New York, NY) between 2001 and 2011. All subjects with suspected FPIES were offered participation; for those who declined OFCs, the medical records database at the Mount Sinai Allergy and Immunology Clinic was searched. The terms FPIES, food protein–induced enterocolitis syndrome, and enterocolitis were used to identify the subjects who chose not to participate in OFCs. These patients were followed clinically to assess the natural history of FPIES. The diagnosis of FPIES was based on Powell's clinical criteria
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome: case presentations and management lessons.
      • Powell G.K.
      Food protein-induced enterocolitis of infancy: differential diagnosis and management.
      : (1) exposure to the incriminated food elicits repetitive vomiting, diarrhea, or both within 4 hours without any other cause for the symptoms; (2) symptoms are limited to the gastrointestinal tract; (3) avoidance of the offending protein from the diet results in resolution of symptoms; and (4) a standardized OFC or isolated re-exposure elicits the typical symptoms. Age of diagnosis was defined as the age at which the physician suspected a diagnosis of FPIES, which could have been before presenting at Mount Sinai. The age of resolution was determined with either OFCs or parental report of food introduction at home. Exclusion criteria included a history of immediate (<2 hours) allergic reactions with cutaneous symptoms, respiratory symptoms, or both to the suspect food and a biopsy-confirmed diagnosis of allergic eosinophilic gastroenteritis or allergic eosinophilic esophagitis.

       Study procedures

      During the study visits, subjects underwent SPTs with bifurcated needles (Allergy Labs of Ohio, Columbus, Ohio) with commercial food extracts (Greer Laboratories, Lenoir, NC), saline (negative control), and histamine (positive control); results were read between 10 and 15 minutes after the prick.
      • Knight A.K.
      • Shreffler W.G.
      • Sampson H.A.
      • Sicherer S.H.
      • Noone S.
      • Mofidi S.
      • et al.
      Skin prick test to egg white provides additional diagnostic utility to serum egg white-specific IgE antibody concentration in children.
      Serum food-specific IgE antibody levels were measured with UniCAP (Thermo Fisher, Portage, Mich), with a lower limit of detection of 0.35 kUA/L and an upper limit of detection of 100 kUA/L.
      OFCs were performed after a minimum of 12 months after the most recent FPIES reaction to a specific food. A subset of subjects underwent periodic OFCs to the same food to evaluate possible resolution. In children with multiple food–induced FPIES, history of severe reactions to 1 food, or both, OFCs were sometimes done to foods that had never been ingested previously but were avoided as a precaution. OFCs were undertaken in the Mount Sinai Clinical Research Unit according to the guidelines of Powell,
      • Powell G.K.
      Food protein-induced enterocolitis of infancy: differential diagnosis and management.
      as modified by Sicherer and colleagues.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome: case presentations and management lessons.
      • Sicherer S.H.
      • Eigenmann P.A.
      • Sampson H.A.
      Clinical features of food protein-induced enterocolitis syndrome.
      A peripheral intravenous line was placed before the OFC. Subjects were given from 0.06 g up to 0.6 g of food protein per kilogram of body weight (usually 0.3 g of protein/kg body weight; maximum, 3 g of protein) in 3 equal doses over a 45-minute period and remained under observation for 4 to 8 hours after the ingestion of the challenge food. In the case of grains with low protein content, such as rice and oat, an age-appropriate food portion was served. We used modified Powell criteria to determine the positivity of an OFC result: (1) emesis, diarrhea, or both and (2) increase in peripheral polymorphonuclear leukocyte count of 3500 cells/mm
      • Järvinen K.M.
      • Nowak-Wegrzyn A.
      Food protein-induced enterocolitis syndrome (FPIES): current management strategies and review of the literature.
      or greater. The OFC result was considered positive if both criteria were met, equivocal if 1 criterion was met, and negative if both were negative.
      • Powell G.K.
      Food protein-induced enterocolitis of infancy: differential diagnosis and management.
      If symptoms of a reaction developed, subjects were treated and kept under observation for approximately 6 hours. A CBC with differential was obtained immediately before the OFC and before discharge (ie, at 6 hours if the OFC result was positive). The treatment was administered at the attending physician's discretion and included intravenous normal saline bolus, intravenous corticosteroids, or both. In the subjects who had specific IgE to the incriminated food during follow-up, the OFC protocol was adapted by administering incrementally increasing doses of the food protein every 15 minutes, as per the protocol for IgE-mediated food allergy.
      • Sampson H.A.
      • Gerth van Wijk R.
      • Bindslev-Jensen C.
      • Sicherer S.
      • Teuber S.S.
      • Burks A.W.
      • et al.
      Standardizing double-blind, placebo-controlled oral food challenges: American Academy of Allergy, Asthma & Immunology-European Academy of Allergy and Clinical Immunology PRACTALL consensus report.

       Statistical analysis

      Analyses were performed with SAS/STAT Version 9.2 software (SAS Institute, Cary, NC). Comparisons of risk factors and between-group data for continuous variables were assessed with a t test for independent variables or a Mann-Whitney U test. A P value of less than .05 was considered significant. Cumulative survival curves were drawn by using the Kaplan-Meier method to investigate the natural course of FPIES.

      Results

       Subjects' characteristics

      Subjects' characteristics and the offending foods are summarized in Table I and Fig 1. One hundred sixty subjects were enrolled. Eighty-six 86 (54%) were male, and the median age at diagnosis was 15 months (25% to 75% interquartile range [IQR], 9-24 months). Among those 160 patients, 82 (51%) were enrolled prospectively, and 78 were identified through the retrospective chart review. Thirteen (8%) subject received a diagnosis after 5 years of age; 5 of them had FPIES to fish or shellfish with late onset. Six subjects with FPIES to milk and 2 with FPIES to soy or egg were given diagnoses after 5 years of age, although symptoms appeared earlier in life (median age of 9 months; IQR, 8-12 months). Forty-eight (30%) children had at least 1 positive OFC result confirming the diagnosis of FPIES; the other 112 (70%) subjects presented with a history of typical symptoms after ingestion of the food less than 12 months before the initial visit, and a confirmatory OFC was not performed.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome: case presentations and management lessons.
      The median delay between first reaction and diagnosis was 7 months (range, 0-143 months).
      Table ICharacteristics of subjects with FPIES
      No. of patients160
      Sex, no. (%)
       Male86 (54)
       Female74 (46)
      Age at diagnosis (mo)
       Median15
       IQR
      25% to 75% interquartile range.
      9-24
      Age at most recent follow-up (mo)
       Median45
       IQR
      25% to 75% interquartile range.
      23; 82
      Diagnosis, no. (%)
       Confirmed by positive OFC result48 (30)
       Based on positive clinical history112 (70)
      Incriminated food, no. (%)
       Milk only35 (22)
       Soy only36 (22)
       Both milk and soy21 (23
      Representing the proportion of children with FPIES to both milk and soy among children with milk or soy FPIES.
      )
       Milk and/or soy and solid food(s)18 (11
      Representing the proportion of children with FPIES to both milk and soy among children with milk or soy FPIES.
      )
       Solid food(s) only50 (31)
      FPIES reaction to: no. (%)
       1 Food104 (65)
       >1 Food56 (35)
      ED for the initial reaction, no. (%)
       Milk and/or soy FPIES49 (31)
       Solid food FPIES21 (13)
      Specific IgE positivity, no. (%)
      IgE positivity defined as a positive SPT response (mean wheal ≥3 mm) or detectable serum food-specific IgE (>0.35 kUA/L). Diagnosis of FPIES was based on modified Powell clinical criteria (see text).2,19
       To the FPIES food39 (24)
       To other foods63 (39)
      Additional allergic conditions, no. (%)
       Atopic dermatitis91 (57)
       Allergic rhinitis60 (38)
       Asthma40 (25)
      Birth history
       Cesarean section46 (29)
       Vaginal delivery114 (71)
      Positive family history, no. (%)
       Allergic disease123 (77)
       Food allergy55 (34)
       FPIES10 (6)
      25% to 75% interquartile range.
      Representing the proportion of children with FPIES to both milk and soy among children with milk or soy FPIES.
      IgE positivity defined as a positive SPT response (mean wheal ≥3 mm) or detectable serum food-specific IgE (>0.35 kUA/L). Diagnosis of FPIES was based on modified Powell clinical criteria (see text).
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome: case presentations and management lessons.
      • Powell G.K.
      Food protein-induced enterocolitis of infancy: differential diagnosis and management.
      The majority of subjects had an allergic background, and 39% were sensitized to another food(s) (ie, detectable specific IgE, positive SPT response, or both). Thirty-nine (24%) subjects with FPIES had positive specific IgE to the FPIES food, and 82% of those children had FPIES to milk, soy, or both. Comparison of characteristics of children with milk FPIES and a positive specific IgE to milk with those with milk FPIES but who were not sensitized to milk is shown in Table II. None of the children with IgE sensitization to milk resolved their milk allergy while in the study. Among those who had no IgE sensitization to milk, the median age of milk FPIES resolution was 61 months (P = .003, Fig 2). Detailed information regarding the individual subject's course of IgE positivity is presented in Table E1 in this article's Online Repository at www.jacionline.org.
      Table IIComparison of subjects with milk FPIES sensitized to milk with those not sensitized to milk
      Subjects with milk FPIES, negative specific IgE to milkSubjects with milk FPIES, positive specific IgE to milk
      The detailed clinical and laboratory characteristics of the subjects with IgE sensitization to cow's milk are presented in Table E1. The median cow's milk–specific serum IgE antibody level was 5.14 kUA/L (range, 0.39 to >100 kUA/L; IQR, 0.702-37.125 kUA/L).
      No. of patients, no. (%)53 (76)17 (24)
      Age of onset
       Median3.55
       IQR
      25% to 75% interquartile range.
      0.5-93-7.5
      Age of tolerance
       Median61NA
      None of the children with positive milk-specific IgE outgrew milk FPIES during the study (P = .003).
       IQR
      25% to 75% interquartile range.
      49-173
      Atopic dermatitis, no. (%)24 (45)10 (59)
      Asthma, no. (%)13 (25)5 (29)
      FPIES to >2 foods, no. (%)27 (51)9 (53)
      Duration of breast-feeding
       Median0.30.5
       IQR
      25% to 75% interquartile range.
      0-2.80-2.8
      The detailed clinical and laboratory characteristics of the subjects with IgE sensitization to cow's milk are presented in Table E1. The median cow's milk–specific serum IgE antibody level was 5.14 kUA/L (range, 0.39 to >100 kUA/L; IQR, 0.702-37.125 kUA/L).
      25% to 75% interquartile range.
      None of the children with positive milk-specific IgE outgrew milk FPIES during the study (P = .003).
      Figure thumbnail gr2
      Fig 2Kaplan-Meier survival curves for cow milk, soy, rice, and oat FPIES. Median age at tolerance for milk was 13.82 years (95% CI, 4.67 years; not established), that for soy was 6.67 years (95% CI, 5.00-15.42), that for rice was 4.67 years (95% CI, 3.5-5.92), and that for oat was 4.00 years (95% CI, 2.25-7.92).

       Causative foods

      Cow's milk and soy were the most commonly reported individual foods (n = 70 [44%] and n = 66 [41%], respectively), followed by grains (n = 70 [44%], including rice [n = 36], oat [n = 26], barley [n = 6], and wheat [n = 2]; Fig 1). The majority of subjects reacted to 1 food (n = 104 [65%]), 42 (26%) children reacted to 2 foods, and 14 (9%) children reacted to 3 or more foods (median, 3 foods, range, 3-10 foods). Comparison between single- and multiple-food FPIES is presented in Table E2 in this article's Online Repository at www.jacionline.org. Twenty-six (16%) children reacted to both cow's milk and soy, of whom 5 (19%) reacted also to a solid food. Among the 46 subjects with FPIES to oat, rice, or both, 35% had concomitant FPIES to oat and rice. Most of the children reacting to more than 3 foods reacted to milk, soy, or both and a solid food(s) (n = 9 [64%], Table III). More than half of the subjects with solid food FPIES reacted to 2 or more foods.
      Table IIIRates of concomitant FPIES to multiple foods
      If FPIES to:MilkSoySolid
       MilkNA38%20%
       Soy37%NA14%
       Solid
      More than 50% of the subjects with solid FPIES reacted to 2 or more foods. Forty-four percent of subjects with FPIES to rice reacted to oat.
      20%13%44%
      More than 50% of the subjects with solid FPIES reacted to 2 or more foods. Forty-four percent of subjects with FPIES to rice reacted to oat.
      NA, Not applicable.
      More than 50% of the subjects with solid FPIES reacted to 2 or more foods. Forty-four percent of subjects with FPIES to rice reacted to oat.
      We compared 50 children with solid food FPIES with 92 children with milk FPIES, soy FPIES, or both observed over the same time period (Table IV). The median age at the onset of solid food FPIES was significantly greater (P = .001) and exclusive breast-feeding was significantly longer lasting (P = .0001) compared with the milk FPIES, soy FPIES, or both group. Three subjects had chronic FPIES while being exclusively breast-fed. One hundred four (65%) of 160 subjects in our study were fed a hypoallergenic formula. Among 104 subjects fed a hypoallergenic formula, 64 (61.5%) were fed an extensively hydrolyzed casein-based formula (Nutramigen [Mead Johnson, Glenview, Ill] or Alimentum [Abbott Nutrition, Columbus, Ohio]), and 40 (38.5%) were fed an amino acid–based formula. Children with milk FPIES, soy FPIES, or both were exposed to an infant formula significantly earlier than children with solid FPIES (P < .001). Fifty-seven (62%) of 92 infants who had FPIES to milk, soy, or milk and soy were exposed to only milk, soy, or both milk and soy formulas, respectively, in the first few weeks of life. The majority (47/75 [63%]) of subjects with milk FPIES, soy FPIES, or both of those for whom early feeding history was available, reported chronic symptoms (diarrhea, colitis, reflux, and/or failure to thrive) in early infancy, occurring shortly after the introduction of milk-based formula, soy-based formula, or both. Removal of milk, soy, or both from the diet resolved chronic symptoms, but subsequently, these subjects had acute episodes with typical FPIES symptoms after ingestion of milk, soy, or both (mean age, 7 months; range, 0.03-60 months). Among the 26 subjects with FPIES to both milk and soy, the majority (61.5%) had symptoms to both foods within a period of less than 2 months. Nine subjects had FPIES to soy several months after onset of FPIES to milk (median, 6 months; IQR, 2.6-6.5 months).
      Table IVComparison between milk FPIES, soy FPIES, or both and solid food(s) FPIES
      Food involvedM/SSolid food(s) onlyM/S and solid food(s)P value
      P value for comparison between solid food FPIES and cow's milk/soy FPIES. The 18 patients having concomitant FPIES to solid food and FPIES to milk, soy, or both were excluded from this analysis.
      No. of patients (%)92 (58)50 (31)18 (11)
      Age at diagnosis (mo)
       Median151312.5.9
       IQR
      25% to 75% interquartile range.
      9-249-337-21.3
      Age at onset of FPIES (mo)
       Median574.0007
       IQR
      25% to 75% interquartile range.
      2-106-122-6
      Sex, no. (%)
       Male54 (59)26 (52)6 (33).45
       Female38 (41)24 (48)12 (67)
      Diagnosis based on: no. (%)
       Positive history59 (64)45 (90)8 (44).005
       Positive OFC result33 (36)5 (10)10 (56)
      Reaction to: no. (%)
       1 Food71 (77)33 (66)0 (0).99
       >1 Food21 (23)17 (34)18 (100)
      Age at formula introduction (mo)
       Median0.031.50.2.0002
       IQR
      25% to 75% interquartile range.
      0.03-1.50.05-9.50.03-1.3
      Age at solid food introduction (mo)
       Median5.36.05.17
       IQR
      25% to 75% interquartile range.
      4-65-63.3-6
      Duration of exclusive breast-feeding (mo)
       Median0.033.80.4.0001
       IQR
      25% to 75% interquartile range.
      0.0-2.10.1-60-4.3
      Past history, no. (%)
       Cesarean section30 (33)12 (24)4 (22).27
       Eczema50 (54)29 (58)12 (67).58
       Rhinitis38 (41)16 (32)6 (33).32
       Asthma22 (24)12 (24)6 (33).94
      Positive family history, no. (%)
       FPIES10 (11)0 (0)0 (0).1
       Food allergy36 (39)14 (28)5 (28).2
       Atopy70 (76)39 (78)5 (28).5
      M/S, Cow's milk, soy, or both.
      P value for comparison between solid food FPIES and cow's milk/soy FPIES. The 18 patients having concomitant FPIES to solid food and FPIES to milk, soy, or both were excluded from this analysis.
      25% to 75% interquartile range.

       OFCs and outcomes

      Subjects' characteristics and symptoms observed during the OFC are presented in Table V. One hundred eighty standardized OFCs were performed in the Mount Sinai Clinical Research Unit to 15 different foods in 82 subjects. Seventy-four challenge results were positive in 47 subjects; 106 (59%) were negative, indicating tolerance to the incriminated food (see Table E3 in this article's Online Repository at www.jacionline.org). One hundred thirty-four (75%) challenges were performed in subjects with a prior history of FPIES reaction to the index food; 46 (25%) challenges were performed to foods that had been avoided as a precaution. Prior reactions to incriminated foods were reported in 93% of subjects with positive OFC results and in 60% of subjects with negative OFC results (P < .001). The challenge foods resulted in a positive OFC result in 58% of OFCs for milk (total number of OFCs, n = 72), 45% for soy (n = 44), 21% for grains (n = 48), 20% for egg (n = 5), and 33% for beef (n = 3) and in none of the challenges for chicken (n = 3), fish (n = 1), sweet potatoes (n = 1), apple (n = 1), pear (n = 1), and banana (n = 1). There were no significant differences regarding OFC outcome by food.
      Table VCharacteristics of OFCs in subjects with FPIES
      Food involved (no.)Milk (n = 72), soy (n = 44), grains
      Grains include oat (n = 14), rice (n = 13), corn (n = 3), wheat (n = 10), and barley (n = 8).
      (n = 48), others
      Other foods include egg (n = 5), beef (n = 3), chicken (n = 3), fish (n = 1), apple (n = 1), pear (n = 1), sweet potatoes (n = 1), and banana (n = 1).
      (n = 16)
      No. of positive OFC results/no. of total OFCs (%)74/180 (41)
      No. of patients82
      Age at OFC (y)
       Median2.8
       IQR
      25% to 75% interquartile range.
      ,
      Range: 0.8-30.3.
      2-5
      Sex, no. (%)
       Male41 (50)
       Female41 (50)
      Symptoms during the positive OFC, no. (%)
       Vomiting70 (96)
       Abdominal pain59 (80)
       Hypotension14 (19)
       Diarrhea5 (7)
       Altered state of consciousness/lethargy5 (7)
      Timing of the reaction in positive OFC (min [range])
       From the first dose150 (35-370)
       From the last dose120 (5-320)
       To complete recovery50 (0-460)
      Treatment of positive OFC, no. (%)
       Intravenous normal saline bolus70 (96)
       Intravenous corticosteroids69 (94)
       None3 (4)
       Hospitalization0 (0)
      No. of reactions before OFC
       Median2
       Range0-multiple
      Timing from the most recent reaction (mo)
       Median13
       IQR
      25% to 75% interquartile range.
      7-53
      There were no statistically significant differences between the characteristics of the OFCs to milk, soy, grains, and other foods.
      Grains include oat (n = 14), rice (n = 13), corn (n = 3), wheat (n = 10), and barley (n = 8).
      Other foods include egg (n = 5), beef (n = 3), chicken (n = 3), fish (n = 1), apple (n = 1), pear (n = 1), sweet potatoes (n = 1), and banana (n = 1).
      25% to 75% interquartile range.
      § Range: 0.8-30.3.
      The majority of severe reactions associated with hypotension occurred after an OFC to milk or soy (12/14 [86%]); 2 occurred after ingestion of oat and wheat (see Table E4 in this article's Online Repository at www.jacionline.org). The severity of the reaction during an OFC did not correlate with the reported severity of the initial reaction to the incriminated food. Fifteen (20%) subjects with a positive OFC result reported having ingested a small quantity of the incriminated food without reaction. In the majority (83%) of the OFCs, the symptoms started 120 minutes or more after the first dose (maximum of 370 minutes); in the remainder (17%) of the OFCs, symptoms started as soon as 35 minutes after the first dose. Among those children with an onset of symptoms sooner than 120 minutes after food ingestion, 40% had detectable food-specific IgE levels. The median time from the onset of symptoms to complete recovery was 50 minutes (IQR, 13-95). Because of the potential risk of severe reaction, these OFCs have to be performed under medical supervision, independently of the food involved.
      The absolute neutrophil count (ANC) measured before and 5 to 8 hours after the positive challenge result showed a median increase of 1850 cells/mm3, with an average of 3228 cells/mm3 (IQR, 825-4200 cells/mm3). The absolute lymphocyte and basophil counts were not changed significantly after the OFC, whereas the eosinophil peripheral count was significantly decreased, with an average of −60 cell/mm3 (range, −50 to 300 cells/mm3; P < .05). There was a mild increase in the platelet count (median, 19,000/mm3; range, −121,000 to 110,000/mm3).

       Resolution of FPIES

      We followed the 160 subjects for a median of 23 months (IQR, 3-56 months). Among the 60 subjects with 3 years or more of follow-up, 34 (57%) recovered at least 1 food during the study period. Twenty-three (68%) patients had a negative OFC result that confirmed tolerance, whereas the other 11 patients reported introducing the food at home. The cumulative probability of recovery was calculated for each food and presented in Fig 2. The median age when tolerance was documented either by an OFC or by a parental report of food reintroduction at home was 4.7 years for rice, 4 years for oat, and 6.7 years for soy. The overall median age for milk tolerance was 13.8 years; the median age of milk FPIES resolution for subjects with undetectable milk-specific IgE was 5.1 years, whereas none of the subjects with detectable milk-specific IgE became tolerant to milk in the study (P = .003, Fig 2). Five subjects remained intolerant after 16 years of age (milk, n = 3; soy, n = 2); all had serial OFCs performed approximately every 2 years, documenting their FPIES reactions at these ages. Subjects with persistent milk FPIES after age 3 years had a higher (46%) proportion of positive specific IgE measurements to milk compared with none (0%) of the children with milk FPIES resolved by age 3 years (P < .05, Table VI). Among the 17 subjects with milk FPIES and detectable milk-specific IgE, 7 (41%) subjects had symptoms of IgE-mediated food allergy at follow-up (see Table E1). Two subjects had multisystem reactions suggestive of anaphylaxis. Of note, 5 subjects underwent a supervised OFC to milk confirming an IgE-mediated allergy. The remaining children were not offered the challenge because of the recent history of unintentional ingestion with reactions or because of the cow's milk–specific IgE antibody levels exceeding the 95% decision point for clinical reactivity.
      Table VIComparison of children with FPIES to milk that resolved by age 3 years and those with persistent milk FPIES
      Resolved by age 3 yPersistent after age 3 yNo follow-up after age 3 yP value
      Comparison of FPIES resolved by age 3 years and subject with persistent FPIES after age 3 years.
      No. of patients (%)11 (16)37 (53)22 (31)
      Age (mo [at resolution])
       Median268019.01
       IQR
      25% to 75% interquartile range.
      23-3256-11913-26
      Sex, no. (%)
       Female5 (45)19 (51)13 (59).75
       Male6 (55)18 (49)9 (41)
      Age at first symptoms (mo)
       Median0.823.34
       IQR
      25% to 75% interquartile range.
      0.2-60.5-60.5-5
      History of early exposure to milk, no. (%)7 (64)29 (78)18 (82).43
      Diagnosis, no. (%)
       Confirmed by an OFC1 (9)26 (70)6 (27).0001
       Based on positive history10 (91)11 (30)16 (73)
      Detectable milk-specific IgE, no. (%)
      Positive serum IgE or SPT responses0 (0)11 (30)6 (27).04
      FPIES to other foods, no. (%)
       Soy5 (45)14 (38)8 (36).80
       Grains2 (18)4 (11)3 (14).53
       Beef0 (0)2 (5)4 (18).75
      History, no. (%)
       IgE-mediated allergy to other foods3 (27)12 (32)7 (32).60
       AD6 (55)18 (49)10 (46).51
       Asthma5 (45)9 (24)4 (18).15
      AD, Atopic dermatitis.
      Comparison of FPIES resolved by age 3 years and subject with persistent FPIES after age 3 years.
      25% to 75% interquartile range.
      In addition to cow's milk, we found that 16 subjects with FPIES to soy, 3 subjects with FPIES to rice, 2 subjects with FPIES to oat, and 1 patient with FPIES to egg had specific IgE to the offending food.

      Discussion

      We report data from a large cohort of subjects with FPIES and the largest number of OFCs performed for FPIES to date. In this cohort 24% had detectable specific IgE antibodies to the food that triggered the FPIES, and 39% had concomitant IgE sensitization to other foods; to our knowledge, this latter observation has not been reported previously. Among children with specific IgE to cow's milk, 41% changed from the milk FPIES phenotype to an IgE-mediated phenotype over time. We confirmed that the presence of specific IgE to cow's milk is a risk factor for persistence of milk FPIES beyond 3 years of age. Although IgE antibodies to the causal food are typically not detected in subjects with FPIES, there are reports in which children had detectable IgE to the causal protein either at presentation or during follow-up, which is referred to as “atypical” FPIES.
      • Leonard S.A.
      • Nowak-Wegrzyn A.
      Food protein-induced enterocolitis syndrome: an update on natural history and review of management.
      In our study 24% of subjects had positive specific IgE measurements to the FPIES-induced food(s). It has been suggested that acute FPIES reactions are associated with TH2 skewing of the T-cell cytokine profile.
      • Mori F.
      • Barni S.
      • Cianferoni A.
      • Pucci N.
      • de Martino M.
      • Novembre E.
      Cytokine expression in CD3+ cells in an infant with food protein-induced enterocolitis syndrome (FPIES): case report.
      One can speculate that a TH2-skewed lymphocyte response might predispose to developing specific IgE to food, which can eventually lead to a classic IgE-mediated reaction, as we have documented during the study in 41% of subjects with positive milk-specific IgE levels. Of note, Onesimo et al
      • Onesimo R.
      • Dello Iacono I.
      • Giorgio V.
      • Limongelli M.G.
      • Miceli Sopo S.
      Can food protein induced enterocolitis syndrome shift to immediate gastrointestinal hypersensitivity? A report of two cases.
      described 2 cases of atypical FPIES that turned into IgE-mediated gastrointestinal anaphylaxis.
      The inflammation in the intestinal tract caused by FPIES might enhance the penetrability of food proteins and their presentation to the immune system. Sicherer et al
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome: case presentations and management lessons.
      suggested that the presence of specific IgE to the involved antigen in subjects with FPIES, which is usually negative, predicted a prolonged sensitivity to the antigen. We found that none of the children with positive milk-specific IgE outgrew milk FPIES during the study. Although this observation is likely confounded by the delayed attempts at introduction because of the awareness of the role of specific IgE in prolonged FPIES, it is nevertheless intriguing and worth further investigation. Because subjects initially presenting with or later having food-specific IgE antibodies might be at risk for more persistent FPIES and/or progression to an IgE-mediated food allergy phenotype (observed in approximately 33% in this study), SPTs, measurement of serum food-specific IgE levels, or both in the initial and follow-up evaluations is useful. In children with positive specific IgE levels, it is particularly important to adapt the OFC protocol and carry out the OFC under conditions in which an anaphylactic reaction can be properly treated. Another previously unrecognized finding is the relatively high (39%) prevalence of IgE sensitization to foods and clinical expression of IgE-mediated food allergy (30% of the children with IgE sensitization) among children with FPIES. This finding emphasizes the utility of allergy testing for the most common food allergens before introduction into the diet of children with FPIES.
      In our clinical practice an initial diagnosis of FPIES is based on clinical criteria.
      • Powell G.K.
      Food protein-induced enterocolitis of infancy: differential diagnosis and management.
      OFCs are performed to assess whether tolerance to the offending food has developed. Because of the risk of a severe reaction (particularly hypotension), the foods that have caused FPIES reactions in the past should be introduced under medical supervision, regardless of the food involved. Negative OFC results in subjects with infantile FPIES reduce a parent's inconvenience and the social isolation resulting from strict dietary management. There is debate as to when follow-up OFCs are appropriate. In this study 160 subjects with FPIES were followed for a median 23 months. Sixty-four (40%) subjects achieved tolerance to at least 1 food. According to the results of the Kaplan-Meier curves, cumulative tolerance probability was 50% at 6 years of age. However, resolution of FPIES appears to be population dependent, particularly for cow's milk and soy.
      • Sicherer S.H.
      • Eigenmann P.A.
      • Sampson H.A.
      Clinical features of food protein-induced enterocolitis syndrome.
      • Nowak-Wegrzyn A.
      • Sampson H.A.
      • Wood R.A.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome caused by solid food proteins.
      • Katz Y.
      • Goldberg M.R.
      • Rajuan N.
      • Cohen A.
      • Leshno M.
      The prevalence and natural course of food protein-induced enterocolitis syndrome to cow's milk: a large-scale, prospective population-based study.
      • Onesimo R.
      • Dello Iacono I.
      • Giorgio V.
      • Limongelli M.G.
      • Miceli Sopo S.
      Can food protein induced enterocolitis syndrome shift to immediate gastrointestinal hypersensitivity? A report of two cases.
      Although we found that cow's milk FPIES resolved in 20% by 3 years of age, a Korean cohort showed more than 60% resolution by 10 months of age and an Israeli birth cohort showed 90% resolution by 30 months.
      • Sicherer S.H.
      • Eigenmann P.A.
      • Sampson H.A.
      Clinical features of food protein-induced enterocolitis syndrome.
      • Nowak-Wegrzyn A.
      • Sampson H.A.
      • Wood R.A.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome caused by solid food proteins.
      • Katz Y.
      • Goldberg M.R.
      • Rajuan N.
      • Cohen A.
      • Leshno M.
      The prevalence and natural course of food protein-induced enterocolitis syndrome to cow's milk: a large-scale, prospective population-based study.
      • Hwang J.B.
      • Sohn S.M.
      • Kim A.S.
      Prospective follow-up oral food challenge in food protein-induced enterocolitis syndrome.
      Similarly, we found that 20% of cases of soy FPIES resolved by 3 years of age, whereas in the Korean cohort more than 90% of children showed resolution of soy FPIES by 10 months of age.
      • Hwang J.B.
      • Sohn S.M.
      • Kim A.S.
      Prospective follow-up oral food challenge in food protein-induced enterocolitis syndrome.
      These differences are likely explained by the higher proportion of subjects with detectable food-specific IgE levels and atopic dermatitis among the subjects who were referred to a major allergy center and the differences in methodology compared with the Israeli and Korean studies.
      • Katz Y.
      • Goldberg M.R.
      • Rajuan N.
      • Cohen A.
      • Leshno M.
      The prevalence and natural course of food protein-induced enterocolitis syndrome to cow's milk: a large-scale, prospective population-based study.
      • Hwang J.B.
      • Sohn S.M.
      • Kim A.S.
      Prospective follow-up oral food challenge in food protein-induced enterocolitis syndrome.
      The data for resolution of solid food FPIES are scarce. In our studies resolution by 5 years of age occurred in 65.5% of children reactive to grains, whereas resolution of FPIES to meat and fish/shellfish took more time (50% and 0% at 5 years of age, respectively). Forty-six OFCs were performed to assess tolerance to foods avoided for precaution; results of 11% of these OFC were positive, highlighting the importance of reintroducing these foods under medical supervision.
      Several studies examined the relative prevalence of causative foods in subjects with FPIES.
      • Nowak-Wegrzyn A.
      • Sampson H.A.
      • Wood R.A.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome caused by solid food proteins.
      • Ruffner M.A.
      • Ruymann K.
      • Barni S.
      • Cianferoni A.
      • Brown-Whitehorn T.
      • Spergel J.M.
      Food Protein-induced enterocolitis syndrome: insights from review of a large referral population.
      • Sopo S.M.
      • Giorgio V.
      • Dello Iacono I.
      • Novembre E.
      • Mori F.
      • Onesimo R.
      A multicentre retrospective study of 66 Italian children with food protein-induced enterocolitis syndrome: different management for different phenotypes.
      • Katz Y.
      • Goldberg M.R.
      • Rajuan N.
      • Cohen A.
      • Leshno M.
      The prevalence and natural course of food protein-induced enterocolitis syndrome to cow's milk: a large-scale, prospective population-based study.
      • Mehr S.
      • Kakakios A.
      • Frith K.
      • Kemp A.S.
      Food protein-induced enterocolitis syndrome: 16-year experience.
      In our series milk and soy were the single most common foods inducing FPIES, followed by rice, which is similar to a report from another large US center.
      • Ruffner M.A.
      • Ruymann K.
      • Barni S.
      • Cianferoni A.
      • Brown-Whitehorn T.
      • Spergel J.M.
      Food Protein-induced enterocolitis syndrome: insights from review of a large referral population.
      In contrast, in Australia, Israel, and Italy soy is not frequently reported. The majority reacted to a single food, whereas approximately one third reacted to multiple foods (n = 56 [35%]). The majority of the subjects with milk FPIES, soy FPIES, or both had chronic symptoms (diarrhea, colitis, reflux, and/or failure to thrive) in early infancy, occurring shortly after the introduction of milk-based formula, soy-based formula, or both. Chronic symptoms resolved on avoidance of milk/soy, but subsequent acute episodes with typical FPIES symptoms occurred after ingestion of milk/soy. These data are in line with the concept of acute-on-chronic FPIES previously described in infants.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome: case presentations and management lessons.
      Although higher rates (up to 60%) of concomitant FPIES to both milk and soy have been reported, in our series the risk of soy FPIES in a child with milk FPIES was 38%, and the risk of milk FPIES in a child with soy FPIES was 37%. Given the above and the recent studies documenting low rates of soy FPIES in infants with milk FPIES,
      • Ruffner M.A.
      • Ruymann K.
      • Barni S.
      • Cianferoni A.
      • Brown-Whitehorn T.
      • Spergel J.M.
      Food Protein-induced enterocolitis syndrome: insights from review of a large referral population.
      • Sopo S.M.
      • Giorgio V.
      • Dello Iacono I.
      • Novembre E.
      • Mori F.
      • Onesimo R.
      A multicentre retrospective study of 66 Italian children with food protein-induced enterocolitis syndrome: different management for different phenotypes.
      • Katz Y.
      • Goldberg M.R.
      • Rajuan N.
      • Cohen A.
      • Leshno M.
      The prevalence and natural course of food protein-induced enterocolitis syndrome to cow's milk: a large-scale, prospective population-based study.
      soy formula might represent an acceptable alternative in subjects with milk FPIES, although supervised introduction is prudent. However, in subjects with FPIES to milk or soy within the first 1 month of life, it is prudent to delay soy or milk introduction until age 12 months.
      • Fleischer D.M.S.J.
      • Assa'ad A.H.
      • Pongracic J.A.
      Primary prevention of allergic disease through nutritional interventions.
      For these patients, exclusive breast-feeding should be encouraged. Alternatively, either a casein hydrolysate–based formula or an amino acid formula are recommended. These formulas lead to resolution of symptoms in all our subjects with milk FPIES, soy FPIES, or both. The mean age at onset of solid food FPIES tends to be older than that of milk FPIES, soy FPIES, or both, reflecting the usual sequence of food introduction, whereas the age of first solid food introduction was not different. Our data regarding the prevalence of reactions to foods from similar food groups are likely confounded by the fact that after the onset of FPIES to 1 food, subsequent introduction of other foods was usually delayed. Nevertheless, infants with solid food FPIES were likely to react to other foods: 52% reacted to more than 1 food protein, and 26% reacted to milk, soy, or both. In those with cereal-induced FPIES, sensitivity to other cereal grains was evident in nearly half of the patients. In particular, we found that 44% of subjects with FPIES to rice reacted to oat, likely reflecting introduction of these foods usually at the same age. On the basis of these results, infants with cereal FPIES appear particularly vulnerable and might benefit from delayed introduction of grains beyond the first year of life. It remains to be determined whether delayed introduction of other foods with high protein content, such as legumes and poultry, is indicated. This might avoid sensitization/reactions to other foods during a possible period of developmental susceptibility. However, confirming previous reports,
      • Fernandes B.N.
      • Boyle R.J.
      • Gore C.
      • Simpson A.
      • Custovic A.
      Food protein-induced enterocolitis syndrome can occur in adults.
      onset after early infancy can occur with some foods, such as fish or shellfish (n = 11; median age, 30; range, 5-415).
      Subjects with milk/soy FPIES were breast-fed for significantly shorter periods than subjects reacting to solids. This suggests that an early exposure to formula predisposes to milk/soy FPIES (Table IV). Milk/soy FPIES in exclusively breast-fed infants is extremely rare, suggesting an important protective role of breast-feeding.
      • Monti G.
      • Castagno E.
      • Liguori S.A.
      • Lupica M.M.
      • Tarasco V.
      • Viola S.
      • et al.
      Food protein-induced enterocolitis syndrome by cow's milk proteins passed through breast milk.
      • Nomura I.
      • Morita H.
      • Hosokawa S.
      • Hoshina H.
      • Fukuie T.
      • Watanabe M.
      • et al.
      Four distinct subtypes of non-IgE-mediated gastrointestinal food allergies in neonates and infants, distinguished by their initial symptoms.
      In our cohort no patient had acute FPIES while being exclusively breast-fed; however, 3 (2%) infants had chronic FPIES symptoms to milk in the maternal diet. Breast milk might provide protection (eg, IgA antibodies and partially processed food proteins) or the threshold dose of allergen might not be reached in breast milk to trigger a full acute FPIES phenotype.
      • Lake A.M.
      Food-induced eosinophilic proctocolitis.
      Diagnosis of FPIES is fraught with difficulties. A delayed diagnosis is common and might be due to the lack of definitive diagnostic tests and the unusual nature of symptoms that lack the “classic allergic” skin and respiratory involvement. Low index of suspicion in case of grains and vegetables that are widely considered to be hypoallergenic might also contribute to the delay in diagnosis. FPIES diagnosis is based on the characteristic history and/or an in-hospital OFC performed with full resuscitation facilities at hand, particularly for rapid intravascular volume repletion. Diagnostic criteria have been proposed for milk FPIES
      • Powell G.K.
      Food protein-induced enterocolitis of infancy: differential diagnosis and management.
      and have not been validated in all types of FPIES. Our data suggest that these criteria could be modified to improve recognition of this potentially severe disease and not include an age limit for either age of onset or age for OFC. The vast majority of the subjects experienced repetitive vomiting (96%) and abdominal pain (80%), but diarrhea was observed in only 7% during the food challenge. These results are similar to other descriptions of acute FPIES-related reactions in which diarrhea were not as prevalent, which is in contrast to chronically exposed children.
      • Nowak-Wegrzyn A.
      • Sampson H.A.
      • Wood R.A.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome caused by solid food proteins.
      • Sopo S.M.
      • Giorgio V.
      • Dello Iacono I.
      • Novembre E.
      • Mori F.
      • Onesimo R.
      A multicentre retrospective study of 66 Italian children with food protein-induced enterocolitis syndrome: different management for different phenotypes.
      • Lake A.M.
      Food-induced eosinophilic proctocolitis.
      Other symptoms included hypotension (19%) and lethargy (7%). Regarding laboratory tests, the Powell criteria for a positive OFC include an increase in the ANC of greater than 3500 cells/mm3.
      • Sicherer S.H.
      Food protein-induced enterocolitis syndrome: case presentations and management lessons.
      • Powell G.K.
      Food protein-induced enterocolitis of infancy: differential diagnosis and management.
      Although an increase in ANCs was observed in the vast majority of subjects with a positive OFC result, the mean increase was less than 3500 cells/mm3 (ie, 3228 cells/mm3). In addition, our data are likely confounded by treatment with intravenous steroids, which increases the peripheral blood neutrophil count. Considering OFCs were done within a minimum of 12 months of the most recent reaction to the offending food and thus excluded the subjects with the most severe phenotype, revising the criteria to decrease the required ANC increase as a major criterion could be considered. Of note, an isolated ANC shift without associated symptoms is not sufficient to consider the OFC result positive according to the Powell criteria and should not lead to avoidance of the food. Hypotension that was observed in our population could be added as a minor criterion. Although the limitations of our study include a mixed prospective and retrospective method of patient ascertainment, single-center experience, limited follow-up, and absence of OFC in a subgroup of patients, our observations raise clinically important questions and warrant validation by other investigators.
      In conclusion, FPIES is an underrecognized non–IgE-mediated food allergy that most commonly affects infants and is usually caused by cow's milk, soy, rice, and oat. Early recognition of the symptoms of FPIES and removal of the offending food are imperative to prevent misdiagnosis and mismanagement of symptoms that might mimic viral illness or sepsis and lead to failure to thrive when food is chronically present in the diet. Our data suggest that introduction of milk formula, soy formula, or both within the first weeks of life is an important risk factor for the development of FPIES to milk, soy, or both. The diagnosis of FPIES is based on clinical criteria that should not include an age restriction because FPIES can occur at any age. Although a physician-supervised OFC remains the gold standard for diagnosis of FPIES, diagnostic criteria for a positive OFC result might need to be modified to remove diarrhea and stool studies as major positive criteria because diarrhea is relatively uncommon during an OFC in older children. Close follow-up is required to determine when foods might be added back into the diet. Although solid food FPIES resolves in most subjects by 5 years of age, subjects with milk and soy FPIES, as well as those with detectable food-specific IgE, might have a more protracted course. Further studies are needed to determine the pathophysiology and biomarkers of FPIES, as well as the features of natural history that are unique for specific foods.
      Clinical implications
      FPIES typically resolves by age 6 years, but milk FPIES, especially with detectable food-specific IgE, can have a protracted course and eventually transition to acute reactions/anaphylaxis.
      We thank Erin Moshier, MS, for expert statistical analysis.

      Appendix

      Table E1Characteristics of subjects with milk FPIES and detectable milk-specific IgE
      Age at initial FPIES reaction
      All of these 17 patients have a history of classic FPIES reactions to milk.
      to milk (mo)
      Initial
      Earliest available test results of SPT and/or milk-specific IgE.
      milk-specific IgE (kUA/L); age
      Initial
      Earliest available test results of SPT and/or milk-specific IgE.
      milk SPT (mm) at diagnosis; age
      Follow-up (mo)IgE-mediated clinical milk allergy; ageMilk-specific IgE (kUA/L) at follow-up; ageMilk SPT (mm) at follow-up; ageAtopic diseasesFPIES to other food(s)
      612; 12 moPositive; 17 mo64Sneezing and cough, hives; 21 mo33.2; 36 moNot doneA, AD, AR, beef, lamb IgE allergySoy
      6<0.35; 12 mo0/0; 18 mo83After 25% of OFC serving: nasal congestion, emesis, Benadryl, Solu-Medrol, ranitidine, epinephrine; 84 mo0.73; 81 mo4/26; 60 moADNo
      8Not done0/0; 16 mo91Acute throat symptoms, hives; 48 mo0.95; 11 y16/28; 11 yA, AR, ADNo
      <121.08; 36 mo0/0; 9 mo72Acute urticaria; 62 mo>100; 5 y, 25; 7 yNot doneEgg IgE allergyNo
      126.01; 48 mo7/24; 72 mo56Skin rash to contact with milk and immediate (30 min after food ingestion) abdominal pain; 60 mo12.3; 8 y4/12; 8 yAD, AR, pork IgE allergy, penicillin allergyNo
      3Not availableNot available35Immediate emesis and skin rash; 5 moNot done9/17; 32 moARRice, oat
      326.8; 48 mo50Wheezing, emesis; 30 mo9.7; 60 mo22/40; 72 moAD, AR, peanut IgE allergySoy
      4Increased; 8 moLarge; 8 mo0No5.14; 11 moNot doneNoSoy
      0.50.93; 11 moNot done30No0.51; 48 mo0/0; 48 moAD, A; positive sIgE to peanut, cashew, pistachio, egg whiteSoy
      6<0.35; 18 mo0/0; 18 mo86NoNot done6/24ADNo
      70/0; 7 moNot done20No0.91; 28 mo0/0; 23 moNoNo
      4<0.35 to milk and casein; 13 mo7/20; 17 mo14NoNot done0/0; 32 moADNo
      0.037.97; 60 mo12/32; 60 mo42No48.9; 84 moNot doneA, AR, AD; positive sIgE to egg, soybean, peanut, sesameSoy
      42.95; 12.5 moNot done0No0.62; 27 mo0/0; 27 moADSoy
      1Casein 0.55; 7 mo8/30; 11 mo5NoNot doneNot doneAD; positive sIgE to peanut, egg, and soyNo
      12
      This patient had reactions to broccoli, rice, chicken, and spinach at 8 months of age.
      <0.35; 10 mo0/0; 10 mo4No73.7; 16 mo0/0; 16 moPositive sIgE to barley, beef, chickpea, egg, wheatSalmon, broccoli, spinach, chicken
      5Not done0/077No0.39; 10 mo0/0 at multiple time pointsA, AR, ADSoy
      Serum food-specific IgE antibody levels were measured with the CAP System FEIA (Thermo Fisher, Portage, Mich); the lower limit of detection was 0.35 kUA/L, and the upper limit of detection was 100 kUA/L. SPTs were done as described by Leonard and Nowak-Wegrzyn.
      • Leonard S.A.
      • Nowak-Wegrzyn A.
      Food protein-induced enterocolitis syndrome: an update on natural history and review of management.
      The median cow's milk–specific serum IgE antibody level was 5.14 kUA/L (range, 0.39->100 kUA/L; IQR, 0.702-37.125 kUA/L). None of the children with positive milk-specific IgE levels outgrew milk FPIES during the study. Five subjects underwent supervised oral milk challenges with reactions: 3 of them had 2 positive challenges, 2 had 1 positive challenge, and 1 had 3 positive challenges, 2 with classic FPIES symptoms and the most recent challenge with hives. The remaining children were not offered the challenge because of a recent history of unintentional ingestion with reactions or because of cow's milk–specific IgE antibody levels exceeding the 95% decision point for clinical reactivity.
      AD, Atopic dermatitis; AR, allergic rhinitis; A, asthma; sIgE, specific IgE.
      All of these 17 patients have a history of classic FPIES reactions to milk.
      Earliest available test results of SPT and/or milk-specific IgE.
      This patient had reactions to broccoli, rice, chicken, and spinach at 8 months of age.
      Table E2Comparison of subjects with FPIES to single and multiple foods
      No. of food(s) involvedFPIES to 1 food (n = 104)FPIES to 2 foods (n = 42)FPIES to ≥3 foods (n = 14)
      Age at diagnosis (mo)
       Median161311
       IQR
      25% to 75% interquartile range.
      9-288-248-16
      Age at onset (mo)
       Median754
       IQR
      25% to 75% interquartile range.
      4-120.73-6
      Detectable sIgE to the food causing FPIES, no. (%)26 (25)11 (26)2 (14)
      Detectable sIgE to other food(s), no. (%)47 (45)12 (29)4 (3)
      Positive familial history for atopic disease, no. (%)80 (77)30 (71)13 (93)
      Positive familial history for FPIES, no. (%)6 (6)3 (7)0 (0)
      There were no statistically significant differences between those with single-food and those with multiple-food FPIES.
      25% to 75% interquartile range.
      Table E3Comparison of failed and passed OFCs
      Failed OFCPassed OFCP value
      No. of challenges performed, no. (%)74 (41)106 (59)
      Prior reactions to the challenge food, no. (%)69 (93)65 (61).0001
      No. of patients challenged
      Of note, some patients underwent multiple OFCs to the same or different foods; some patients had both positive and negative OFC results.
      47 (57)54 (66)
      Age (y)
       Median3.52.3.001
       IQR
      25% to 75% interquartile range.
      2.3-8.71.8-4.3
      Sex, no. (%)
       Male25 (53)26 (48).1
       Female22 (47)28 (52)
      Incriminated food, no. (%)
       Milk42 (57)30 (28).1
       Soy20 (27)24 (23).1
       Grains
      Grains include oat, rice, wheat, and corn.
      10 (13)38 (36)
       Others
      Other foods include egg, chicken, beef, apple, and pear.
      2 (3)14 (13)
      Time from the last FPIES reaction (mo)
       Median2723.3
       IQR
      25% to 75% interquartile range.
      20-4520-45
      Of note, some patients underwent multiple OFCs to the same or different foods; some patients had both positive and negative OFC results.
      25% to 75% interquartile range.
      Grains include oat, rice, wheat, and corn.
      § Other foods include egg, chicken, beef, apple, and pear.
      Table E4Comparison of patients who had documented hypotension during the OFCs with children who did not have hypotension
      Severe reaction during OFC (n = 14)Nonsevere reaction during OFC (n = 60)
      Age (mo)
       Median54.540.4
       IQR
      25% to 75% interquartile range.
      18-9628-105
      Sex, no. (%)
       Male8 (57)33 (55)
       Female6 (43)27 (45)
      Food, no. (%)
       Milk9 (64)33 (55)
       Soy3 (21)17 (28)
       Grains (oat or wheat)2 (14)8 (13)
       Other (beef or egg)0 (0)2 (3)
      Positive history of hospitalization because of FPIES no. (%)6 (43)23 (38)
      Detectable sIgE to the food causing FPIES, no. (%)3 (21)17 (28)
      Detectable sIgE to other food(s), no. (%)1 (7)12 (20)
      There were no statistically significant differences between the characteristics of the subjects with a severe reaction and those with a mild reaction.
      25% to 75% interquartile range.

      References

        • Boyce J.A.
        • Assa'ad A.
        • Burks A.W.
        • Jones S.M.
        • Sampson H.A.
        • Wood R.A.
        • et al.
        Guidelines for the diagnosis and management of food allergy in the United States: report of the NIAID-sponsored expert panel.
        J Allergy Clin Immunol. 2010; 126: S1-58
        • Sicherer S.H.
        Food protein-induced enterocolitis syndrome: case presentations and management lessons.
        J Allergy Clin Immunol. 2005; 115: 149-156
        • Järvinen K.M.
        • Nowak-Wegrzyn A.
        Food protein-induced enterocolitis syndrome (FPIES): current management strategies and review of the literature.
        J Allergy Clin Immunol Pract. 2013; 1: 317-322
        • Sicherer S.H.
        • Eigenmann P.A.
        • Sampson H.A.
        Clinical features of food protein-induced enterocolitis syndrome.
        J Pediatr. 1998; 133: 214-219
        • Nowak-Wegrzyn A.
        • Sampson H.A.
        • Wood R.A.
        • Sicherer S.H.
        Food protein-induced enterocolitis syndrome caused by solid food proteins.
        Pediatrics. 2003; 111: 829-835
        • Ruffner M.A.
        • Ruymann K.
        • Barni S.
        • Cianferoni A.
        • Brown-Whitehorn T.
        • Spergel J.M.
        Food Protein-induced enterocolitis syndrome: insights from review of a large referral population.
        J Allergy Clin Immunol Pract. 2013; 1: 343-349
        • Caubet J.C.
        • Nowak-Wegrzyn A.
        Food protein-induced enterocolitis to hen's egg.
        J Allergy Clin Immunol. 2011; 128: 1386-1388
        • Sopo S.M.
        • Giorgio V.
        • Dello Iacono I.
        • Novembre E.
        • Mori F.
        • Onesimo R.
        A multicentre retrospective study of 66 Italian children with food protein-induced enterocolitis syndrome: different management for different phenotypes.
        Clin Exp Allergy. 2012; 42: 1257-1265
        • Nowak-Wegrzyn A.
        • Muraro A.
        Food protein-induced enterocolitis syndrome.
        Curr Opin Allergy Clin Immunol. 2009; 9: 371-377
        • Bruni F.
        • Peroni D.G.
        • Piacentini G.L.
        • De Luca G.
        • Boner A.L.
        Fruit proteins: another cause of food protein-induced enterocolitis syndrome.
        Allergy. 2008; 63: 1645-1646
        • Monti G.
        • Viola S.
        • Tarasco V.
        • Lupica M.M.
        • Cosentino V.
        • Castagno E.
        A case of severe allergic reaction to cooked potato.
        Acta Paediatr. 2011; 100: e236-e238
        • Fernandes B.N.
        • Boyle R.J.
        • Gore C.
        • Simpson A.
        • Custovic A.
        Food protein-induced enterocolitis syndrome can occur in adults.
        J Allergy Clin Immunol. 2012; 130: 1199-1200
        • Leonard S.A.
        • Nowak-Wegrzyn A.
        Food protein-induced enterocolitis syndrome: an update on natural history and review of management.
        Ann Allergy Asthma Immunol. 2011; 107 (quiz 62): 95-101
        • Katz Y.
        • Goldberg M.R.
        • Rajuan N.
        • Cohen A.
        • Leshno M.
        The prevalence and natural course of food protein-induced enterocolitis syndrome to cow's milk: a large-scale, prospective population-based study.
        J Allergy Clin Immunol. 2011; 127 (e1-3): 647-653
        • Caubet J.C.
        • Nowak-Wegrzyn A.
        Current understanding of the immune mechanisms of food protein-induced enterocolitis syndrome.
        Expert Rev Clin Immunol. 2011; 7: 317-327
        • Fogg M.I.
        • Brown-Whitehorn T.A.
        • Pawlowski N.A.
        • Spergel J.M.
        Atopy patch test for the diagnosis of food protein-induced enterocolitis syndrome.
        Pediatr Allergy Immunol. 2006; 17: 351-355
        • Jarvinen K.M.
        • Caubet J.C.
        • Sickles L.
        • Ford L.S.
        • Sampson H.A.
        • Nowak-Wegrzyn A.
        Poor utility of atopy patch test in predicting tolerance development in food protein-induced enterocolitis syndrome.
        Ann Allergy Asthma Immunol. 2012; 109: 221-222
        • Boyce J.A.
        • Assa'ad A.
        • Burks A.W.
        • Jones S.M.
        • Sampson H.A.
        • Wood R.A.
        • et al.
        Guidelines for the diagnosis and management of food allergy in the United States: summary of the NIAID-Sponsored Expert Panel Report.
        J Allergy Clin Immunol. 2010; 126: 1105-1118
        • Powell G.K.
        Food protein-induced enterocolitis of infancy: differential diagnosis and management.
        Compr Ther. 1986; 12: 28-37
        • Knight A.K.
        • Shreffler W.G.
        • Sampson H.A.
        • Sicherer S.H.
        • Noone S.
        • Mofidi S.
        • et al.
        Skin prick test to egg white provides additional diagnostic utility to serum egg white-specific IgE antibody concentration in children.
        J Allergy Clin Immunol. 2006; 117: 842-847
        • Sampson H.A.
        • Gerth van Wijk R.
        • Bindslev-Jensen C.
        • Sicherer S.
        • Teuber S.S.
        • Burks A.W.
        • et al.
        Standardizing double-blind, placebo-controlled oral food challenges: American Academy of Allergy, Asthma & Immunology-European Academy of Allergy and Clinical Immunology PRACTALL consensus report.
        J Allergy Clin Immunol. 2012; 130: 1260-1274
        • Mori F.
        • Barni S.
        • Cianferoni A.
        • Pucci N.
        • de Martino M.
        • Novembre E.
        Cytokine expression in CD3+ cells in an infant with food protein-induced enterocolitis syndrome (FPIES): case report.
        Clin Dev Immunol. 2009; 2009: 679381
        • Onesimo R.
        • Dello Iacono I.
        • Giorgio V.
        • Limongelli M.G.
        • Miceli Sopo S.
        Can food protein induced enterocolitis syndrome shift to immediate gastrointestinal hypersensitivity? A report of two cases.
        Eur Ann Allergy Clin Immunol. 2011; 43: 61-63
        • Hwang J.B.
        • Sohn S.M.
        • Kim A.S.
        Prospective follow-up oral food challenge in food protein-induced enterocolitis syndrome.
        Arch Dis Child. 2009; 94: 425-428
        • Mehr S.
        • Kakakios A.
        • Frith K.
        • Kemp A.S.
        Food protein-induced enterocolitis syndrome: 16-year experience.
        Pediatrics. 2009; 123: e459-e464
        • Fleischer D.M.S.J.
        • Assa'ad A.H.
        • Pongracic J.A.
        Primary prevention of allergic disease through nutritional interventions.
        J Allergy Clin Immunol Pract. 2012; 1: 29-36
        • Monti G.
        • Castagno E.
        • Liguori S.A.
        • Lupica M.M.
        • Tarasco V.
        • Viola S.
        • et al.
        Food protein-induced enterocolitis syndrome by cow's milk proteins passed through breast milk.
        J Allergy Clin Immunol. 2011; 127: 679-680
        • Nomura I.
        • Morita H.
        • Hosokawa S.
        • Hoshina H.
        • Fukuie T.
        • Watanabe M.
        • et al.
        Four distinct subtypes of non-IgE-mediated gastrointestinal food allergies in neonates and infants, distinguished by their initial symptoms.
        J Allergy Clin Immunol. 2011; 127 (e1-8): 685-688
        • Lake A.M.
        Food-induced eosinophilic proctocolitis.
        J Pediatr Gastroenterol Nutr. 2000; 30: S58-S60