Key points

Introduction

Food protein–induced enterocolitis syndrome (FPIES) is a rare, non–immunoglobulin E (IgE)-mediated gastrointestinal food allergy primarily diagnosed in infancy. Acute FPIES reactions typically present with delayed, repetitive vomiting, lethargy, and pallor within 1 to 4 hours of food ingestion. Chronic FPIES typically presents with protracted vomiting, diarrhea, or both, accompanied by weight loss or poor growth. Foods regularly included in the diet are thought to induce chronic symptoms, such as cow’s milk (CM) or soy formula, while acute reactions are commonly reported after intermittent ingestion of the causative food. The pathophysiology of FPIES is unknown but is thought to be immunologic and cellular in nature.

Owing to nonspecific symptoms and the lack of diagnostic testing, FPIES is often initially misdiagnosed, leading to a delay in diagnosis and increased morbidity from extensive workups and hospitalizations. Varying clinical manifestations of FPIES in different populations further complicate the picture. An understanding of the various presentations of FPIES along with increased awareness of the condition by general practitioners and emergency room providers will help to improve the management and quality of life of patients with FPIES.

Introduction

Food protein–induced enterocolitis syndrome (FPIES) is a rare, non–immunoglobulin E (IgE)-mediated gastrointestinal food allergy primarily diagnosed in infancy. Acute FPIES reactions typically present with delayed, repetitive vomiting, lethargy, and pallor within 1 to 4 hours of food ingestion. Chronic FPIES typically presents with protracted vomiting, diarrhea, or both, accompanied by weight loss or poor growth. Foods regularly included in the diet are thought to induce chronic symptoms, such as cow’s milk (CM) or soy formula, while acute reactions are commonly reported after intermittent ingestion of the causative food. The pathophysiology of FPIES is unknown but is thought to be immunologic and cellular in nature.

Owing to nonspecific symptoms and the lack of diagnostic testing, FPIES is often initially misdiagnosed, leading to a delay in diagnosis and increased morbidity from extensive workups and hospitalizations. Varying clinical manifestations of FPIES in different populations further complicate the picture. An understanding of the various presentations of FPIES along with increased awareness of the condition by general practitioners and emergency room providers will help to improve the management and quality of life of patients with FPIES.

Epidemiology

Large population studies to assess the overall prevalence of FPIES are lacking. In the only such study with more than 13,000 Israeli infants prospectively enrolled, the cumulative incidence of CM FPIES was reported at 0.34%. IgE-mediated CM allergy, a condition considered more common than FPIES, was diagnosed in 0.5% of the same population.

Several studies have reported an increase in FPIES diagnosis, frequently attributed to increased awareness of the syndrome as separate from other food sensitivities. In an Italian cohort, a significant increase in the number of FPIES cases was noted in 2008, when the diagnostic criteria for FPIES were modified at one of its centers, with a constant level of cases reported thereafter. Alternatively, in an Australian cohort, although the number of FPIES cases increased steadily over a 16-year period, the median number of prediagnosis FPIES reactions did not substantially change, suggesting that increased awareness did not play a role. In a recent large United States cohort, no increase in the number of FPIES cases was observed over a 6 year period.

Although FPIES develops primarily in infancy, reports in older children and adults indicate that it may develop at any age. There does not seem to be a gender predilection in childhood, however a female predominance in a small cohort of adults with FPIES has been reported, similar to that seen in IgE-mediated food allergy. In a large United States cohort, 65% of FPIES subjects were Caucasian, mirroring the referral population evaluated their allergy clinic. In Israel, FPIES was more common in the Jewish population ( P = .03). Overall, FPIES has been reported across all races and ethnicities.

Many FPIES patients have an atopic background. Eczema was reported in 9% to 57% of FPIES subjects, similar to or higher than the overall population. Wheezing or asthma was reported in 3% to 25% of FPIES subjects, similar to or lower than overall population. Allergic rhinitis was reported in 38% of FPIES subjects, IgE-mediated food allergy to other foods in 11% to 30%, and eosinophilic esophagitis in 1 FPIES subject. A family history of allergic disease was reported in 20% to 77% of FPIES cases. In one cohort, 34% of FPIES subjects had a family history of food allergy, and 6% (n = 10) had a family history of FPIES.

CM FPIES in the Israeli cohort was not found to be associated with gestational age, birth weight, maternal age, number of siblings, dairy consumption by mother, or age of CM introduction. However, infants with CM FPIES were more likely than healthy infants to have been born by cesarean section (27% vs 15%, respectively) ( P = .003). Conversely, in a United States FPIES cohort, the rate of cesarean section was 29%, lower than the overall population. In the Italian FPIES cohort, 95% were breastfed for a median duration of 4 months (range 0.5–12), whereas the breastfeeding rate in a large United States cohort was 47%, similar to the national rate.

Clinical presentation

The rates of FPIES symptoms in different cohorts are presented in Table 1 . Vomiting is the predominant FPIES symptom, while diarrhea is typically reported in 50% or fewer FPIES cases. Isolated vomiting and isolated diarrhea have been reported. It has been suggested that diarrhea is likely associated with more severe cases, exacerbating fluid loss, and infancy. In the study by Sopo and colleagues, 47% versus 27% of acute reactions with and without diarrhea, respectively, required hospitalization and were treated with intravenous fluids (IVF). However, diarrhea was reported in only 7% of a large United States cohort of oral food challenges (OFCs), while hypotension was reported in 19%.

Hypotension has been reported in 5% to 77% of different cohorts. This large range is likely due to differences in methodology, whether blood pressure was measured, and the number of cases observed during OFCs or treated at a medical facility compared with symptoms reported by parents. Sopo and colleagues reported that pallor and hypotension/lethargy never occurred alone. There was no mention of pallor or hypotension in the adult cohort, although these symptoms were reported in an adult case report of scallop FPIES. In the adult cohort, abdominal pain was reported in 77% of cases, more so than vomiting (71%).

Other reported presenting symptoms included dehydration, lethargy, irritability, loss of consciousness, clamminess, hypotonia, cyanosis, and other stool changes, including bloody diarrhea, melena, and malodorous, pale or sticky stools. Mehr and colleagues reported that 6 of 25 (24%) patients who had body temperature measured were hypothermic (<36°C). Weight loss and poor growth has been reported in chronic FPIES.

The amount ingested that causes FPIES symptoms can be very low, as in the case of a 6-month-old diagnosed with rice FPIES who developed symptoms after chewing on a wrapper from a rice cake. Similarly, Bansal and colleagues reported 4 cases where increasingly smaller amounts were needed to trigger subsequent FPIES reactions. Conversely, Katz and colleagues reported that 54% of subjects with CM FPIES tolerated 121 mL or more of CM before having symptoms during OFCs.

Although approximately 60% of FPIES cases occur on first exposure, many cases occur after a period of tolerance. In the study by Mehr and colleagues, 12% of the cohort developed FPIES symptoms on second exposure, 12% on third, and 15% on fourth. Katz and colleagues reported that 16% of infants with CM FPIES tolerated CM for more than 4 days and 11% tolerated CM for 14 to 30 days before developing symptoms. FPIES in adults has been reported to be acquired after tolerating the trigger food previously.

The appearance of symptoms after ingestion of trigger foods is typically delayed in comparison with IgE-mediated food allergy. The median reported time from first dose to development of symptoms during OFCs in 2 cohorts was 1.5 to 2 hours, with a range of 0.5 to 4 hours. This is similar to the findings of Katz and colleagues, who reported a range of 30 minutes to 5.25 hours between ingestion and symptoms, with 60% occurring after 2 hours. In the fish-related FPIES cohort of Zapatero Remón and colleagues, two patients reported symptoms within 5 to 10 minutes and one showed symptoms at 6 hours; during observed OFCs the range of time from ingestion to symptoms was between 30 minutes and 5 hours. The duration of symptoms has been reported to be between 2 and 48 hours. During observed OFCs, the median time to recovery was 50 minutes (interquartile range [IQR] 13–95 minutes) in the cohort of Caubet and colleagues.

The age at onset of FPIES seems to differ according to food trigger ( Table 2 ). Likely because of delays in diagnosis, the overall median age at FPIES diagnosis was 15 months (IQR 9–24 months) in one cohort, and the overall mean age at diagnosis was 14 months in another cohort. Several cohorts report CM or soy FPIES presenting earlier than solid food FPIES, mostly before 6 months of age. Late-onset fish or shellfish FPIES was diagnosed after infancy in 7% of one cohort with a median age at diagnosis of 30 months. Zapatero Remón and colleagues reported the age of onset of fish FPIES to be between 9 and 12 months.

Ruffner and colleagues found no significant difference in age at FPIES presentation between breastfed and formula-fed infants. Caubet and colleagues found that in solid food FPIES, formula was introduced later (median 1.5 months) and breastfeeding lasted longer (median 2.8 months) than in those with CM or soy FPIES, for whom formula was introduced and breastfeeding lasted a median 0.03 months ( P = .0002). In this cohort, 62% of CM or soy FPIES infants were exposed to CM or soy formula in the first few weeks of life. In addition, most CM or soy FPIES presented with chronic symptoms, such as diarrhea, colitis, reflux, or failure to thrive, occurring shortly after CM or soy formula introduction, which resolved with avoidance. Subsequent acute episodes to CM or soy occurred at a mean age of 7 months (range, 0.3–60 months).

Acute CM or soy FPIES in exclusively breastfed infants is rare. Theories about breast milk being protective include the presence of protective immunoglobulin A, that the food is highly processed, or that the threshold dose may not be present in breast milk. Three breastfed infants in one cohort (N = 160) had symptoms of chronic FPIES to CM in the maternal diet. By contrast, in a parent survey (N = 263), a surprising proportion (42%) reported reactions through breast milk exposure. This result may be affected by recall bias and may depend on the definition of FPIES symptoms.

The amount ingested that causes FPIES symptoms can be very low, as in the case of a 6-month-old diagnosed with rice FPIES who developed symptoms after chewing on a wrapper from a rice cake. Similarly, Bansal and colleagues reported 4 cases where increasingly smaller amounts were needed to trigger subsequent FPIES reactions. Conversely, Katz and colleagues reported that 54% of subjects with CM FPIES tolerated 121 mL or more of CM before having symptoms during OFCs.