A 5-Month-Old Infant with Cardiorespiratory Arrest Caused by Food Protein-Induced Enterocolitis Syndrome to Cow's Milk Proteins

Abstract

Food protein-induced enterocolitis syndrome (FPIES) is a nonimmunoglobulin E cell-mediated food allergy, which occurs predominantly in infants and young children. The most commonly incriminated triggers are cow's milk (CM), soy, and grains. Acute FPIES can be potentially life-threatening and culminate in shock requiring fluid resuscitation in at least 15% of the cases. To our knowledge, there have been no reports in literature of cardiorespiratory arrest induced by acute FPIES. We describe the first case of cardiorespiratory arrest occurred after accidental ingestion of a CM-based formula in a 5-month-old infant with previous diagnosis of FPIES to CM.

Introduction

Food protein-induced enterocolitis syndrome (FPIES) is a nonimmunoglobulin E (non-IgE) cell-mediated food allergy ¹ that usually has an onset in young infants. Its incidence ranges from 0.015 to 0.7% but it is reportedly increasing. ² The most commonly incriminated triggers are cow's milk (CM), soy, and grains, but there are reports of reactions to many other foods. FPIES can manifest as acute or chronic symptoms according to the frequency of food trigger in the diet. The first reaction of acute FPIES usually occurs after the first exposure to the offending food, presenting with repetitive projectile vomiting, pallor, and altered behavior ranging from decreased activity level to lethargy within 1 to 3 hours after ingestion of the causative food. Diarrhea can follow for up to 24 hours, occasionally with blood in stool, mostly reported in infants at less than 2 months of age. Hypothermia, dehydration, and hypotension are other potential clinical findings in severe acute FPIES. If the trigger food is eliminated from the diet, acute symptoms resolve spontaneously within 24 hours of the onset of the reaction. During elimination diet the child remains asymptomatic with normal growth. ¹ Chronic FPIES occurs in infants daily fed with the offending food, typically CM or soy-based formula. Chronic symptoms appear within 4 months of age and include intermittent but progressive emesis, chronic diarrhea (occasionally bloody), and failure to thrive. Switching to an elimination diet, symptoms resolve within 3 to 10 days but an accidental trigger food exposure or an oral food challenge (OFC) induces an acute reaction (acute-on-chronic FPIES). ¹

Despite the potential severity of FPIES, misdiagnosis and delayed diagnosis are common because of difficulties in differential diagnosis of FPIES and both allergic and nonallergic diseases, including first sepsis and infectious gastroenteritis, but also metabolic and toxic disorders, head injury, and surgical emergencies. ³

To our knowledge, there have been no reports in literature of cardiorespiratory arrest induced by acute FPIES. We describe the first case of a young infant with FPIES to CM who went into cardiorespiratory arrest after accidental ingestion of a CM-based formula (CMF). The purpose of this case report is to demonstrate the potential severity of acute FPIES even when it occurs with few visible water losses and the patient is well-appearing without signs of dehydration.

Case Report

A hospitalized 45-day-old infant developed a vomiting episode 30 minutes after the first ingestion of a CMF used to supplement breastfeeding. The patient was a preterm female infant (33 weeks of gestational age), with multiple bowel and colonic atresias surgically repaired on the second day of life. She was fed with total parenteral nutrition (TPN) until the twentieth day of life. The exclusive breastfeeding was then started and supplemented on the 45th day of life, when the acute episode occurred. Extensively hydrolyzed protein formula (eHF) was introduced, but vomiting recurred whenever the infant was fed with eHF. Intestinal subocclusion as postoperative complication was then considered and abdominal X-ray revealed intestinal dilatation. Therefore the patient underwent exploratory laparotomy and subsequent lysis of identified abdominal adhesions. She was fed with TPN and successively with both maternal and donor human milk, being well.

While she was still hospitalized (66th day of life), 1to 2 hours after the new attempt to administer her CMF, another vomiting episode occurred, followed by bloody diarrhea (lag time of 3–4 hours). Not even eHF was tolerated (vomiting after 2 hours) and after rapid and spontaneous symptom resolution the patient was fed with an amino acid-based formula (AAF), without developing further symptoms. A surgical evaluation was performed and abdominal postoperative complications were excluded; sepsis and metabolic work-up also proved negative and then the patient was referred to our Pediatric Allergy Unit when she was 75 days old.

CM allergy was suspected, specific IgEs to CM were negative and although there was a high index of suspicion for FPIES to CM confounding factors which made the diagnosis unclear, an OFC to CM was performed in our medically supervised setting, when she was 4 months old. Approximately 3 hours after ingestion of 100 mL of CMF, she developed repetitive vomiting and she looked pale and lethargic. Within 2 hours of rehydration, her conditions returned to baseline. FPIES to CM was diagnosed and the patient was discharged with CM dietary restrictions.

At the age of 5 months she was accidentally fed with 60 mL of CMF. An hour later the patient presented an episode of projectile vomiting, followed by mild diarrhea with blood and mucous after 3 hours of ingestion. She was therefore admitted to Pediatric Emergency Department. At the time of presentation she was well appearing and physical examination did not reveal anything significant, except for fever (body temperature of 38.7°C). Muscle tone and state of consciousness were preserved and there were no evident signs of dehydration. Just a further mild diarrhea episode with blood and mucous occurred.

In spite of her allergy history, an infectious gastroenteritis was also suspected at that time and the infant was kept under medical supervision in short stay observation unit. Routine blood chemistry, blood gas analysis, and radiographic testing were not obtained. A peripheral intravenous line was not placed. The patient was fed with AAF, without vomiting. No further symptoms developed.

Twelve hours after admission the patient's conditions deteriorated significantly. She became profoundly lethargic and went into cardiorespiratory arrest. Physical examination was characterized by cyanosis, cold and mottled extremities, absent capillary refill, no central pulses, and irrevocable unresponsiveness. The blood gas analysis highlighted a metabolic acidosis (pH 6.79; pCO ₂ 46.8 mm Hg; HCO ₃ ⁻ 5.5 mEq/L; base excess −24.5 mmol/L; lactates 10.7 mmol/L). Cardiopulmonary resuscitation was promptly performed, in addition to two intraosseous bolus of epinephrine, one bolus of normal saline (20 mL/kg), and another bolus of sodium bicarbonate. She was intubated and transferred to the intensive care unit. A distributive shock was diagnosed and after stabilization a broad work-up was initiated. Sepsis was taken into account as the most probable etiology. Laboratory tests showed neutrophilia (white blood count 18.690/mm ³ , absolute neutrophil count 8.750/mm ³ ) and inflammatory markers elevation (C-reactive protein level 30.5 mg/L, platelet count 430.000/mm ³ ). In spite of these laboratory findings septic shock was ruled out because of negative bacterial cultures in biological samples (blood, urine, and stools). The search of both fecal blood and viral antigens in feces was negative. Furthermore, no foci of infection was detected by thoracic X-ray as well as by abdominal examination which was consistently performed by pediatric surgeons. Surgical etiologies were also ruled out.

At the end, our allergy consultation was requested and the diagnosis of a cardiorespiratory arrest induced by acute FPIES to CM was made. According to this diagnosis, the patient was persistently fed with AAF and in few hours she returned to her usual state of health. At present, our 24-month-old patient continue to follow a strict CM-free diet and no further FPIES symptoms occurred.

Discussion

Acute FPIES can be severe and culminate in shock necessitating fluid resuscitation in at least 15% of reactions. ¹ However, to our knowledge no cases of cardiorespiratory arrest had ever been reported in literature. The aim of this report is to focus the attention of emergency physicians and general pediatricians on the fact that acute FPIES may represent a medical emergency in any case, even in patients with few or no visible water losses and/or with no signs of dehydration since the onset of symptoms.

Profuse vomiting and diarrhea may obviously contribute to dehydration in acute FPIES, but shock may also occur in absence of significant visible water losses. In effect it has been suggested that increased tumor necrosis factor-α (TNF-α) expression by activated T cells in the epithelium and lamina propria of the small intestine increases intestinal permeability, leading to third spacing fluid with subsequent dehydration. ⁴ TNF-α also potentiates synthesis of prostaglandins eliciting the relaxation of vascular smooth muscle tone. ⁵ TNF-α-induced vasodilation leads to third space fluid too, as well as hypotension, contributing to the pathogenesis of shock. Furthermore, a neuroimmune mechanism may be involved based on the reported successful use of ondansetron. ^{1 6} Serotonin has been indeed implicated in symptoms of FPIES ^{1 6} including effects on the blood pressure. ⁷ Interestingly, as seen in our case, between the onset of symptoms and the occurrence of cardiorespiratory arrest the patient presented only a vomiting episode and mild diarrheic stools with no signs of acute dehydration, showing that visible water losses not always are the only pathogenetic mechanism in acute FPIES shock. Mild symptoms do indeed not exclude the occurrence of other mechanism described above, as well as the cessation of clinical manifestations does not mean that the reaction is definitively resolved.

After elimination of trigger food acute symptoms may indeed occur up to 24 hours of the onset of the reaction, ¹ during which hemodynamic instability can develop even when unexpected on the basis of clinical conditions, as suggested by our case.

We may hypothesize that in our patient all mechanisms described above occurred, causing either a loss of effective blood volume circulation and also a vasodilatation which led to such a severe hemodynamic instability culminating in cardiorespiratory arrest.

It is unknown how much past intestinal atresias and surgeries in this patient contributed to the intravascular volume depletion and vasodilation. Indeed major surgical interventions induce an increase in levels of proinflammatory cytokine including TNF-α ⁸ with its known effects on the intestinal permeability, as well as on the vascular tone.

Thus we may hypothesize that in this patient either FPIES pathogenesis and presumably also past surgeries increased the intestinal inflammation and then the damage to the mucosal barrier, as well as the relaxation of vascular smooth muscle tone.

Fever was a confounding factor which masked the etiopathogenesis of gastrointestinal symptoms as well as shock: it is well-known that acute FPIES may at most progress to hypothermia. ¹ Nevertheless, fever was reported in a small cohort of Japanese infants with acute FPIES, occurring approximately 6 hours after the ingestion of CMF and subsided within 24 hours after elimination of trigger food. ⁹ This fever pattern is not typically associated with an untreated infectious disease, especially if bacterial. Temporal relationship between symptoms, including fever, and specific food intake is indeed an important clinical pointer to acute FPIES. ¹

Thus, even if unusual, fever with these features does not exclude acute FPIES reaction. Furthermore, increased reactive C-protein, neutrophilia, and thrombocytosis are described during acute FPIES reactions. ^{1 9} More in detail, diagnostic criteria for the interpretation of OFC in patients with a history of possible or confirmed FPIES include indeed an increased neutrophil count of ≥1.500/mm ³ above the baseline count. ¹ However, especially if hemodynamic instability occurs, it is mandatory to rule out sepsis. In addition to the specific food intake relation, as suggested by our case, distinguishing features are inflammatory markers elevation and neutrophilia without infection findings (negative bacterial cultures in biological samples, no foci of infection detected) and a positive history of allergy.

However, in a contest of previous diagnosis of FPIES, pediatric health care providers should be vigilant for prompt treatment of acute reaction even when diagnosis is not convincing.

According to the International Consensus Guidelines for the management of FPIES, mild reactions can resolve with oral rehydration. ¹ Furthermore, the dramatic evolution of this case suggests that even mild acute FPIES episodes may result in delayed and potentially fatal systemic reactions necessitating intravenous fluid resuscitation. It may be therefore considered securing with a peripheral intravenous access children with previous diagnosis of FPIES presenting symptoms after accidental ingestion of the trigger food, even when the patient is well-appearing without significant visible water losses or signs of acute dehydration.