Milk-protein intolerance affects approximately 3% of infants(1-4). Typically, 30% to 50% of infants with cow's milk intolerance are also allergic to soy (1,2). Unfortunately, there are also increasing reports of infants with persistent allergies to casein hydrolysate formulas(5-9). In the past, these infants often required prolonged parenteral nutrition. However, most infants with allergies to casein hydrolysate can now be successfully treated with amino acid formulas (7-9). Nonetheless, we report two patients with persistent symptoms of severe gastrointestinal food allergies while receiving amino acid-based formulas.
AA was the full-term product of an uncomplicated pregnancy, who was initially fed with a cow's milk formula (Enfamil; Mead Johnson, Evansville, IN, U.S.A.). On (postnatal) day 2 of life, bloody, mucous stools developed. Her formula was changed to a soy-based formula (Isomil; Ross Laboratories, Columbus, OH, U.S.A.). When the blood and mucus persisted, the formula was again changed on day 4 to a casein hydrosylate formula (Nutramigen; Mead Johnson, Evansville, IN, U.S.A.). On day 7, forceful emesis, fever, and bacteremia with bifidobacterium developed. She was admitted to a hospital and treated with antibiotics. After recovery, the patient was again fed with a cow's milk formula (Lactofree; Mead Johnson, Evansville, IN, U.S.A.), and profuse, bloody, mucous diarrhea and acidosis developed within 2 days, with no other signs of atopy. Stool cultures were negative. AA was transferred to the pediatric intensive care unit at Robert Wood Johnson Hospital, New Brunswick, New Jersey, U.S.A.
After initial bowel rest, the symptoms resolved, and her stools were heme negative. She was fed an amino acid formula (pediatric Vivonex; Eaton Laboratories, Nowich, NY, U.S.A.), and after 5 days, she had more than eight liquid, bloody, mucous stools per day. Flexible sigmoidoscopy revealed patchy erythema and hemorrhage. Analysis of rectal biopsy specimens showed markedly increased eosinophils in the lamina propria and mild cryptitis. Attempts to feed her with a different amino acid-based formula (Neocate; Scientific Hospital Supplies, Gaithersburg, MD, U.S.A.) resulted in persistent vomiting and bloody, mucous stools, occurring as many as 15 times daily. Upper endoscopy revealed patchy erythema of the esophagus and antrum. Histologic studies showed marked eosinophilic infiltration of the esophagus and antrum. Findings in an immunologic evaluation, which included determination of complement levels, immunoglobulins, IgG subclasses, nitroblue tetrazolium test, T-cell subsets, and antiepithelial antibodies, were normal. A central venous catheter was placed, and the patient received nothing by mouth and remained asymptomatic for 2 weeks. Flexible sigmoidoscopy and endoscopy were repeated before feedings were reinstituted, and esophageal mucosa revealed a mild eosinophilic infiltrate. Results of examination of antral, duodenal, and rectal specimens were normal. Feedings were again attempted with an amino acid-based formula (Neocate). Initially, the patient tolerated the feedings without difficulty (stools less than three times a day; all stools heme negative); however, after 1 week of feeding, she again had persistent vomiting and explosive, bloody, mucous stools occurring 10 to 15 times or more daily. A lamb's meat-based hypoallergenic formula(10) was tried after bowel rest. After 4 days, the patient again developed bloody, mucous diarrhea. Sigmoidoscopy again showed increased patchy erythema. Histologic study showed increased eosinophils in the lamina propria and mild cryptitis. Normal ganglion cells were present in the rectum. The patient was treated with a 6-week course of corticosteroids(intravenous, methylprednisolone; oral, prednisone; 1 mg/kg per day for 2 weeks, then slowly tapered). Nasogastric administration of the amino acid-based formula (Neocate) was slowly advanced to full energy requirement. Pureed foods were gradually introduced at when the patient was approximately 5 months of age. Radioallergosorbent test results for milk and soy were negative. At 1 year of age, she is thriving on a milk- and soy-free diet.
BB was the full-term product of an uncomplicated pregnancy, who was was initially breast fed without difficulty. At 9 weeks of age, she had low-grade temperatures and heme-positive diarrhea with cramps six to seven times a day. Stool cultures were negative. Her mother was instructed to withdraw all milk products from her child's diet. Because of persistent cramps and diarrhea, the child was fed a casein hydrolyzed formula (Nutramigen; Mead Johnson, Evansville, IN, U.S.A.). Worsening, bloody, mucous diarrhea occurred more than eight times daily for the next week, with no other signs of atopy. Sigmoidoscopy revealed patchy erythema and nodules with superficial ulceration and friability. Histologic studies showed markedly increased eosinophils with cryptitis. The patient was fed an amino acid-based formula(Neocate) and was asymptomatic during the next 5 weeks, with 2 to 3 heme-negative, pasty stools daily.
Acute vomiting, cramps, and diarrhea then developed. Stools cultures for ova and parasites and C. difficile toxin were negative. Results of a kidney and upper bladder examination were normal. The diarrhea worsened to 15 to 20 liquid, bloody, mucous stools a day. Upper endoscopy findings were normal. Sigmoidoscopy showed patchy erythema and nodules. Histologic study again revealed markedly increased eosinophils in the lamina propria, cryptitis, and Paneth's cell metaplasia. Nothing was fed the patient by mouth for approximately 1 week. Slow nasogastric rechallenge with the amino acid formula (Neocate) was initially tolerated, and the patient had normal, heme-negative stools less than three times daily. However, after approximately 10 days, profuse, bloody, mucous diarrhea again developed that occurred 20 times or more daily. A central catheter was placed. She received no oral feedings, and was asymptomatic for the next week. Nasogastric administration of a different amino acid formula (pediatric Vivonex) was slowly advanced to full energy requirment during a 2-week period. At 6 months of age, the patient was fed rice cereal without difficulty and continues to thrive and grow consuming an amino acid formula (pediatric Vivonex). Radioallergosorbent test results remained negative for milk and soy.
We report two infants with severe milk-protein allergy despite treatment with amino acid-based formulas. The diagnosis of milk protein colitis is based on characteristic clinical and histologic findings. Radioallergosorbent test results are usually negative in children with no other signs of atopy(5). The most definitive diagnosis requires rechallenge, although this is often not possible in patients with severe symptoms(11). Endoscopic and histologic features of allergic colitis were recently reviewed by Odze et al. (12) Typical endoscopic features usually include patchy erythema, increased nodules, and friable-appearing mucosa, with intervening areas of normal mucosa. Pathologic features of milk-protein allergy are often patchy, but may include increased eosinophilic infiltration of the rectum and perhaps cryptitis and Paneth's cell metaplasia (12-15). Esophageal eosinophilia and villous blunting have also been reported (16,17). Clinical, endoscopic, and histologic features typical of milk protein allergy were present in both patients, and both patients had clinical recurrence with rechallenge by amino acid formula. Both patients had significant clinical improvement at approximately 5 months of age, and neither patient showed evidence of persistent gastroenterologic symptoms or immune dysfunction.
The histologic improvement that occurred in the first patient while she was receiving no oral feedings, followed by the clinical and histologic relapse after oral feeding began again, is strongly consistent with an allergic origin. Unfortunately, no biopsies were performed during the quiescent period in the second patient that would have enabled us to distinguish between persistent colitis that was unresponsive to the amino acid formula and colitis resulting from allergic reaction to components of the amino acid formulas. However, the extended periods when these patients had normal, heme-negative stools between formula rechallenges argues against the possibility of persistent severe colitis. In both patients, the severity of symptoms precluded the possibility of challenging with unprocessed cow's milk or soy antigens.
We hypothesize that small amounts of milk or vegetable protein survive the manufacturing process. The reintroduction of unprocessed cow's milk protein(lactose-free) in the first patient or the presence of an unidentified intestinal infection may have heightened the immunologic response to even minute amounts of antigens in these patients. Using a double-antibody-linked immunosorbent assay, Isolauri et al. (18) isolated minute amounts of β-lactoglobulin in two amino acid formulas (1.6 - 31× 10-3 µg/g; detection limit, 0.3 × 10-3µg/g). The amount of β-lactoglobulin in one of the amino acid formulas (Nutri-Junior; Nutricia, The Netherlands) exceeded that of the casein and whey hydrolysates, whereas the amount of β-lactoglobulin in another amino acid formula (Neocate) was approximately 10% to 15% of that of the casein and whey hydrolysates. In contrast, Sampson et al.(19) found no detectable proteins in the amino acid formula Neocate using sodium dodecyl sulfate-polyacrylamide gel electrophoresis, immunoblot analysis, and inhibition enzyme-linked immunosorbent assay (19).
Because we did not perform any direct analysis on the formulas, it is also possible that small amounts of antigen were introduced during the mixing of the formula or the cleaning of the formula containers. We encourage families to use fresh formula containers and not to clean the containers with utensils used for other foods. Infants may also have an allergic reaction to the preservatives in these formulas. Multiple reports suggest that infants can have severe anaphylactic reactions to parenteral amino acid formulas, intralipids, multivitamins, or intravenous dipeptides or tripeptides(20-23). Although specific allergens have not been identified in parenteral nutritional formulations, it has been assumed that sulfites or other preservatives (e.g., butylated hydroxyanisole, butylated hydroxytoluene, polysorbate emulsifiers, and others) contribute to the development of severe allergic reactions.
The prevalence of allergies to whey or casein hydrolysates is unknown. Sampson et al. (24) suggested that less than 10% of infants with cow's milk allergy cannot tolerate a casein hydrolysate formula(24). In contrast, de Boissieu et al.(8) suggested that as many as 19% of infants with cow's milk protein allergy may also react to extensively hydrolyzed formulas. Vanderhoof et al. (7) suggested that 95% of patients with cow's milk allergy tolerate protein hydrolysate formulas, although few data are offered to support this claim. In the past, many infants with persistent feeding intolerance to soy formulas or casein hydrolysates were labeled as having intractable diarrhea of infancy (25). Typically, these patients are malnourished and have watery, diarrhea and malabsorption after an acute intestinal illness. Blood and mucus in the stool are not typical. Prolonged parenteral nutrition is sometimes necessary. It is certainly possible that many of these infants have a milk-protein enteropathy and may benefit from an amino acid formula.
Previous reports on the use of amino acid-based formulas have documented their usefulness. Vanderhoof et al. (7) reported that 25 of 28 patients with symptomatic allergic colitis had complete resolution of symptoms with an amino acid formula (Neocate). Seventeen of the 25 patients who were rechallenged with casein hydrolysate formula had recurrence of their symptoms. Hill et al. (9) treated 18 patients with symptomatic milk protein allergy with an amino acid-based formula (Neocate). Symptoms in all 18 infants improved. Twelve of 18 patients had recurrence of symptoms with reintroduction of soy or hydrolyzed formulas. McLeish et al.(26) reported the successful use of an amino acid formula in 20 infants with persistent diarrhea. One patient treated with the amino acid formula had necrotizing enterocolitis. Details of this case are not available. However, that all patients in the study were more than 36 weeks gestational age raises the possibility that this patient had persistent allergic colitis instead.
Although symptomatic improvement is usually reported in infants receiving amino acid formulas, these formulas do not completely normalize intestinal function in infants with milk-protein allergy. de Boissieu et al.(8) showed increased intestinal permeability in infants with persistent symptoms while consuming extensively hydrolyzed cow's milk formulas. The use of an amino acid formula (Neocate) resulted in significant improvement in mucosal permeability; however, a minority of patients showed complete normalization of permeability while consuming amino acid formula. In addition, three of the 28 infants studied by Vanderhoof et al.(7) continued to have occult blood in their stool after 2 weeks of amino acid formula (Neocate), although all other symptoms had resolved. These results suggest that some infants with milk-protein allergy who are fed with an amino acid formula continue to have mild gastrointestinal dysfunction.
Finally, in patients with cow's milk allergy, clinical challenge with cow's milk antigens usually results in profound symptoms within 24 hours(27). However, late-onset reactions (4 to 7 days) often occur in challenges with extensively hydrolyzed formulas(9). Because of the minimal amounts of antigens present in amino acid formulas and the use of slow, continuous nasogastric feedings in these two patients, longer periods until reaction would not be unexpected.
In summary, results of most studies indicate that amino acid formulas are safe for infants with persistent allergic symptoms, despite the use of hydrolyzed casein and whey formulas. However, infants treated with amino acid formulas may have persistent gastrointestinal dysfunction, including continuing allergic symptoms. We recommend gross or histologic confirmation of milk-protein allergy in infants without other atopic symptoms before feeding the patient an amino acid formula. Every attempt must be made to prevent the contamination of amino acid formulas with even minute amounts of milk or soy proteins. The diagnosis of allergy to amino acid formulas requires rechallenge with the formula for confirmation; in particular, documentation of normal mucosa between challenges may be necessary.
1. Gerrard JW, MacKenzie JWA, Goluboff N, Garson JZ, Maningas CS. Cow's milk allergy: Prevalence and manifestations in an unselected series of newborns. Acta Paediatr Scand Suppl
2. Jakobsson I, Lindebert TA. A prospective study of cow's milk protein intolerance in Swedish infants. Acta Paediatr Scand
3. Host A, Halken S. A prospective study of cow milk allergy in Danish infants during the first 3 years of life. Clinical course in relation to clinical and immunological type of hypersensitivity.Allergy
4. Schrander JJ, van den Bogart JP, Forget PP, Schrander-Stumpel CT, Kuijten RH, Kester AD. Cow's milk protein intolerance in infants under 1 year of age: A prospective epidemiological study.Eur J Pediatrics
5. Kelso JM, Sampson HA. Food protein-induced enterocolitis to casein hydrolysate formulas. J Allergy Clin Immunol
6. Rosenthal E, Schlesinger Y, Birnbaum Y, Goldstein R, Benderly A, Freier S. Intolerance to casein hydrolysate formula. Acta Paediatr Scand
7. Vanderhoof JA, Murray ND, Kaufman SS, et al. Intolerance to protein hydrolysate infant formulas: An unrecognized cause of gastrointestinal symptoms in infants. J Pediatr
8. de Boissieu, Matarazzo P, DuPont C. Allergy to extensively hydrolyzed cow milk proteins in infants: Identification and treatment with an amino acid-based formula. J Pediatr
9. Hill DJ, Cameron DJS, Francis SRD, Gonzalez-Anadaya AM, Hosking CS. Challenge confirmation of late-onset reactions to extensively hydrolyzed formulas in infants with multiple food protein intolerance.J Allergy Clin Immunol
10. Weisselberg B, Dayal Y, Thompson JF, Doyle MS, Senior B, Grand RJ. A lamb-meat-based formula for infants allergic to casein hydrolysate. Clin Pediatr
11. Powell GK. Milk- and soy-induced enterocolitis of infancy: Clinical features and standardization of challenge. J Pediatr
12. Odze RD, Weshil BK, Leichtner AM, Antoniolli DA. Allergic colitis in infants. J Pediatr
13. Winter HS, Antoniolli DA, Fukagawa N, Marcial M, Goldman H. Allergy-related proctocolitis in infants: Diagnostic usefulness of rectal biopsy. Mod Pathol
14. Machida HM, Catto Smith AG, Gall DG, Trevenen C, Scott RB. Allergic colitis in infancy: Clinical and pathological aspects. J Pediatr Gastroenterol Nutr
15. Odze RD, Bines J, Leichtner AM, Goldman H, Antonioli DA. Allergic proctolcolitis in infants: A prospective clinical-pathological biopsy study. Hum Pathol
16. Kelly KJ, Lazenby AJ, Rowe PC, Yardley JH, Perman JA, Sampson HA. Eosinophilic esophagitis attributed to gastroesophageal reflux: Improvement with an amino acid-based formula. Gastroenterology
17. Manuel PD, Walker-Smith JA, France NE. Patchy enteropathy in childhood. Gut
18. Isolauri E, Sutus Y, Makinen-Kiljunen S, Oja SS, Isosomppi R, Turjanmaa K. Efficacy and safety of hydrolyzed cow milk and amino acid-derived formulas in infants with cow milk allergy. J Pediatr
19. Sampson HA, James JM, Bernhisel-Broadbent J. Safety of an amino acid-derived infant formula in children allergic to cow milk.Pediatrics
20. Market AD, Lew DB, Schropp KP, Hak EB. Parenteral nutrition-associated anaphylaxis in a 4-year old child. J Pediatr Gastroenterol Nutr
21. Kamath KR, Berry A. Acute hypersensitivity reaction to intralipid (letter). N Engl J Med
22. Pomeranz S, Gimmon Z, Ben Zvi A, Katz S. Parenteral nutrition-induced anaphylaxis. JPEN
23. Bullock L, Etchason E, Fitzgerald JF, McGuire WA. Case report of an allergic reaction to parenteral nutrition in a pediatric patient. JPEN
24. Sampson HA, Bernhisel-Broadbent J, Yang E, Scanlon SM. Safety of casein hydrolysate formula in children with cow milk allergy.J Pediatr
25. Lo CW, Walker WA. Chronic protracted diarrhea of infancy: A nutritional disease. Pediatrics
26. McCleish CM, MacDonald A, Booth IW. Comparison of an elemental with a hydrolyzed whey formula in intolerance to cow's milk.Arch Dis Child
27. Powell GK. Milk- and soy-induced enterocolitis of infancy. J Pediatr