Intolerance to infant formulas containing cow's milk protein or soy protein is a diagnosis commonly made in pediatrie practice. The diagnosis is frequently made empirically, without adequate documentation. Consequently, the diagnosis is probably overestimated in the pediatrie population. This article will discuss intolerance to these two proteins- presented together because of their similar clinical presentations, a shared proposed pathophysiology, and a similar diagnostic work-up and treatment.
Cow's milk protein and soy protein intolerance (CMPSPI) is one of nine major causes of chronic diarrhea in infancy (Table 1). CMPSPI should also be included among the long list of differential diagnoses of malabsorption syndromes of childhood. Since so many infants use either cow's milk or soy formula as their sole nutritional support, many are at risk of developing intolerance.
CMPSPI is frequently suspected, often diagnosed, but rarely documented. An average of 0.3 percent to 7 percent of children aged zero to two years have cow's milk intolerance; only O.I percent to 0.3 percent of children above two years of age are intolerant.1 Because different population groups have been studied in the literature, without uniformity, there is a wide range of "average incidence." The incidence of soy protein intolerance is not known at this time.
Cow's milk protein and soy protein intolerance (CMPSPI) usually first presents in an infant less than six months of age, and in approximately 60 percent of the patients, within the first month of life. Symptoms usually begin within two weeks of contact with the cow's milk or soy protein formula. However, onset of intolerance may not develop until the second six months of life. In a patient initially breastfed, the mother's milk was well tolerated. However, once formula is introduced, the child's weight gain may be poor, and weight loss may occur. There is usually a strong family history of allergy. Among children less than four months of age, patients with CMPSI are predominately male, but among older children the sex ratio is equal.
Clinical manifestations of CMPSPI include changes in the gastrointestinal tract, the respiratory system, the skin, the blood, the child's behavior, and a number of miscellaneous changes. Patients present with almost any combination of the above, but foremost are gastrointestinal symptoms. The child will have a history of diarrhea. The stool is usually watery and may have no odor or a sweet smell, and may contain blood and polymorphonuclear neutrophils (PMNs). The stool pH may be below 5.5, and when the stool is tested with Clinitest® tablets, a positive reaction greater than or equal to 0.5 percent. Two-thirds of these patients vomit. "Failure to thrive," abdominal distention, protein-losing enteropathy, malabsorption, and steatorrhea may also be part of the clinical signs and symptoms of CMPSPI. Some patients have only blood and PMNs in their stools, with or without other symptoms, suggesting a diagnosis of colitis.2
The respiratory manifestations of CMPSPI can include recurrent infections, wheezing, rhinitis, and nasal stuffiness. The patient may have atopic dermatitis, and rarely will also have urticaria and angioedema. The child may be irritable, restless, and, at times, lethargic. Iron deficiency anemia secondary to chronic blood loss in the stools may be seen, as well as leukocytosis and neutrophilia.
CMPSPI commonly presents with fever and dehydration secondary to diarrhea, but only rarely with anaphylaxis. The child may have signs of malnutrition, with diminished subcutaneous fat, and, on occasion, peripheral edema. The serum electrolytes may be deranged secondary to the diarrhea, and vomiting may develop secondary to electrolyteimbalance. Occasionally, a child will appear toxic, and require an evaluation to rule out sepsis. Whether intolerant to cow's milk protein or soy protein, the patient will present in a similar fashion.
There is no simple laboratory test to confirm the diagnosis of CMPSPI. Patients will usually have a leukocytosis with a left shift, as well as anemia, hypoalbuminemia, and acidosis. IgA coproantibodiesto milk protein fractions, complement, and serum immunoglobulins are usually normal for age. Eosinophi) counts are usually normal. Skin tests, precipitin tests, and serum antibodies to cow's milk proteins are not reliable tests, since normal and affected infants may have similar results, in vitro studies of a lymphocyte migration inhibition factor to beta-lactaglobulin may in the future be a useful diagnostic test for cow's milk protein intolerance;3 such a test has not yet been developed for soy protein isolate.
Unfortunately, the pathological findings in small bowel biopsies from patients with CMPSPI are not unique. The majority of patients have varying degrees of villous damage. Typically, the villi are flattened and broadened, and the lamina propria contains an increased number of plasma cells and lymphocytes. It may also contain polymorphonuclear leukocytes. The epithelial cells are more cuboidal with less basal alignment of the nuclei, and there is an increased number of goblet cells. The small bowel lesion may, at times, be indistinguishable from that seen in celiac sprue.
MAJOR CAUSES OF CHRONIC DIARRHEA IN INFANCY
Examination of the colon at protosigmoidoscopy reveal a hemorrhagic, friable mucosa with Rectal biopsies may show crypt abscesses and a lack mucus within the rectal glands. The lamina typically shows increased numbers of plasma cells and lymphocytes. An increase in mitotic Figures in the crypts is also seen. The findings in CMPSPI are indistinguishable from those seen in infectious colitis or mild ulcerative colitis.
The pathogenesis of CMPSPI is not known. Damage to the intestinal mucosa is most likely secondary to toxic factors released by immune mechanisms. There are several immune mechanisms which may mediate the initial injury, including immediate hypersensitivity reactions, immune complex reactions and cell mediated reactions. These have all been shown to be abnormal in children with CMPSPI.
There are at least 20 antigenic components found in cow's milk.' The most common components are: I) casein; 2) alpha-lactalbumin; 3) beta-lactaglobulin; 4) bovine serum albumin; and S) gamma globulin. Betalactaglobulin is thought to be a likely antigenic stimulus, since it is not found in breast milk; Casein is found in reduced amounts and albumin in similar amounts in breast milk as compared with cow's milk.
It has been postulated that beta-lactaglobulin leads to intolerance in the following manner: Ordinarily, antigens are continuously absorbed by the enterocytes and transported into the interstitial space.4 A local immune response then produces specific antibodies which appear in the mucous coat over the cell surface. Antigen (cow's milk protein, soy protein, bacteria, etc.) penetration is inhibited by the interaction of antibody with antigen after immunization. Under normal conditions, immunological factors within the intestinal lumen, on the surface of the epithelial cell, and within the lamina propria combine to limit the access of antigens to the systemic circulation. However, when these natural immune defenses are disrupted, excessive antigenic material may be allowed to enter the circulation and lead to clinical disease.4
Factors contributing to pathological absorption of antigens include: 1) decreased intraluminal digestion; 2) disrupted mucosal barrier (increased permeability); 3) dosage and physical properties of the particles themselves; and 4) decrease in IgA-producing plasma cells in the lamina propria.5 Physiologic IgA deficiency in the newborn has been proposed as an important determinant in gastrointestinal sensitizaron to food antigens. High titers of serum antibodies to cow's milk proteins are found in babies fed cow's milk. Since antibodies to cow's milk protein are normally low in newborns, this suggests leakage of antigen through the intestinal mucosa.
The antigenicity of alpha-lactalbumin, beta-lactaglobulin and casein is only partially inactivated by commercial processing or heating. There is a striking tendency forthe infant to develop soy protein intolerance when soy protein is fed immediately after cow's milk protein. The bowel may have been previously damaged from earlier exposure to cow's milk protein, and the increased intestinal permeability may contribute to subsequent development of soy protein intolerance. This must be considered when feeding an infant an entirely new, whole protein immediately following a severe reaction to another.
Confirmation of a patient's intolerane to cow's milk or soy protein may be accomplished by eliminating the protein from the diet and then rechallenging with the same protein. The suspected formula should be eliminated for at least three weeks, or even for a few months, in order for the symptoms to resolve and to give the mucosa sufficient time to heal. The patient should be stable, with a documented appropriate weight gain. The challenge must be done in the hospital, since symptoms may be severe. Sucrose and lactose tolerance tests should be performed prior to the challenge with the complete formula, to eliminate sucrose and lactose intolerance as a cause of symptoms. The appearance of sucrose and lactose malabsorption when complete cow's milk and soy formulas are fed during challenges will be used as an indicator of mucosal injury.
Following a six to eight hour fast, the patient is given I ml of the suspected formula. If symptoms do not occur within one hour, the dose is doubled. If, after one hour, there are still no symptoms, the dose is again doubled. This is repeated hourly until I20 ml of formula are given. If the patient remains asymptomatic, he should be fed ad libiium for at least five days, since symptoms can occur at any time within this period. The rectum and small bowel should be biopsied 24 to 48 hours after the appearance of symptoms- the development of stools which are hemoccult positive, CHnitest positive of 0.5 percent or more, and pH less than or equal to 5.5. Proctoscopy should also be performed at this time.
The test is positive if symptoms develop anywhere within hours to five days after exposure to the formula. All stools during this period must be evaluated for blood, PMNs, pH, and reducing substances with Clinitest tablets. The white count usually increases six to ten hours after the challenge, if the patient is acutely sensitive to the formula. The clinical seventy does not correlate with the increased PMNs, and the white count usually returns to normal within 24 hours, if the test is positive, the patient will have a return of the original presenting symptoms; anaphylaxis during rechallenge has been reported rarely.
Once CMPSPI is diagnosed, the antigen should be completely eliminated from the patient's diet. A more basic formula, such as Pregestimil, Nutramigen, or an elemental diet containing only peptides and /or amino acids, may be used. Goat's milk may have high cross reactivity with cow's milk or soy protein, and should be avoided. Some authors recommend sodium cromolyn or antihistamines as anti-allergy measures; in our experience, the use of sodium cromolyn has not been beneficial in patients with cow's milk or soy protein intolerance. The elimination of the antigen from the diet is far simpler.
The prognosis for children with CMPSPI will vary according to the age at presentation. Infants who develop intolerance before six months of age tend to tolerate the protein by age one to two years.' They should be rechallenged every six months to see if tolerance to the protein has appeared. Approximately 50 percent of patients with intolerance to one protein will be intolerant to the other as well.
The mucosal injury usually repairs in three to 12 months following elimination of the allergen. These patients may have a high incidence of other food allergies, including eggs, cereals, chocolate, oranges, legumes and poultry. This is probably a result of the original injury which then allows penetration of other antigens.
The incidence, clinical symptoms, proposed pathogenesis, work-up, treatment and prognosis of cow's milk protein and soy protein intolerance have been discussed. It is an important diagnosis to make and document, since treatment is simple and effective, and can prevent severe morbidity and malnutrition in the young child.
1. Bahna Sl., Heiner DC. Cow's milk allergy: Paihogenesis, manifestations, diagnosis and management. In: Advances in Pediatrics. New York, Year Bk Med. 1978. p l-37.Halpin TC, Byrne W.I. Ameni ME. Colitis, persistent diarrhea, and soy protein intolerance. J Pediair 9I(3):404-407, 1977.
3. Ashkenazi A, et al. In vitrti cell-mediated immunologie assay for cow's milk allergy. Pediatrics 66(31:399-402. 1980.
4. Walker WA. Isselbacher KJ. Uptake and transpon of macromolecules by the intestine; Possible role in clinical disorders. Gaslroenterology 67:531550. 1974.
5. Walker WA. Antigen absorption in the small intestine and gastrointestinal disease. Ped CHn North Am 22:731-745.
6. Powell GK. Milk and soy induced enterocolitis of infancy. J Pediatr 93(4):553-560. 1978.
1. East ham EJ. Walker WA. Effect of cow's milk on the gastrointestinal tract: A persisten! dilemma for the pediatrician. Pediatrics 60(4):477-48I. 1977.
2. Ameni ME, Rubin CE. Soy protein another cause of the flat intestinal lesion. Gaslroenterology 62(2):227-234, 1972.
3. Shiner M, Ballard J, Smith ME. The small intestinal mucosa in cow's milk allergy. Lancet 1:136-140, January 18, 1975.
4. Whitington PF. Gibson R. Soy protein intolerance: Four patients with concomitant cow's milk intolerance. Pediatrics 59(5);730-732, 1977.
5. Gryboski .ID, Kocoshis S. lmmunoglobulin deficiency in gastrointestinal allergies. .1 Clin Gasiroenterol 2:71-76. 1980.
6. Manuel O. Walker-Smith. I A, France N E. Patchy enteropathy in childhood. Ciut 20: 211-215. 1979.
7. Butler H I., et al. Depressed neutrophil cherriotaxis in infants with cow's milk and or soy protein intolerance. Pediatrics 67(2): 264-268, 1981.
MAJOR CAUSES OF CHRONIC DIARRHEA IN INFANCY