The Natural History of Atopic Disease

The association of allergy with cow milk has been documented in the literature for decades. ^{, ,} The incidence of this allergy in the general population has been noted to increase progressively since the original comments on the subject by Rowe in 1931. The incidence has reportedly increased 10 times in 20 years. It has been attributed to increased recognition; increased incidence of exposure to known allergens; and a gradual decrease in infection as a source of morbidity, because the use of antibiotics and immunization revealed an underlying allergic component to chronic symptoms. Glaser attributed this rapid increase in the development of allergic diseases to the abandonment of breastfeeding when safe, pasteurized milk became available. It was noted that 20% of all children were allergic by 20 years of age.

Studies of office pediatrics have shown that one-third of the visits are a result of allergy. One third of all chronic conditions in patients younger than age 17 result from allergy and one third of lost school days from asthma. In the evaluation of 2000 consecutive, unselected newborns in pediatric practice, 50% had family histories of allergy. Grulee et al. observed, as early as 1934, that eczema was seven times more common in infants fed cow milk than in breastfed infants. McCombs et al. reported in 1979 that asthma caused more than 2000 deaths and the loss of 94 million days of activity. It initiated 183,000 hospital admissions and more than 1 million hospital days in 1 year in the United States alone.

Asthma is the most common chronic disease of childhood, affecting an estimated 6.3 million children, according to a Centers for Disease Control and Prevention (CDC) report in 2001. ^, These data indicate that, in the United States, people with asthma collectively have more than 100 million days of restricted activity and 470,000 hospitalizations annually, with more than 5000 deaths annually. Asthma hospitalization rates have been highest among black adults and children, with mortality rates consistently highest among black individuals ages 15 to 24 years. Asthma costs the American public billions of dollars every year. Decades of investigation have resulted in conflicting results. In a systematic review of the literature on how to prevent the development of food allergy, there was one clear conclusion. Cow milk should be avoided for at least the first 4 months of life in children at risk for allergy. Maternal avoidance of allergens during pregnancy also produced conflicting results, except for the avoidance of cow milk during pregnancy when there was a family history. Maternal avoidance of cow milk resulted in lower levels of mucosal-specific IgA and a lower incidence of cow milk allergy in the infant. The American Academy of Pediatrics (AAP) also agreed that cow milk and dairy products should be avoided in at-risk infants for the first year of life. The AAP and others have also declared that soy milk has no role in the prevention of allergy. The AAP is very supportive of breastfeeding for at least 6 months, as well as the delay in starting solids until 6 months.

Bone turnover is increased when mothers are on elimination diets that include cow milk, cow milk products, and eggs, even when they are on supplemental calcium. Mothers who were found to have some bone mobilization for 6 months recovered quickly when breastfeeding was discontinued. Breastfeeding has increased in the new millennium. The incidence of allergy should diminish.

The Question of Heredity

Heredity undoubtedly plays a part in the development of allergic disease, an observation first recorded by Maimonides in his Treatise on Asthma in the twelfth century. Most studies in the past 60 years have concurred with the concept of a recessive mode of inheritance.

Kern has noted that the outstanding etiologic factor in human hypersensitivity is heredity. He states that few diseases exist in which heredity is so clearly identified and so common.

Hamburger reported that children with two atopic parents had a 47% chance of developing atopic disease. One atopic parent meant a 29% chance of developing atopy, and the risk dropped to 13% with no allergic parent.

In a study of asthmatic monozygotic twins, Falliers et al. observed similar serum immunoglobulin E (IgE), blood eosinophil counts, and positive skin tests to allergens in both twins. However, they had dissimilar responses to infection and methacholine. This finding suggests an acquired component to bronchial hyperactivity. Apparently several mechanisms are involved in antigen processing.

To identify infants at high risk for developing atopy, several approaches have been suggested. Cord serum total IgE levels of greater than 100 U/mL are associated with a five to ten times greater risk than lower levels. Eosinophilia and lymphocytes may prove to be markers, but, at present, only the family allergy history and the cord blood IgE have been significantly reliable predictors.

In the 1930s, Glaser speculated that if a child was at a high risk for developing allergy, prophylaxis should be able to change the outcome. The original work on prophylaxis was done by Glaser and Johnstone and reported in 1953. Only 15% of a group of children whose mothers controlled their own diet in pregnancy and the infants’ diets and environments at birth did develop eczema. In contrast, 65% of the sibling controls and 52% of the nonrelated controls who received cow milk developed similar allergic illnesses. As a retrospective study, it was open to some criticism, although it did begin to look at a significant issue—reducing the incidence of allergic manifestations in high-risk individuals by a new type of preventive measure: avoidance of known antigens.

A second study was designed and carried out prospectively by Johnstone and Dutton, in order to investigate dietary prophylaxis of allergic disease. They observed a difference of more than 10 years in the incidence of asthma and perennial allergic rhinitis in those fed soybean milk (18%) and those fed evaporated milk (50%). No infant in this study of 283 children was breastfed, however. A study of 1753 children fed breast milk, soy milk, and cow milk from birth to 6 months of age, who were followed until they were 7 years or older, was published. The children included those with high-risk, low-risk, and no-risk family histories for allergy. No difference in outcome was related to early diet, but a relationship to the family history was seen.

In a prospective study to identify the development of reaginic allergy, infants of allergic parents were placed in a study or control group. The study group followed an allergen-avoidance regimen, including breastfeeding. At 6 months and 1 year, the study infants had less eczema than the control infants, as well as lower serum total IgE levels.

Prophylaxis of Atopic Disease

Efforts to alter the incidence of atopic illness have continued to challenge investigators, who now have access to increased methodologic sophistication. Prevention of IgE-mediated disorders can be directed at the practice of interfering with any of the major forces responsible for the phenotypic expression of atopy. Practically, however, it is not yet possible to mask IgE genes or manipulate cellular components of the response organ. Clinicians are limited to manipulating the effect of the environment, by reducing the allergenic load.

Review of the plethora of studies directed at measuring the impact of dietary manipulations on the incidence of atopic disease demonstrates that retrospective studies show little or no difference in the incidence of asthma and eczema. Prospective studies, however, tend to demonstrate a significant reduction in atopic disease in the treated group ( Table 17-1 ). In looking at these data, it is important to recognize that some studies did not consider the risk for the population developing atopic disease on a hereditary basis. In other studies, breastfeeding may have been carried out for only a few weeks or months. The evidence is clear that exclusive breastfeeding for 6 months or longer makes a difference. None of these studies controlled for smoking in the household, and no data were reported on the incidence of respiratory syncytial virus. In addition, some studies did not control for the breastfeeding mother’s diet, the weaning foods, or the use of cow milk beverages. However, when the long-term effects of breastfeeding, maternal smoking during pregnancy, and recurrent lower respiratory tract infections on asthma in children were examined, some discordance was observed. Breastfeeding for less than 3 months was not an effective deterrent. Breastfeeding reduced the effect of lower respiratory tract infections on asthma. Similarly, it reduced the effect of smoking. The authors concluded that asthma in childhood can be prevented by promoting breastfeeding, preventing smoking in pregnancy, and avoiding recurrent lower respiratory tract infections in early childhood. Recurrent wheezing episodes were evaluated for associated risk factors. Cigarette smoking in the household, heating mode (open stove), and breastfeeding for less than 6 months were significant.

Hamburger et al. carried out prospective prophylactic studies to include measuring IgE and skin radioallergosorbent tests (RASTs) on mothers, fathers, and infants. They found a significant correlation between maternal IgE and infant IgE and potential allergy in the infants ( Tables 17-2 and 17-3 ). This study was done by controlling the environment and the diet. The process was initiated in pregnancy, in order to begin by protecting the fetus, and was then continued at birth. Therefore, considerable attention was directed toward breastfeeding in this and other studies.

Human milk consistently contains antibodies, especially secretory immunoglobulin A (IgA), to major food proteins. The levels are influenced by the mother’s own external antigen exposure.

In a study of 500 babies born to families at a high risk for allergies, one group was deliberately not given cow milk and was fed soy milk by random assignment. No benefit resulted from withholding cow milk, but breastfeeding, even for a short period, was clearly associated with a lower incidence of wheezing, prolonged colds, diarrhea, and vomiting. Smoking and environmental molds were also associated with wheezing. Merrett et al. concluded from this that breastfeeding played a significant role in prophylaxis.

The effect of breastfeeding on allergic sensitization is both direct, through the elimination of nonhuman milk protein as an exposure to antigen, and indirect, by affecting the absorption of antigen through the intestinal tract. ^, Maternal antibodies are transferred to breastfed infants as part of what has been called the enteromammary immune system ( Figure 17-1 ). The secretory IgA antibody present in milk is the result of a mother’s enteric immune response to antigens in her gut. Secretory IgA in a mother’s milk provides protection against bacterial, viral, and toxic exposures. Prospective studies have shown that infants at a high risk for atopic illness, from a hereditary standpoint, had significantly less disease when breastfed, especially if reared in a protected environment with delayed use of solid foods. This was compared with children of similar risk fed cow milk and regular solid foods. Serum IgE concentrations were also greatly reduced in infants younger than 6 months and 12 months of age in the breastfed group.

Maternal serum antibodies affect passage of foreign antigens into milk and processing of antigen in infant’s intestine.

(Redrawn from Kleinman RE: The role of developmental immune mechanisms in intestinal allergy, Ann Allergy 51:222, 1983.)

Infants with a low incidence of T lymphocytes are at greater risk to develop allergies if fed cow milk rather than breast milk, according to Juto and Juto and Bjorksten. Infants with reduced T cells fed cow milk also demonstrated higher serum IgE levels and peripheral eosinophil counts. Juto reported that with careful prophylaxis, more than 50% of infants who had both parents with IgE levels greater than 100 mg/mL showed elevated cord and 4-month IgE levels. More than 80% of those infants whose parents had IgE levels less than 100 mg/mL, however, had both low cord blood and low 4-month IgE levels. Such data confirm the genetic effect of both maternal and paternal genes.

Considerable work was reported from a laboratory in Newfoundland that promoted certain formulas as protective. This work has since been considered misleading.

In a prospective study designed to examine asthma and atopy outcomes in male/female patients, Mandhane et al. reported specific parental history of atopy and breastfeeding. The study members were born in New Zealand in 1972 to 1973 and followed to adulthood. Breastfeeding was considered positive if it lasted 4 weeks or more. There was no mention of exclusivity or when cow milk was introduced. Parental history was obtained. Skin testing, spirometry, and bronchial challenge to methacholine was done from age 9 at intervals forward. They found that breastfeeding, in spite of its brevity, influenced development of atopy and asthma by sex and parental history. They found greater incidence of atopy in girls who were breastfed than boys who were breastfed and both who were bottle fed. They acknowledged the importance of a thorough breastfeeding history. The need to correlate the incidence of parental atopy, whose offspring were more likely to be breastfed, was also emphasized. Mai et al. studied the relationship of breastfeeding, overweight, and asthma. They queried whether overweight and asthma shared common environmental influences, such as breastfeeding. They found that children who had been breastfed exclusively for less than 12 weeks had a risk for being overweight and of having asthma by 10 years of age. They also noted that they existed together but not separately. These mothers were obese in most cases, and the authors suggested that their obesity decreased their probability of successful breastfeeding. The long-term effects of maternal smoking during pregnancy and recurrent lower respiratory infections were associated with asthma in children. These effects were mitigated by breastfeeding in the long-term study of children in the Isle of Wight birth cohort.

Long-Term Effects of Allergy Prophylaxis

In an 18-month study of atopic outcome, atopic mothers were randomly allocated to an intervention group or an unrestricted-diet group, and both were compared with nonatopic mothers on unrestricted diets. The intervention was a milk/dairy product-free diet during late pregnancy and lactation. After 7 weeks of the diet, serum β-lactoglobulin and immunoglobulin G (IgG) levels in the mothers were collated to the levels in cord blood. The infants were examined at 12 and 18 months, utilizing a single-blind allergy assessment by a pediatrician. Infants born to nonatopic parents had significantly less allergy than those born to atopic mothers with unrestricted diets. The “restricted-diet group” of infants had comparable levels to the atopy-free group and had significantly less allergy than the unrestricted-diet group. The nature of the parents’ disease also played a role in the type of illness in both groups.

Mothers who consumed a diet similar to the Mediterranean diet, rich in fruits, vegetables, and fish and ample in vitamin D, showed greater impact on suppression of atopic disorders than those who did not. The role of vitamin D has just been recognized as being important in lactating women, especially those with restricted diets. All breastfeeding women should consume at least 1000 units of vitamin D while lactating. A prospective longitudinal study of 988 healthy infants, from birth to 6 years of age, recorded feeding-history episodes of lower respiratory tract infection in the first 3 years of life and recurrent episodes of wheezing. Being breastfed was associated with lower rates of recurrent wheeze (3.1% vs. 9.7%, p < 0.01) for nonatopic children. The authors concluded that recurrent wheeze at age 6 years is less common among nonatopic children who were breastfed as infants. This effect was independent of whether or not the child had a wheezing lower respiratory tract illness in the first 6 months of life ( Table 17-4 ). These authors recorded smoking history, but it did not alter the compelling influence of breastfeeding on the outcome.

Additional long-term studies have demonstrated that children who had ever been breastfed had a 50% lower incidence of wheezing than those who had not been breastfed. The effect persisted for the 7 years of the study in nonatopic children. The authors attribute this, in part, to breastfeeding’s protective effect against respiratory illness. They did not distinguish minimal from prolonged breastfeeding.

In a 17-year prospective study of 150 healthy children, researchers did consider length of breastfeeding. The three groups had been breastfed: less than 1 month or not at all, 1 to 6 months, or more than 6 months. Prolonged breastfeeding was associated with the least eczema at 1 to 3 years, as well as fewer food and respiratory allergies. At age 17 years, the trends continued, leading the authors to conclude that breastfeeding is protective against atopic eczema, food allergy, and respiratory asthma throughout childhood and adolescence ( Figures 17-2 and 17-3 ).

Prevalence of atopic eczema, food allergy, and respiratory allergy in full cohort of initial 236 children during follow-up for 17 years.

(Modified from Saarinen UM, Kajossari M: Breastfeeding as prophylaxis against disease: prospective follow-up study until 17 years old, Lancet 346:1065, 1995.)

Prevalence of atopic eczema, food allergy, and respiratory allergy in infant feeding groups during follow-up for 17 years. Tests for differences during the appropriate age periods (eczema 1 to 3 years, food allergy 1 to 3 years, respiratory allergy at 5, 10, and 17 years) were done by analysis of variance and covariance with repeated measures.

(Modified from Saarinen UM, Kajossari M: Breastfeeding as prophylaxis against disease: prospective follow-up study until 17 years old, Lancet 346:1065, 1995.)