Correspondence
Peanut Allergy
N Engl J Med 2003; 349:301-303July 17, 2003
- Article
To the Editor:
Lack et al. (March 13 issue)1 used a case–control design to obtain retrospective data showing a relation between topical application of creams containing peanut oil (but not creams that did not contain peanut oil) and the development of peanut allergy. Presumably, the creams were used to treat eczema. Since the patients with peanut allergy and the controls (a healthy group and an atopic group without peanut allergy) were not matched, it is possible that the patients with peanut allergy used creams more often and used a larger number of different creams. Indeed, the prospectively acquired cohort data showed a strong relation between the severity of eczema and the risk of peanut allergy. It is possible that the children with peanut allergy had more severe and troubling eczema, that creams were used more often, and that, therefore, a greater number of topical preparations were used. Clearly, the more creams that were used, the greater the likelihood that some of those creams contained peanut oil. There may be a relation between the topical use of peanut oil and peanut allergy simply because both are related to atopic eczema.
1 ReferencesJohn B. Ziegler, M.D.
Sydney Children's Hospital, Randwick, NSW 2031, Australia
j.ziegler@unsw. edu. au 1
Lack G ,Fox D ,Northstone K ,Golding J . Factors associated with the development of peanut allergy in childhood. N Engl J Med 2003;348:977-985
Full Text | Web of Science | Medline
To the Editor:
The article by Lack et al. raises several red flags. First, the authors do not specify the type of peanut oil used in their clinical studies. In England, the peanut oil is primarily crude oil that contains peanut protein, and it is the protein in peanuts that produces a reaction in peanut-sensitive persons. This might explain the reactions in the patients in this study.
In the United States, all peanut oil is processed in the same way. According to the manufacturer (Standard Brands), this processed peanut oil has a very consistent proximate composition: 100 percent fat, with no detectable protein, carbohydrate, water, ash, or fiber, and only trace amounts of minerals. The article by Lack et al. may arouse concern about peanut-oil products even though many such products have established safety profiles.
Jerry S. Roth
Hill Dermaceuticals, Sanford, FL 32773To the Editor:
We believe that the choice of controls by Lack et al. may have exaggerated the association between exposure to soy protein and peanut allergy. When the analysis was performed with the use of randomly selected controls, exposure to soy protein was independently associated with peanut allergy. However, children who have food allergies are more likely than children who do not have such allergies to have been given soy formula because of atopic dermatitis, gastroesophageal reflux, or immediate hypersensitivity to a food. The authors' results may therefore reflect a general association between food allergy and exposure to soy protein, rather than a specific association with peanut allergy.
Although Lack et al. addressed this bias by performing the analysis using controls with atopic dermatitis, controls with food allergy should be used in order to determine whether there is a peanut-specific risk associated with exposure to soy products. The authors did state that neither milk allergy nor egg allergy was associated with peanut allergy but did not discuss whether either was associated with exposure to soy products. An association between egg allergy and exposure to soy products would underscore the potential importance of the selection bias.
Elizabeth C. Matsui, M.D.
Robert A. Wood, M.D.
Johns Hopkins University, Baltimore, MD 21287
ematsui@jhmi.edu To the Editor:
We would like to suggest another possible source of peanut allergy. Vaginal progesterone suppositories, which may be used in early pregnancy, have often contained peanut oil. The effect of this maternal variable may merit further investigation.
Dorothy H.B. Wilson, R.N., M.S.N.
2 Drake Ct., Wyoming, DE 19934
salmondot@aol.com Samuel M. Wilson, M.D.
Kent General Hospital, Dover, DE 19901To the Editor:
Lack et al. identify exposure to skin creams containing peanut oil as a significant risk factor for peanut allergy in children. In the accompanying editorial,1 Metzger outlines approaches to the management of the risk of peanut allergy. The Food and Drug Administration (FDA) has been aware of the concern about peanut oil and has been evaluating options for reducing the risk. There is an additional strategy that may lessen the burden of peanut allergy: the removal of allergens from peanut oil destined for use in therapeutic products. Heating peanut oil at high temperatures during refining may reduce the levels of residual proteins sufficiently so that the oil would not cause allergic reactions.2,3
Current requirements of the National Formulary for refined peanut oil4 do not specify a heating step. Since many drug and cosmetic products contain “peanut oil, NF,” the addition of a sufficient heating step to the National Formulary requirements could have a public health benefit. The review division of the FDA contacted Larry Ouderkirk of the compendial operations staff of the Office of Pharmaceutical Science, who solicited the aid of the International Pharmaceutical Excipients Council of the Americas in the effort to ensure that current compendial standards for peanut oil are sufficient to ensure safety.
The International Pharmaceutical Excipients Council has evaluated the current compendial standards along with the Institute of Shortening and Edible Oils. As a result, the United States Pharmacopeial Convention has given notice that a revised monograph for peanut oil, NF, will be proposed in the Pharmacopeial Forum newsletter, requiring full refinement of peanut oil including heating to 230 to 260°C. The identification of an opportunity for upgrading the compendial requirements for peanut oil in order to enhance public safety emerged from the close, collaborative drug-review process involving clinicians, pharmacologists, and chemists at the Center for Drug Evaluation and Research of the FDA, in cooperation with the partner organizations mentioned above.
4 ReferencesJonathan K. Wilkin, M.D.
Ernest G. Pappas, B.S.
Wilson H. DeCamp, Ph.D.
Food and Drug Administration, Rockville, MD 208571
Metzger H . Two approaches to peanut allergy. N Engl J Med 2003;348:1046-1048
Full Text | Web of Science | Medline2
Hourihane JO ,Bedwani SJ ,Dean TP ,Warner JO . Randomised, double blind, crossover challenge study of allergenicity of peanut oils in subjects allergic to peanuts. BMJ 1997;314:1084-1088
CrossRef | Web of Science | Medline3
Teuber SS ,Brown RL ,Haapanen LAD . Allergenicity of gourmet nut oils processed by different methods. J Allergy Clin Immunol 1997;99:502-507
CrossRef | Web of Science | Medline4
The National Formulary. USP 26-NF 21. Rockville, Md.: United States Pharmacopeial Convention, 2003:2805.
Author/Editor Response
Dr. Ziegler suggests that children with peanut allergy have more severe eczema and therefore use more creams than children with mild eczema. However, most of the preparations containing arachis (peanut) oil were used for treating diaper rash. Preparations for eczema were used regularly as emollients, rather than for exacerbations, and are used similarly in children with eczema of varying severity. There was a selective increase in the use of preparations containing peanut oil in the group of patients with peanut allergy. The level of exposure to preparations not containing peanut oil was equivalent in all groups of children. The level of usage of corticosteroid creams on the skin was equivalent in the atopic control group and the group with peanut allergy.
Our study was a birth-cohort study, in which comparisons were made between patients with peanut allergy and the entire cohort with regard to prospectively collected information. Nested within this study was a case–control study involving retrospectively collected information. We did not match controls to patients with peanut allergy because the potential factors for matching were allowed for in the regression analysis.
Refined peanut oil is used in products for the skin in the United Kingdom as in the United States. Mr. Roth states that only crude oil contains protein and that refined oil does not elicit allergic reactions. Although the ingestion of refined peanut oil will not elicit allergic symptoms in persons with allergy,1 the smaller quantities of protein present (as small as picograms) may lead to sensitization through the skin. One must distinguish between “no detectable protein” and “no protein present.” Until recently, enzyme-linked immunosorbent–assay techniques for the detection of peanut protein had a limit of detection of one part per million; this limit may be considerably higher than the levels of peanut protein required for sensitization. Even refined peanut oil heated at high temperatures contains detectable protein, albeit at reduced levels.2
Dr. Wilkin and colleagues suggest that heating peanut oil may reduce the level of protein enough so that it would not cause allergic reactions. However, the question remains whether these levels would be below the threshold for sensitization. The products used by patients in our study contained refined peanut oil, which is heated. Furthermore, excessive heating may enhance the allergenicity of the residual protein.3
Drs. Matsui and Wood suggest that the association between the consumption of soy products and peanut allergy may have been confounded by atopy and other food allergies. Information on soy consumption was obtained prospectively, as were details of other food allergies. Although children with milk and egg allergies were more likely to drink soy milk, their consumption of soy milk was initiated after these allergies were suspected. In contrast, in 90 percent of instances, reactions to peanut products occurred only after the soy milk was introduced. The association between soy consumption and peanut allergy could not be explained away by entering the variables for a history of allergy to cow's milk, allergy to eggs, and eczema into the regression analysis.
Wilson and Wilson suggest a new source of sensitization. In our study, maternal use of breast creams containing peanut oil, which was indirectly ingested by babies who were breast-fed, was not associated with peanut allergy. There was an association only with oils that were applied to the infant's skin. Thus, the cutaneous route of exposure seems to be the important determinant for allergic sensitization.
3 ReferencesGideon Lack, M.B., B.Ch.
St. Mary's Hospital, London WZ 1NY, United KingdomKate Northstone, M.Sc.
Jean Golding, Ph.D.
University of Bristol, Bristol BS8 1TQ, United Kingdom1
Hourihane JO ,Bedwani SJ ,Dean TP ,Warner JO . Randomised, double blind, crossover challenge study of allergenicity of peanut oils in subjects allergic to peanuts. BMJ 1997;314:1084-1088
CrossRef | Web of Science | Medline2
Teuber SS ,Brown RL ,Haapanen LAD . Allergenicity of gourmet nut oils processed by different methods. J Allergy Clin Immunol 1997;99:502-507
CrossRef | Web of Science | Medline3
Maleki SJ ,Chung SY ,Champagne ET ,Raufman JP . The effects of roasting on the allergenic properties of peanut proteins. J Allergy Clin Immunol 2000;106:763-768
CrossRef | Web of Science | Medline
