Add as bookmark

Eliminating and IgG Food Allergy Testing

by Dr Sidney MacDonald Baker, MD(more info)

listed in allergy testing, originally published in issue 35 - December 1998

Urinating, defecating, combing, picking, spitting, vomiting, sweating, trimming, shaving, bathing, washing, scrubbing, cleaning, shampooing, flatulating, weeping, expressing, draining, coughing, sneezing, blowing, paring, cutting, wiping, rubbing, scratching, belching, burping, farting. Am I confusing you?

One of the most entertaining challenges of my daily life as a physician is choosing words that avoid confusing my patients. The vocabulary of elimination requires the most delicate adjustment of my words to my patient's age, geographic and social origins. I prefer to shoot low, getting the sympathy of my youngest patients and avoiding unnecessary confusion among the elder by following the rule that clarity is served by sliding my language toward the more vulgar end of the scale. Meanwhile I have accumulated a repertoire of euphemisms upon which I can draw when confronted by quizzical looks that seem to say, "What do you suppose the doctor is talking about?"

The problems of ambiguous language surround a subset of all the words we use to describe ridding our bodies of unwanted substances. Especially the kinds of ridding that are least acceptable in public for those of us who have graduated to the loo from the nappy, to use two expressions that would be incomprehensible to most of my American patients.

Ridding ourselves of unwanted substances has always been a part of medical conversation where one describes the details of activities that would be unmentionable in any other social setting. I offer you the above list of detoxifying activities to remind you that removing offensive material from our bodies is a generalised activity that actually requires a fair amount of attention. If we include materials of which we are usually unaware at the conscious level such as our bodily excess of acid, nitrogen, an abundance of used molecules formed as by-products of our normal metabolism, and toxins and allergens that enter our body, we find that our daily energy expenditure on bodily sanitation is quite enormous, quantitatively overshadowing what we spend on apparently more noble activities such as thinking and moving about.

Detoxification is so costly that it should be a concern for anyone who feels unwell or stops to think about preserving health against the tide of sickness and age. However global and important the removal of unwanted materials from our body, we do not have a medical specialty of eliminology that applies scientific principles to guide us in our efforts to be more efficient in our varied efforts at bodily sanitation. Perhaps the closest we have to such specialists are those of us doctors who ask two simple questions when pondering all the ways to help our patients recover and preserve health:

1 "Could this patient have unmet needs to get some nutritionally helpful substance for which he or she may have a deficiency or a special need?"

2 "Could this patient have unmet needs to avoid or eliminate some toxin or allergens?"

Avoid is the key word in the second question. If our bodies are fairly burdened by the metabolic costs of detoxification, it makes sense to stay as clean as possible. Wash your hands before dinner. Stay away from poisons. Eliminate allergens from your diet. Now here is a new spin on the notion of elimination. Elimination diets mean getting rid of the offending substance before it has a chance to become a burden to your body's costly chemistry of sanitation. My medical training more than 30 years ago did not place a high priority on the notion that food intolerance could be bothersome to many people. More than three decades of medical practice has led me to another conclusion: most of us suffer some consequence from eating apparently healthy foods against which our immune system has taken a negative attitude.

One way to discover whether or not the elimination of one or more foods may be the key to feeling better is to just go ahead and try avoiding them assiduously for at least 5 or, better, up to 14 days. But which foods should you consider taking out of your diet? You could guess. A good guess would identify foods that you eat frequently, common allergens such as cereal grains and milk products, leavened or fermented foods, citrus, spices, egg, coconut, fish, tea, coffee, and so on. But what if you go to all that trouble and miss something that just happens to be your problem, such as potato or tomato or peppers? The point is that some sort of test can be quite useful in sorting it out to the extent of giving you a short list of likely possibilities.

There is no perfect test. None will identify foods that definitely produce symptoms in you so that you would want to stay away from such a food indefinitely. The test I find the most useful measures IgG (Immune Globulin, type G) antibodies to foods with the idea that an abnormally high level of antibody is an indication that your immune system has taken a strange posture with regard to that food. When done at a laboratory with reliable methods IgG ELISA test is valid, and in certain clinical situations is essential to a medical practice involving patients with complex chronic illness.

Here is how I look at the science behind my statement.

Basic science

The Immune System, like the Central Nervous System, functions to perceive and remember environmental stimuli that we encounter in sickness and in health. The realm of the Immune System is the invisibly small world of molecules and cell surfaces. We use the word recognition (knowing again) to specify the processes involved in perceiving substances at the molecular level as well as in the realm of our central nervous system's daily conscious experience of the world around us. Recognition at both levels involves the concerted use of varying modalities that may be appropriate to apprehend a given stimulus. Thus I recognise my wife by her face, her voice, the touch of her hand, and her perfume and never have I thought of specifying any particular modality as the mediator of my perception. Eye-witnessing is considered more reliable than ear-witnessing, but we all agree that it is unreasonable to reject any kind of credible evidence a priori, as we try to take in the world around us.

Under normal circumstances immune perception takes place below the level of consciousness so we are less attuned to the setting in which antibodies, complement, T-cell and macrophage processing, and various cytokines are orchestrated to recognise a given antigen and, at times, bring that recognition to a conscious level with the production of symptoms such as fever, inflammation, itching, or nausea. It is scientifically reasonable, however, to insist that immune mechanisms, like conscious perceptions, are orchestrated in flexible and subtle ways depending on the situation at hand. It is unreasonable to insist that any one mechanism, such as IgE, for example, is the sole reliable mediator of a whole class of immunological experiences. I make no claim in this review that IgG is the mediator of delayed food reactions but there is abundant evidence that it is a clinically reliable marker for such reactions, provided analytical competence in the laboratory.

I accept as scientific fact, however, that to the extent that IgG is a mediator of immune reactions it is (apart from T-cell malignancies) invariably specific to a given antigen so that when high levels of specific IgG are detected there must be a reason for the excess that has to do with the relationship between the organism and the specific antigen or a coincidentally cross reacting antigen.


1 Insulin-Dependent Diabetes Mellitus (IDDM) & IgG anti-Bovine serum albumin (BSA)

Part of the debate over IgG antibodies has revolved around the question implied by the statement that: "IgG antibodies simply reflect what the patient has been eating and have nothing to do with symptoms." If there are instances in which IgG antibodies are clearly associated with a distinctive pathology, then the preceding statement does not hold water. It matters not whether the IgG antibodies are "the cause" or what might more properly be referred to as a marker for a pathological condition, measurement of them is clinically appropriate in assessing a given individual. Consider then the literature on insulin dependent diabetes mellitus (IDDM) and increased levels of cow's milk antibodies.3,4,5,6,7,8,9 The evidence for an association between cow's milk IgG BSA antibodies and IDDM is strong. Its pathophysiologic role in autoimmune mechanism as described in the literature is well beyond speculation, so that a prudent clinician, weighing the present evidence, would have difficulty crediting the notion that IgG antibodies in this situation are an innocent reflection of what a particular infant has been eating. Should a prudent clinician, dealing with a family with IDDM be deprived access to reliable testing for IgG antibodies to bovine serum albumin which may represent the target antigen for islet cell autoimmunity? Pending the complete elaboration of the IDDM – cow's milk story, should the prudent physician not be permitted to conduct tests that may be speculative from the standard of pure scientific proof but are a very good bet on behalf of his or her patient? Is the IDDM-BSA story in itself not sufficient to rebut all claims that IgG antibodies are just an innocent reflection of dietary intake or some sort of immunologic bystanders?

2 Inflammatory Illnesses

Serum IgG antibodies that react with different dietary proteins have been detected in a significant proportion of adult patients with coeliac disease,[10] Crohn's disease and ulcerative colitis,[11] dermatitis herpetiformis and atopic eczema.[12] In studies of coeliac disease, the focus on serologic immune function is mainly concerned with IgG antibodies which are assumed in all of the literature to be pathological and presumed to be of significant diagnostic value.[13],[14] Coeliac disease, by itself, may be so widespread and so occult in its manifestations[15] that dependence on IgG testing is not only clinically appropriate but probably should be provided for a wide spectrum of patients. Authoritative articles cite the role of IgG in both sensitising and protective functions.[16] The gluten sensitivity literature alone should be sufficient to lay to rest the claim that IgG is of no pathologic significance with respect to food intolerance. In any given individual with such a condition a clinician may be free to speculate that the presence of antibody is innocent, is the result of increased gut permeability, or is sensitising and therefore an indication that the patient deserves a diagnostic trial of specific food avoidance that takes into consideration the pattern of IgG antibody responses. Should a conscientious physician be denied the opportunity to give a full answer to his or her patients' inquiries as to the possible role of diet in their conditions?

3 Food allergy per se

"Children who exhibited slowly appearing untoward reactions to cow's milk feeding had significantly higher titres of IgG antibodies against both native and digested beta lactoglobulin than children with an immediate type of reaction or no intolerance."[17] This statement represents the consensus among those who recognise that delayed food sensitivity has little to do with IgE and correlates with IgG. The literature on IgG sub-class antibodies to foods in patients with specific food intolerance makes it abundantly clear that IgG is involved in pathologic responses to foods in some patients and may be irrelevant in patients selected for study on the basis of their IgE mediated immediate food reactions.[18] The literature is also clear that IgG is not invariably related to delayed (non IgE) food intolerance. Shakib et al reported[19] that an ELISA for IgG antibodies to milk identified milk intolerance causing eczema in six patients, asthma in one and both asthma and eczema in one. While failing to identify milk induced eczema and asthma in a total of eleven other subjects, the authors conclude that the presence of high levels of IgG antibody to the three milk proteins was diagnostic of milk intolerance because such levels were not observed in twenty patients with inhalant allergy and thirty nine out of forty blood donors. A 1993 study[20] of food and house dust mite antigen in atopic dermatitis showed that anti beta lactoglobulin and soy protein IgG levels were elevated in patients who improved when placed on the appropriate elimination diet. The authors propose that specific IgG antibody concentrations against food be determined as a means of predicting the response to an exclusion diet. The practising physician is left to conclude that IgG mediated food intolerance is worth considering in diagnostic puzzles and that it would be negligent to favour blanket rejections of IgG mechanisms over descriptions in the peer reviewed literature that clearly indicate the pathologic significance of food specific IgG. If there is a controversy in which an old view is being replaced by a new one, should the practising physician be held to a standard that demands waiting for a shift in public policy before being permitted to apply the new view in his practice? Consider, for example, the controversy over the use of folic acid supplementation for the prevention of neural tube birth defects. A conscientious physician could easily have concluded following the initial reports by Smithells et al[21] in 1976 that it was a good bet for his or her child-bearing patients to take a folic acid supplement. Those who waited nearly two decades for the folic acid story to unfold to the point where public policy favoured supplementation would have done a serious disservice to their patients. All physicians should at least have had the choice of interpreting the scientific literature in a responsible fashion and not have been deprived of access to appropriate clinical behaviour while a very slow and conservative process finally supported Smithells' original observations.[22]

Clinical common sense

In 1985 Cohen et al reported a case of severe anaemia and chronic bronchitis associated with a markedly elevated specific IgG to cow's milk protein.[1] The child in question had no detectable specific IgE to milk but specific anti cow's milk IgG was found at very high levels and symptoms and antibody levels cleared with removal of cow's milk from the child's diet. Can I, as a clinician, ignore such a convincing report published in a peer review journal when confronted with a patient displaying mysterious symptoms that raise the question of a hypersensitivity reaction? Two years following the publication of the Cohen paper, the Council of Scientific Affairs of the American Medical Association published[2] a review of about a dozen allergy diagnostic methods, including fifty words and three references devoted to measurement of specific IgG to detect food allergy. There was no attempt in the superficial review or in the papers cited (two of which date from the 1970s) to confront the obvious contradiction, between observations exemplified by the Cohen paper and the opinion stated by the council and other authoritative medical bodies since that time, that IgG is not a reliable marker for food allergies. It should be noted that the Council Report includes the specific statement: "This report is not intended to be construed or to serve as a standard of medical care. Standards of medical care are determined on the basis of all the facts and circumstances involved in an individual case and are subject to change as scientific knowledge and technology advance and patterns of practice evolve." Does not the present use of IgG ELISA by more than one thousand physicians constitute a pattern of practice that must be considered by anyone attempting, as the Council Report does not attempt, to set a standard of medical care?

As I interact with other physicians at meetings and in courses that I teach I am struck by the passion associated with allergy diagnostic techniques. Various specialty societies scrimmage on the subject with a vehemence more suited to religious or political difference than to scientific discourse. A common sense ear to such debate detects definite tones of ego and economics as well as the understandable difficulty we all have in abandoning a public stance. I believe that the strongest undertone comes, however, from the legacy of the evolution of allergy as a specialty. Just beyond the memory of most practising physicians today was an era in which allergy was an empirical movement within medicine struggling to attain the high ground of scientific credibility. My professors at Yale in the early 1960s were down on it. The only specialty clinic without leadership from the full time faculty was allergy. The field of allergy has evolved, therefore, with a special thirst for academic respectability that can only be quenched by standards of proof that are difficult to attain, considering that the biology of allergy is all tied up with individuality. Properly conducted scientific studies that are based on the assumption that all participants are "the same" with respect to certain variables are especially difficult to conduct in a field in which the very difference of the individual from others' reactions to foods, pollens, mould, etc. is the whole idea. If someone were to try to set a standard of care, what standard of proof would be required to show that IgG ELISA testing is sufficiently valid for use with an individual patient in the care of a qualified physician?

When the subject of IgG ELISA comes up in discussion with colleagues who have tried to set it up in their labs I usually hear statements to the effect that: "We tried to set up IgG testing and just could not get consistent results." Further discussion leads me to the conclusion that my colleagues have encountered significant obstacles at the technical level. I believe that the inter-test reliability of the IgG ELISA performed by Dr John Rebello at Immuno Laboratories reflects his having overcome most of the technical difficulties. Once the technical problems (such as being able to get consistent results in split samples) are overcome there are still biological vagaries of IgG that are bound to make IgG ELISA imprecise in the same way that all allergy tests require common sense and clinical judgement in their clinical application.

Summary and conclusions

In the decade since the publication of the previously cited AMA council report, more than a hundred thousand IgG ELISA tests have been performed on patients of more than a thousand physicians whose interpretation of the published literature and their own clinical experience find these tests to be a cost effective aid to diagnosis of symptoms that may be caused by delayed food hypersensitivity. In addition to helping to find reactive foods, the test also identifies foods that are non-reactive, which, in the judgement of many of my colleagues, may be especially important in the many syndromes we encounter in which there is an overall inappropriate immune activation. It seems appropriate under those circumstances to know which foods are associated with the least evidence of immune reactivity. There is no evidence that the IgG ELISA poses any question of safety. There are certain categories of patients in whom IgG testing is so appropriate to an understanding of the pathogenesis of their problem that I believe it would be negligent to ignore this diagnostic option. One final thought. In discussing my research (on the use of IgG testing to identify foods, which if avoided produce significant symptom relief) with various interested colleagues, several have said that they tried to set up IgG testing in their labs but failed to get consistent results. Question: If a laboratory can demonstrate quality control statistics showing a high degree of reliability and consistency in performance of a test, should that laboratory's special expertise go unrecognised because other labs have difficulties?


1 Cohen G, Hartman G, Hamburger R, O’Connor R. Severe anaemia and chronic bronchitis associated with a markedly elevated specific IgG to cow’s milk protein. Annals of Allergy 55: 38-40. 1985
2 Council on Scientific Affairs. In Vitro Testing for Allergy, Report II of the Allergy Panel. JAMA 258: 1639-1643. 1987.
3 Bruining GJ, Molenaar J, Tuck CW, Lindman J, et al. Clinical time-course and characteristics of islet cell cytoplasmatic antibodies in childhood diabetes. Diabetologia 26: 24-29. 1984.
4 Yokota A, Yamaguchi Y, Ueda Y, Nakanishi T, et al. Comparison of islet cell antibodies, islet cell surface antibodies and anti-bovine serum albumin antibodies in type 1 diabetes. Diabetes Research and Clinical Practice 9: 211-217. 1990.
5 Karjalainen J, Martin J, Knif M, Ilonen J, et al. A bovine albumin peptide as a possible trigger of insulin dependent diabetes mellitus. The New England Journal of Medicine 327: 302-307. 1992.
6 Virtanen SM, Saukkonen T, Savilahti E, Ylonen, et al. Diet, cow’s milk protein antibodies and the risk of IDDM in Finnish children. Diabetologia 37: 381-387. 1994.
7 Pietropaolo M, Cstano L, Babu S, Beulow R, et al. Islet Cell Autoantigen 69 kD (ICA69) Molecular cloning and characterisation of a novel diabetes associated autoantigen. J Clinical Invest 92: 359-371. 1993.
8 Savilahti E, Akerblom HK, Tainio VM, Koskimes S. Children with newly diagnosed insulin dependent diabetes mellitus have increased levels of cow’s milk antibodies. Diabetes Research 7: 137-140. 1988.
9 Saukkonen T, Savilahti E, Vaarala O, Virtala E, et al. Children with newly diagnosed IDDM have increased levels of antibodies to bovine serum albumin but not to ovalbumin. Diabetes Care 17: 970-976. 1994.
10 O’Farrelly C, Kelly J, Hekkens W, Bradley B, et al. Gliadin antibody levels: a serological test for coeliac disease. British Medical Journal 286: 2007-2010. 1983.
11 Paganelli R, Pallone F, Montano S, LeMoli S, et al: Isotypic analysis of antibody response to a food antigen in inflammatory bowel disease. Int. Archs Allergy Appl. Immun. 78: 81-85. 1985.
12 Barnes RMR, Harvey M, Blears J, Finn R, Johnson PM. IgG subclass of human serum antibodies reactive with dietary proteins. Int. Archs Allergy Appl. Immun 81: 141-147. 1986.
13 Kieffer M, Frazier PJ, Daniels NWR, Coombs RRA: Wheat gliadin fractions and other cereal antigens reactive with antibodies in the sera of coeliac patients. Clin. Exp. Immunol. 50: 651-660. 1982.
14 Finn R, Harvey MM, Johnson PM, Verbov JL, et al. Serum IgG antibodies to gliadin and other dietary antigens in adults with atopic eczema. Clinical and Experimental Dermatology 10: 222-228. 1985.
15 Catassi C, Ratsch IM, Fabiani E, Rossini M, Bordicchia F, Candela F, Coppa GV, Giorgi PL. Coeliac Disease in the year 2000: exploring the iceberg. Lancet 343: 200-203 1994.
16 Perelmutter L. IgG4: Non-IgE mediated atopic disease. Annals of Allergy 52: 64-70. 1984.
17 Fallstrom SP, Ahlstedt S, Carlsson B, Lonnerdal B, Hanson A. Serum antibodies against native, processed and digested cow’s milk proteins in children with cow’s milk protein intolerance. Clinical Allergy 16: 417-423. 1986.
18 James J, Sampson H. Immunologic changes associated with the development of tolerance in children with cow milk allergy. Journal of Pediatrics 121(3): 371-377. 1992.
19 Shakib F, Morrow Brown H, Phelps A, Redhead R. Study of IgG sub-class antibodies in patients with milk intolerance. Clinical Allergy 16: 451-458. 1986.
20 Casimir GJA, Duchateau J, Gossart B, Cuvelier PH, et al. Atopic dermatitis: role of food and house dust mite allergens. Pediatrics 92: 252-256. 1993.
21 Smithells RW, Sheppard S, Schorah CJ. Vitamin deficiencies and neural tube defects. Arch Dis Child 51(12): 944-50. Dec 1976.
22 Folic acid for the prevention of recurrent neural tube defects. ACOG Committee Opinion: Committee on Obstetrics: Maternal and Fetal Medicine Number 120 March 1993. Int J Gynaecol Obstet 1993 Jul; 42(1):


  1. Reggie Boshes said..

    Anyways, many thanks for having me in this article.

  2. Daniel said..

    I am interested in furthering our testing for IGG with more reliable testing and quality controls , I would like to speak directly to Dr Baker if at all possible.

    Please email me at

    Thank you ,


« Prev Next »

Post Your Comments:

About Dr Sidney MacDonald Baker, MD

Dr Baker is a practising physician in Weston, Connecticut, with a special interest in the environmental and biochemical aspects of chronic health problems, both in children and in adults. He has published many books, the most recent being Detoxification and Healing: the Key to Optimal Health (Keats Publishing Inc, 1997). In the forward to this thorough yet very accessible guide to the subject Dr Jeffrey Bland describes Dr Baker as "a master teacher in the field of functional medicine and the practice of good medicine".

top of the page