In the Sept. 23, 1980 Federal Register, the National Health Care Technology requested opinions from interested parties concerning the following techniques:
(1) Cytoxicity Testing (Bryan's Test)
In response to this request, the Executive Committee and the Committee on Unproved Techniques have prepared the following position statements, and these have been submitted to the National Center for Health Care Technology.
Robert E. Reisman, M.D.
Leukocytotoxic testing is based on the claim that the addition of specific allergen in vitro to whole blood or to serum leukocyte suspensions will result in reduction in whiteblood cell count1 or death of the leukocytes.2-6 The test has been described and critically reviewed by Golbert.7 Although it has been claimed to be useful for diagnosis of both food and inhalant allergy by a number of physicians,1-6 the test has never been proved effective by controlled studies, nor has a scientific basis for its use been demonstrated.7-15 On the other hand, the results of numerous published controlled trials indicate that the procedure is not effective for diagnosis of food and inhalant allergy. Franklin and Lowell10 reported that in tests of blood obtained from ragweed-sensitive patients the white blood cell count was the same in tubes of blood containing ragweed pollen extract as in tubes of the same blood containing saline. Chambers et al11 reported that in serum-leukocyte preparations from patients sensitive to a variety of specific allergens there were no consistent differences between leukocytes exposed to allergens to which the patients were clinically sensitive and those exposed to allergens to which the patients were not sensitive. Lieberman et al,12 in a controlled study of the cytotoxic effect of specific allergens on white cells in plasma suspensions, found that the test did not correlate with atopic reactions to foods or with other untoward reactions to foods(headache, diarrhea, and fatigue) and that the test was time-consuming, dependent on subjective interpretation, and inconsistent in results when repetitive runs were performed on the same patient. Benson and Arkins,13 in a double-blind controlled study, found that a similar cytotoxic test afforded no reliable help in establishing the diagnosis of food allergy because positive cytotoxic effects were frequently obtained to foods that produced no clinical symptoms, and negative cytotoxic reactions were obtained to foods that did produce clinical symptoms. Lehman,14 in a double-blind study, also found the leukocytotoxic test to be ineffective for diagnosis of food allergy.14
Autogenous urine injection therapy for allergic diseases was introduced by Plesch1. He reported that he collected fresh urine aseptically in sterile containers, sterilized it by filtration and/or by boiling, and injected at various time intervals 0.25 to 5.0 ml intramuscularly into the same patient who had excreted the urine. The injections were noted to be associated sometimes with local redness and swelling that cleared within 30 hours, fever, intense psychologic depression, sneezing, dyspnea, diarrhea, vomiting, and with fall of blood pressure of as much as 30 mm Hg. He did not specify the frequency with which these reactions followed injections. The data consisted of brief case histories that concerned 12 patients. He presented scanty and entirely anecdotal evidence of relief of symptoms of diverse illnesses, including jaundice; post-hepatitis weakness, depression, constipation, and flatulence; eczema; asthma; hay fever; urticaria; distention, and food allergy: migraine, abdominal distention and flatulence; rheumatic heart disease with mitral insufficiency, pulmonary edema, and asthma; incessant sneezing, with eye itching and tearing; asthma; and pregnancy, abdominal pain, distention, and flatulence. Treatment of the last patient was associated with abortion. Plesch presented no evidence from suitable controlled studies with reference to the efficacy of this treatment and provided no valid scientific basis for the treatment.
We have found no article published on this subject in a scientific journal since Plesch's article. We are familiar with a number of articles on injection of autogenous urine peptones, which appeared between 1927 and 1937 and which were reviewed by Vaughan and black,2 but we find this treatment to be somewhat different from autogenous urine injection and it is probably never employed today.
Autogenous urine injection therapy continues to be used today and to be advertised to the public.
In 1968 Hawkins and Cochran3 reported that rabbit urine contained antigenic material immunologically similar to rabbit glomerular basement membrane (GBM). Subsequently, Lerner and Dixon4 induced acute glomerulonephritis in rabbits with anti-GBMantibodies. In addition, others have shown that renal tubular antigen given in minute quantities by injection in the foot pad of a rat can cause autologous immune complex disease of the rat resulting in glomerulonephritis5.
McPhaul and Dixon6 have further shown that man like rabbit excretes antigens immunologically related to soluble antigens from normal human GBM. Furthermore the antigens in human urine appear identical in different normal adults, and they are capable of evoking anti-GBM antibodies in heterologous species (sheep).
In human disease states, anti-GBM antibodies are present in the kidneys of all cases of Goodpasture s syndrome and in approximately 50% of humans with subacute glomerulonephritis. Eluting these antibodies from the nephritic human kidney and then transferring the antibody to monkeys has resulted in glomerulonephritis in the monkey. Circulation of anti-GBM antibodies in humans has also resulted in glomerulonephritis in allografted kidneys as well. It is conceivable that urinary GBM antigens given to the human subject would initiate production of anti-GBM antibodies capable of causing autologous immune complex disease and nephritis3.
In view of this background, the California Society of Allergy and Clinical Immunology and the California Blue Shield came forth with the following opinion with the help of the Blue Shield Advisory Panels on Allergy, Internal Medicine, and Dermatology.
It is the unanimous opinion of members of the Advisory Panels on Allergy, Internal Medicine, and Dermatology that urine injections are not acceptable medical practice in the treatment of allergic diseases. Advisory Panel members agreed with the position paper on this subject prepared by the California Society of Allergy and Clinical Immunology. They also emphasized that this treatment modality must be considered experimental until controlled studies are done proving its benefit or harm and further, that physicians using urine injection therapy should be encouraged strongly to do such studies.
The Rinkel method of skin titration is one method of skin test end point titration. The technique is well described by Rinkel1,2 and Willoughby.3 Briefly, the end point dilution is determined by intradermal testing with use of approximately 0.01 ml of serial fivefold dilutions of allergenic extract. It is the weakest dilution that produces a positive skin reaction and initiates progressive increase in the diameter of the wheals with each stronger dilution tested.
Rinkel claimed that the method was useful for determination of starting dose and therapeutic dose, that a safe starting dose was 0.01 to 0.15 ml of end point dilution, that the optimal therapeutic dose was close to 0.5 ml of end point dilution, that the specified starting dose rarely, if ever, caused untoward reactions, and that the specified optimal therapeutic dose rarely, if ever, caused untoward reaction and in many cases promptly relieved symptoms.
The method was critically reviewed by Golbert4 in 1975, who pointed out that the procedure was not proved. This view was concurred in by other authorities.5 Since then, several controlled studies have been accomplished, allowing more definitive conclusions to be made.6-8
The end point skin titration method of Rinkel provides essentially the same information as that provided by other well-studied skin-test end point titration methods, which have been reviewed by Norman et al.9 Suitably controlled studies show that it is a satisfactory method for quantifying skin sensitivity to ragweed pollen extract6-8 and for obtaining information about sensitivity of patients who are highly sensitive to ragweed. The reliability of the method is comparable to that provided by in vitro leukocyte histamine release,6,7 radioallergosorbent tests6,7 and skin test end point titration by Norman technique with Ragweed AgE.6,7 Patients shown by the Rinkel method of skin titration to have a high sensitivity to ragweed pollen extract with a skin-test end point titer of 56(1:312,500 w/v or more) can be expected to have an excellent correlation between symptom-medication scores and ragweed pollen counts during the ragweed pollen season.6,7
These studies show that the method is a rough guide to safe starting dose. For titers of 56 or more a starting dose is 0.1 ml of the 56 dilution. For titers below56 (55,54 and so on) a safe starting dose is 0.1 ml of the end point dilution. However, for the highly sensitive patient with a skin-test end point titer of 57 (1:1,562,500 w/v) or more (58, 59 and so on) the dose predicted by the Rinkel method would be less than the starting dose of 0.1 ml of 1:250,000 to 1:500,000 w/v, which has proven to be tolerated by most patients6,7. The Rinkel method would have predicted a starting dose too low for 45 of 52 patients in one study7 and would have resulted in waste of time. As a practical matter, titrations with concentrations more dilute that 5-7 may be unnecessary for routine purposes.7
The recent controlled studies deal with highly sensitive patients because such patients have been shown to have a good correlation between the rise and fall of symptom-medication score and rise and fall of pollen counts. Patients with low sensitivity to ragweed pollen counts were not selected because they have been previously shown to have a poor correlation between symptom medication scores and ragweed pollen counts.7 Indiscriminate immunotherapy of patients with low skin sensitivity (end point 5-1 to 5-3) should be discouraged in the absence of strong clinical evidence of sensitivity.
The controlled studies provide no support for Rinkel's claim that the usual optimal dose for providing relief from hay fever symptoms was 0.5 ml of the end point dilution (see page 214 of ref. 2 and page 608 of ref. 3.) In the patient highly sensitive to ragweed, 0.5 ml of the end point dilution of ragweed pollen extract has been shown to be no more effective than placebo for relieving symptoms of hay fever6-8 or for producing immunologic effect.6,7 It has also been shown to be much less effective than larger doses,7 which were about 1000 times larger (3.7 to 46.8 mg of ragweed AgE) and produced significant reduction in symptom-medication scores, increase in serum levels of antiragweed IgG, and decrease in the expected seasonal rise in serum levels of antiragweed IgE.7 These studies indicate that some measurement of patient sensitivity, such as skin-test end point titration (by Rinkel or other standard methods), may be useful at the onset of immunotherapy to provide a guide to initial dose. Thereafter, costly repetitive end point titrations are usually unnecessary because regardless of what the titration indicates, the dose will be advanced either until the patient can tolerate no more or until a dose is reached that produces satisfactory results and has been shown in controlled studies to produce an effect superior to placebo.7
Subcutaneous provocation-neutralization testing may be defined as a technique for the diagnosis and treatment of allergic disease in which a subcutaneous injection of antigen of sufficient quantity is administered to elicit symptoms corresponding to the patient complaints. This is followed by the immediate injection of weaker or stronger dilution of the same antigen to relieve the provoked symptoms.
Since first description of the test,1 a number of articles have appeared in the literature supporting its use and describing variations of its application.2-5 This literature is descriptive and efficacy data are anecdotal. Miller6 reported a favorable result from subcutaneous food injection therapy after studying eight subjects on a double-blind crossover basis. Critical review of this article questions whether the wide variety of symptoms observed could stem from a single source. The scoring of symptomatic improvement and the subsequent statistical analysis are open to criticism. On the other hand, Crawford et al.7 performed a double-blind study of 61 atopic subjects to five common foods and was unable to confirm the validity and reproducibility of results from subcutaneous food testing. Bronsky et al.8 also performed a study of symptoms, chest auscultation, pulse, electro-cardiographic patterns, white blood cell fluctuations, and peak expiratory flow rates on 20 asthmatic children after provocation skin testing with foods and found no correlation of these measurements with skin tests. In addition, Draper9 attempted to correlate deliberate feeding tests with intradermal provocative tests in 121 subjects. He found that 38% of the positive provocative tests could be confirmed by deliberate feeding, and therefore verification of provocative skin tests results by deliberate feeding tests is necessary. Kailin and Collier10 also reported a double-blind trial comparing the relieving effect of subcutaneous or sublingual administration of saline vs. food extracts in patients who were previously thought to have shown an adequate response to a relieving dose of food extract. Treatments with active extracts were correctly judged to be active in 24 of 34 trials (70.6%). Saline, however, was judged to be the active extract in 28 of 40 trials (70%), thereby failing to show a difference in the results between saline and aqueous food extracts. Furthermore, despite repeated claims of greater efficacy than that of current conventional methods, no studies comparing this form of diagnosis and treatment with present methods have been forthcoming.
No attempts at scientific establishment of the possible mechanisms involved have been published. From what is currently known about the time sequence of verified immunologic therapeutic response from a relieving or neutralizing dose of allergenic extract based on skin testing. Present conventional methods of diagnosis of food allergy by blinded techniques have been shown to be efficacious.11
The use of sublingual antigen administration as a means of treatment for atopic disease was first suggested by Hansel1 in 1944. The advocates of this technique claim that it can be used for diagnosis and treatment of food-induced respiratory, gastrointestinal, and other systemic symptoms. The method, as currently adopted for diagnosis and treatment of atopic disease, consists of placing three drops of 1:100 (w/v) aqueous extracted andglycerinated allergenic extract under the tongue of the patient and waiting 10 min for the appearance of symptoms2. When the physician is satisfied that he has determined the cause(s) of the symptoms, he then administers a neutralizing dose, which is usually three drops of a dilute solution (e.g. 1:300,000 w/v) of the same extract. The symptoms are expected to disappear in approximately the same temporal sequence in which they appeared. If the neutralizing dose is given prior to challenge (e.g. eating a meal containing the offending food), the prevention of symptoms is also expected.
In recent years a number of anecdotal reports concerning the clinical effectiveness of sublingual antigen administration in the diagnosis and treatment of food allergy have appeared.1-5 None of these reports was based on prospective, double-blind, well controlled protocols, and the ameliorative effects in these clinical case studies were not substantiated by objective scientific evaluation of data.
On the other hand, controlled studies have been performed independently by several investigative teams.6-9 The Food Allergy Committee of the American College of allergists evaluated the use of sublingual provocative testing in 19736 and again in 1974.7 These studies revealed that sublingual provocative food testing did not discriminate between placebo controls and food extracts. As a result of these double-blind, multi center, controlled investigations, this committee could not recommend the sublingual provocative technique as a reliable indicator of food allergy. In a separate investigation Kailin8 evaluated the effectiveness of sublingual provocative testing by five different physicians, all of whom had been using this method of testing for at least 7 yr. These physicians were allowed to select patients previously tested and known to have positive reactions. However, the sublingual provocative technique was performed according to a double-blind protocol. There was no distinction of reactions between placebo and active food extracts. Recently, Lehman9 obtained similar negative results. He reported that changes of symptoms or signs occurred as frequently after sublingual provocation with placebo as with food extract. Critical reviews of sublingual provocation and desensitization techniques are also available from other authoritative sources, none of which could support the validity of such treatment.10-13
JACI, Vol. 67, No. 5, May 1981, pp. 333-338