Histamine reactivity of the skin
Histamine is perhaps the most important mediator released during the first stage of the allergic reaction. But this principle can be questioned, since, using the microdialysis technique, Petersen et al. (1) recently found that after skin testing with allergen, histamine was found only just under the center of the wheal, not in the periphery. Therefore, after allergen challenge of the skin/mucosa, the mediators released by the challenged mast cell, which also stimulate closely related c-fibers, induce an axon reflex, and in turn induce release of neurogenic peptides and mast-cell mediators from “the next” mast cell, are probably the major players in the immediate wheal and flare reaction. However, histamine has its immediate effects and induces a visible reaction, similar, but not identical, to the allergen-induced wheal and flare reaction, and that is why it will probably have a place in clinical work even in the future.
In this issue of Allergy, Ronchetti et al. (2) report on the change in histamine reactivity over 13 years in cohorts of Italian children. They performed their study to investigate whether the cause of the increase in skin reactions during the 1980s and 1990s depends on an increase in skin response to histamine rather than a real increase in the number of sensitized patients. To me, this hypothesis does not sound plausible, since it would indicate either a change in the genetics of Italian children or an increase in the nonspecific reactivity of the skin of most children similar to that in the bronchi of asthmatics.
Skin testing should be considered a “laboratory method”
Skin testing should be subject to the same high standards as laboratory methods. Why? And what does it mean? In the laboratory, all methods are validated and continuously supervised, often duplicate tests are also performed routinely, and the variation within and between runs is continuously supervised. Only results fulfilling given criteria are forwarded to the referring clinician. In the clinic, however, most common measures are approximate: the liver/spleen is found to be 2 or 3 cm or so many fingers under the arch, etc. However, when more precise determination of the size of organs is necessary, then ultrasound, radiography, and other methods are used, and the size of the organs is given by the laboratory in centimeters and millimeters. For some reason, when methods are developed purely within the clinic, the same strict rules are not followed, probably depending on tradition. This lack of precise methodology sometimes also applies to clinical methods used in scientific papers and highly influences diagnostic and epidemiologic trials within allergology.
What do histamine and allergen skin prick tests (SPT) express?
Histamine acts directly on the skin tissue components, causing vasodilation, increased blood flow, and edema. Thus, it measures the reactivity of the skin (see below).
Tests for allergen-specific IgE measure circulating IgE antibody, while allergen SPT measures the release of histamine and other preformed mediators, leukotrienes, and prostaglandins produced by the mast-cell wall after IgE/allergen interaction on the cell surface; i.e., it measures the in vivo histamine-releasing effect of the allergen/IgE interaction and the individual sensitivity to those mediators.
Methods for SPT with histamine
Several investigators have compared different SPT methods (3–6). Most of them have found the original SPT method of Pepys (7), which was a major invention when introduced, to be less precise than the “allergy pricker” method of Østerballe & Weeke (8) with a lancet with a 1-mm tip. To obtain the best result, the lancet should be pressed through a drop of test solution at a 90° angle against the skin. If the lancet is jabbed, the same penetration of the skin is not obtained every time, reducing the precision of the method (9).
The figures on the precision (CV) given in the paper by Ronchetti et al. are surprisingly low and must be questioned, since, in some cases, an individual dose response was not found, but a lower concentration gave a larger wheal than a higher concentration of histamine, indicating large deviations from the expected result in some cases. Furthermore, it is not clearly stated that duplicate tests were performed, such tests being a prerequisite for determining the precision.
Surprisingly, the authors have used the Pepys technique with the Allergy Pricker (8), a combined method that has never been documented regarding its diagnostic capacity or precision in relation to other well-documented methods. I can understand their dilemma since the lancet with a 1-mm tip had just been described (8) when they started the first trial. The majority of investigations on different SPT methods were published after the mid-1980s. At that time, since they had started using the described method, it was certainly correct to continue with the same technique in all the trials described in the paper, because that would best ensure results that can be compared over time. Most other investigators, however, would have used the results on histamine reactivity to ensure that the technique was similar in all trials.
Histamine as a standard
The reaction to histamine was earlier used in skin testing as a standard. In 1992, Aas and Belin published the first Nordic proposal for standardization of SPT (10). At that time, there was no information on the allergenic composition and potency of allergenic extracts used for SPT. They needed a basis for comparison of the results of allergen-induced SPT reactions between patients and within patients over time. Thus, they proposed to compare the size of the allergen-induced wheal with that of histamine dihydrochloride at 1 mg/ml. However, for such comparisons, the potency of the extract must be constant. They used a system with a score of 3+ for an allergen-induced wheal of similar size to that of histamine and a score of 1+ less or more for wheals with half and double the area of that of the 3+ reaction, respectively. This method has been widely used for many years. However, since then, we have learned a number of things.
The concentration of histamine used in routine SPT was increased from 1 to 10 mg histamine dihydro-chloride/ml during the 1980s to make it possible for most testing personnel to obtain wheals of measurable size.
The composition of common inhalant allergenic extracts from several manufacturers are nowadays well documented and their potency given in biologic units and/or micrograms of major allergen/ml.
It is recommended that the size of the wheal be given in mm of diameter or mm2 (4, 11, 12). If there is also information on the size of the histamine wheals, including mean diameter or area, confidence intervals and CV in the samples of patients investigated, and the method of SPT employed is clearly defined, the end point, or the concentration eliciting a 3-mm wheal, or the concentration of allergen eliciting a wheal of the same size as that of histamine can be easily calculated. Both of these procedures are recommended.
In the paper by Ronchetti et al. (2) in this issue, the size of wheals is given in mean diameter (mm). This makes it possible to analyze the data, an analysis which is seldom possible when the +++ system is used. Then the referee has no idea of the actual sizes of the wheals and does not know whether they are positive or not. In the present trial, with most of the cases with histamine wheals less than 2 mm in diameter, a ++ reaction would not be possible to measure at all. As further discussed below, the sizes of the histamine wheals, especially in the beginning of this follow-up study, were below the cutoff limit of the method!
Atopy vs histamine reactivity
Stuckey et al. (13) found that the higher the sensitivity of the patient, as measured by the number of positive SPT or total IgE, the larger the histamine wheal.
In a study of children hospitalized for asthma in Lahore, Pakistan, we found histamine wheals up to 18 mm in diameter in children with multiple sensitization and severe asthma (14).
When the degree of allergy is reduced, as by immunotherapy, not only the skin response to the allergen in question is reduced (15), but also that to other allergens (16). Allergen vaccination also reduces the number of mast cells in the skin (17). Furthermore, the histamine reactivity of the skin is also reduced by immunotherapy (unpublished observation).
The finding by Stuckey et al. and the related influence of allergen immunotherapy on skin reactivity indicate that there might be a hyperresponsiveness to histamine in the skin, as found for the bronchi and the nose. This finding influences the use of histamine as a reference.
The fact that the reactivity of the skin is related to the degree of allergy is an important finding, but whether or not it reduces the value of histamine as a reference is not clear at the moment and must be further investigated. This can only be done if the wheal size is transformed to the histamine concentration eliciting a given wheal size, as discussed below.
Although histamine challenges are used to investigate bronchial hyperresponsiveness, no one has yet used the results of the histamine or methacholine bronchial challenge to adjust the results of the bronchial allergen challenge. Yet, it is obvious that patients with hyperresponsive bronchi more easily react both to indirect stimuli, such as cold air, and to allergens.
Another question is whether or not histamine can be used as a reference. At present, the documentation favors the use of histamine rather as an indicator that the patient has not taken antihistamines before the skin test and as an indicator of dermal hyperresponsiveness like that observed in the airways.
In the paper in this issue (2), there is no information on the number of children with atopic symptoms or their total IgE and whether there were any differences between the years with respect to the frequency of atopy/atopic diseases, despite the hypothesis that there was no increase in sensitization, but in skin response to histamine.
Reading the histamine reaction
Vorhorst & Hooft-van Asbeck (18) found already during the 1970s that, in intradermal skin testing, the histamine wheal reaches a maximum after 10 min, results which have not been questioned since then. In a recent study, using laser doppler flow imaging and scanning of drawn wheal sizes at 5-min intervals, Brinck et al. (personal communication) found the maximum histamine wheal size to be reached 20 min after SPT with histamine dihydrochloride 10 mg/ml.
The recommendation by the EAACI Subcommittee on Allergen Standardization and Skin Tests in 1993 (11) was to use a cutoff limit of the wheal size of ≥3 mm in diameter or 7 mm2, and give the concentration of allergen and histamine, since both can now be determined.
Eigenmann & Sampson (19) investigated the reactivity of nonallergic patients in parallel with children with food allergy. The upper limit of the 95% confidence interval of the mean wheal diameter in “normals” varied from 3 mm for soy and wheat to 5 mm for milk. Unfortunately, they used nonstandardized extracts without known potency; therefore, the data can never be used in any other year with extracts from the same manufacturer or with other manufacturers' extracts. One interpretation is that the milk extract was at least 50 times more biologically active, or contained much more allergen, than the soy extract.
Furthermore, the recommendation was to draw the contour of the wheal on the erythema, transfer the drawing to the record sheet by means of translucent tape, and measure the diameters or, for scientific purposes, the area. This can easily be done by the commercially available (so far free) computer program (UCB Institute of Allergy) (20, 21).
What does the dose response of histamine mean?
We investigated the dose response of histamine and allergen (22–24). Firstly, we confirmed the findings by Haahtela (25) that the dose response to allergen as well as histamine is best fitted to a log/log model; secondly, we confirmed that the slope of the allergen dose-response relationship is flat. The slope of histamine in the model 10log wheal area/diameter=a+b *10log conc, where a is the intercept with the y-axis (close to 0), and b the slope of the dose-response relationship. If we use the area, the slope will be steeper by a factor of 2, due to simple mathematical rules. The slope was found to be 0.38 (∼0.4) using the area (π*r2), i.e., 0.19 (∼0.2) using the diameter (2*r*π). Then, it was possible to derive the formula for estimation of the concentration eliciting a wheal with 3-mm diameter (26). It can be calculated that the wheal diameter increases 1.5 times (area 2.5 times) if the concentration of allergen is increased 10 times; i.e., from 3 to 4.5 to 6.8 mm in diameter for 10 and 100 times increase in allergen concentration. The slope of the histamine dose-response relationship was found to be slightly, but significantly, less steep than the allergen slope; i.e., 0.34 and 0.17 for area and diameter, respectively (24). The dose response is best fitted to the log/log model, meaning that the CV is similar at different parts of the slope – larger than the cutoff – and there is a minimal bow of the curve within normal ranges.
Others have used a lin/log model, without documenting that model. In that model, the slope differs depending on where on the bowed curve it is calculated. Therefore, differences in slope largely depend on where on the curve the slope is estimated.
The report in this issue (2) uses the lin/log model and in addition wheal diameters below the normal cutoff limit; i.e., it is impossible to draw any conclusions from the data on the slope. To make calculations below the cutoff limit of any method is not recommended. The authors did so, and since the histamine wheals were smaller and not really measurable in the first trials, the slope was also flatter.
Turkletaub et al. (27) used not the wheal but the erythema of the i.c. test for evaluation of the skin response to allergen. They found the slope of the erythema to be much steeper than that of the wheal and therefore more suitable for parallel-line bioassay of the potency of allergens in relation to a standard. The wheal after injection of 0.01 ml of test solution was 3–4 mm in diameter but increased to 5–6 mm after 15 min. Furthermore, the slopes of the histamine and allergen dose-response relationships were flat and similar to that of SPT. Therefore, not only the wheal size but also the erythema should be considered with the intradermal method.
Furthermore, there are several ways to calculate the slope. Ronchetti et al. (2) have calculated the mean wheal size and then the slope of the total material. It is more relevant, but has a similar result, to calculate the slope of the response in each patient (22–24). Then it can clearly be shown whether there is a dose-response relationship for histamine in all patients. We found that for such calculations at least quadruplicate tests with each concentration are needed. Therefore, the method cannot be applied to epidemiologic studies, but the median slope in large patient populations can be used (22).
The size of the whealing response to histamine
The size of the whealing response is influenced by the composition and potency of the test solution and the pressure applied to the lancet – provided that the SPT method is used correctly. The histamine concentration is easy to adjust and should not be a problem, but the pressure is difficult to standardize between persons and over time within the same person. I have worked with two persons who obtained mean wheal diameters of 3 and 8 mm in the same patients (n≈30, quadruplicate tests) (personal communication). This would correspond to a difference in potency of about 500 times! It is recommended that a 6-mm wheal be obtained with 10 mg/ml histamine dihydrochloride in most patients, and the precision is acceptable if the variation is ±1 mm at that part of the dose-response curve.
In the paper presented in this issue, Ronchetti et al. (2) reported wheals partly below the recommended cutoff limit; i.e., the data on wheal sizes obtained in the study are irrelevant. This is especially true over time. They also calculated the slope from means as small as 1 mm in diameter. It is impossible to distinguish between 1 mm and 0 mm in diameter. Furthermore, it must be remembered that an increase in wheal diameter from 1.4 to 2.0 or 3.0 to 4.5 mm in diameter represents a 10-fold increase in sensitivity/reactivity or in potency.
The precision of the SPT with histamine
At least two parallel tests must be performed with the same material in every patient to calculate the precision. The CV can then be calculated (=SD * 100/mean). For many laboratory methods, the CV is less than 5%. For allergen determination in dust, it is often 10–15%; for intradermal testing under well-controlled conditions, it is about 10%, using the area (28); but for SPT, it is >20%, using the diameter, and >40%, using the area (22–24, 29). With duplicate tests, the diameter should be about 6 mm in diameter and must not vary more than±1 mm, which reflects the limits given above for variation. In a recent paper, Antico et al. (30) studied the precision of the deposit in the skin of test solution, using the Allergy Pricker method (8). The median volume of the prick test inoculum was equal to 15.9 nl, with a remarkable dispersion of the values (range: 0.418–82.253 nl). They observed a great variability both between subjects and also within subjects. They claimed that the observed variability depends on the individual characteristics of the subjects examined, not of the tester, and that a skilled tester using a standardized technique is not responsible for significant variability. However, they did not present data on any skin response, since they studied the inoculation only by using radioactive substances. Therefore, it was impossible to judge the precision of the final response, i.e., the whealing response to histamine.
In the present study (2), several investigators participated. The precision given in the paper, 8.5–20.5%, does not fit the data obtained in most clinics in Europe (24), i.e., 20–50%, as determined from the diameter values (double for areas).
Determination of the relative potency
SPT can also be used for determination of the relative biologic potency between batches of allergen extracts or products from different manufacturers.
The old method for determination of the potency of allergens (and histamine) was to titrate from a very low dose to a high. This methodology was developed into a precise system by the US Food and Drug Administration (FDA) (27). By intradermal testing, which has a much better precision than the SPT, it is possible to use the steep dose-response relationship for erythema to determine the relative potency of extracts and histamine. Unfortunately, some investigators still use the old, imprecise method of end-point titration not only for allergen extracts of conventional type, but also for testing the relative potency of recombinant allergens, a method which is not scientifically acceptable (31). However, in spite the flat dose response of allergens and histamine, the SPT can be used for determination of potency of allergens and histamine (15), provided quadruplicate tests with at least three 10-fold concentrations are used.
Histamine/SPT with allergens in epidemiologic trials
It is very common to use histamine and allergen skin testing in epidemiologic trials looking at the point prevalence of atopy and specific allergy. As already explained, it is very difficult to document a difference between regions and over time.
Data from the large European asthma study (32) must be viewed with caution, since this study used the allergen-coated Phazet needle, which was initially well standardized (33) but had a varying potency between batches (34). The potency of the batch used in the study was never established.
Another example is the Australian studies on the differences in sensitization between the dry desert climate in the Outback and the high-humidity coast (35, 36); these studies are not reliable since different batches of nonstandardized extracts were used over several years.
I suggested (37) the use of allergen-specific IgE to compare two regions, between which we found a difference in the prevalence of SPT reactivity. However, this is not ideal, since circulating IgE is not equivalent to cell-bound histamine-releasing active mediators. They measure different things.
Some epidemiologic trials comparing East and West have used both allergen-specific IgE and SPT, with similar results (37, 38), whereas others have not, indicating a possible lack of training and proper cooperation between centers.
Thus, because of such flaws, not only the paper in this issue (2) but also many other often-cited papers on the epidemiology of asthma and allergy should be interpreted with caution.
On the basis of the above documentation, I recommended that:
• skin tests be performed only with allergens/histamine with guaranteed potency and composition, i.e. those characterized and potency standardized;
• the Allergy Pricker with 1-mm tip should be used until a better instrument is developed;
• the SPT be performed in a standardized manner; i.e., that the tip be pressed at 90° against the skin surface, through the test solution, for and with similar pressure every time;
• duplicate tests be used routinely as well as in scientific studies, with the exception of infants, in order to be able to supervise the precision. This can be done in routine work, without calculations, if a mean wheal diameter of 6±1 mm is obtained with histamine dihydrochloride 10 mg/ml;
• computer scanning of wheal areas be used to reduce the error of measuring the wheal responses used in scientific studies;
• the same part of the dose-response relationship be used for comparison of the slope between patients or samples of patients;
• the concentration of allergen or histamine calculated to induce a certain wheal response, such as 3 mm in diameter, be used, since the size of the wheal in mm diameter is not comparable to concentrations normally used when challenging other organs. To use the diameter of wheals, on the one hand, and the concentration of allergen, histamine, or methacholine causing bronchial constriction, on the other hand, is like comparing apples and pears;
• differences between groups can be calculated on the basis of information on wheal sizes, but the real meaning of results from SPT can only be understood properly if transformation to concentration eliciting a certain response is used;
• when allergen reactivity is measured in epidemiologic trials, histamine be used for comparison of technique; testing personnel be trained and compared in the same patients to obtain similar results; these results be clearly documented and published; and a standardized in vitro test for specific IgE (at least for one important allergen) be used for comparison of sensitization, and the samples run in the same run to document similarities/differences between test occasions/regions, since the difference between investigators in SPT technique can never be fully corrected by training.
There are many unsolved problems in this area, waiting for scientific studies.
Finally, I would like to emphasize that these comments could have been made in relation to most papers on this subject, since the SPT method usually did not follow the above recommendations.