Epicutaneous patch testing

ARTICLE

Allergic contact dermatitis is a delayed type of hypersensitivity of the skin, for which epicutaneous patch testing is regarded as the best method of diagnosis. It is based upon re-exposing the skin of the patient to suspected allergens under controlled conditions. It is a bioassay that reproduces contact dermatitis. The first epicutaneous tests were carried out by Jadassohn in 1895 [1]. Although it is more than 100 years since the method of application was introduced, it is still the method of choice to establish contact allergy [2, 3]. In this review we discuss epicutaneous allergy testing and features of allergic contact dermatitis relevant for the patch test procedure.

History of patch testing

The first detailed description of an epicutaneous allergy test was probably by a chemist, Städeler at the University of Göttingen, Germany in 1847. He described a method called the blotting paper strip, which he had devised to examine the effects triggered by Anacardium occidentale [1, 4]. In 1895, Josef Jadassohn introduced his "Application Method" [5], inspired by Neisser's work concerning contact dermatitis to iodoform and mercury derivatives [6]. It was more than a decade before the concept of "allergy" was defined in 1906 by von Pirquet. After working with Jadassohn, Marion Sulzberger wrote his first major paper on the subject in 1931, together with Fred Wise: "The Contact or Patch Test in Dermatology" [7]. This test is now commonly called patch test, a term introduced by Cooke in 1916. During the last decades of the 20th century a lot of effort was put into standardization of materials and methods used in patch testing [8]. Patch testing is cost-effective and if used appropriately reduces the length of the episode and/or the severity of the dermatitis and by that, the cost of therapy in patients with allergic contact reactions [9]. However, the accuracy of the test should be established in each patient; in other words does patch testing always accurately distinguish patients who do and do not have allergic contact dermatitis.

Clinical picture of allergic contact dermatitis

The classic clinical picture of contact dermatitis is polymorphous; an erythematous eruption maybe accompanied by vesicles, infiltration, edema, scaling and sometimes crustae or excoriations. Usually it is accompanied by severe itching. The clinical picture or a positive patch test reaction resembles eczema in general, although it tends to be more sharply demarcated than the atopic or seborrhoeic form of dermatitis and the localization is more closely related to the area in contact with the responsible allergen. The classic positive patch test reaction is a miniature form of the same dermatitis: erythema, edema, and small, closely set vesicles, which often extend beyond the borders of the patch.

Etiopathogenesis

Allergic contact dermatitis is a delayed type of hypersensitivity, a type 4 allergic reaction. However, it may not be a traditional type 4 hypersensitivity, due to the importance of the interaction between the irritant and antigenic properties of sensitizing chemicals in the development of allergic contact dermatitis [10]. It is an inflammatory reaction, mediated by antigen specific T-lymphocytes. It can be divided into 3 phases: the sensitization, the challenge and, if the exposure stops, the resolution phase.

Given the clinical nature of this article, we refer the readers to the article by Belsito DV et al. [11].

The patch test procedure

The present patch test technique is the result of a continuous process of development and improvement since its first application in the late 19th century [12].

Patch testing is only indicated if after history taking and clinical examination, allergic contact dermatitis is suspected. Epicutaneous patch testing is especially indicated if: a) a clear relationship is evident between the dermatitis and certain professional or other activities, b) the dermatitis is confined to the hands or the feet, peri-orbital, around ulcera cruris or peri-anal dermatitis, c) acute and wetting dermatitis of any localization and d) any dermatitis that is therapy resistant, exists for over 3 months or worsens during topical treatment [13].

In the original test system the patches, tapes and allergens are supplied separately.

Different patch test units are now commercially available; such as the Finn Chambers or van der Bend square chambers. These test chambers are filled manually. The modern adhesive tapes are acrylate and not colophony based, so the problem of colophony allergy has been eliminated.

Standard patch test allergens are commercially available and have to be chemically defined and pure. The suppliers' recommendations on storage are to be followed to minimize the risk of degradation due to humidity, air or light. Most preparations should be kept in a refrigerator and in the dark; those in diluted liquid preferably in dark bottles. The allergens should not be stored vertically, to prevent sedimentation and concentration changes of the allergens.

The test preparation in petrolatum, kept in syringes, is applied directly onto the test chamber. Liquid test preparations are applied via a digital pipette to allow exact dosing. The preferred test site is the upper back, but the outer sides of the upper arms are also acceptable, especially when retesting. Only areas covered by clothing should be used, because some positive reactions may persist for several weeks and occasionally produce hypo- or hyperpigmentation. Removal of hair on the back is sometimes recommended for practical reasons, but it can contribute to the skin irritation. Oily skin can be degreased with a mild solvent, which must evaporate before applying the test strips.

The skin of the back should not be treated with topical corticosteroids for one week before testing. Preferably oral corticosteroids should also be avoided during testing, because they can suppress positive test reactions. The same goes for cytostatics and cyclosporin. During one week before testing the skin should not be irradiated by the sun or artificial ultraviolet sources [13].

Each test site can be easily delineated with a marking paint, such as gentian violet, felt pens or the nearly colorless "ultraviolet" paint, which shines bright yellow when exposed to black light.

Patients should be informed about avoiding excessive exercise, showers, etc. to keep the test system dry.

The patch test system is usually removed after 48 hrs, as recommended by the International Contact Dermatitis Research Group [14], and readings are done 20 min after removal of the strips (day 2) and after 72 (day 3) or 96 hrs (day 4). For some test series it is preferable to read the tests once more after 7 days, not to miss the late reactions. Gold and certain therapeutic agents such as dermatocorticosteroids and neomycin have the tendency to appear later than reading day 2 or 4 [15, 16]. Patients should be asked to note new positive reactions that arise after the readings at 96 hrs and report them promptly.

Occasionally some severe reactions can cause itching and burning, in which case, that patch can be prematurely removed without disturbing the others.

The interpretation method recommended by the International Contact Dermatitis Research group (ICDG [17]) is:

- Negative reaction

?+ Faint erythema only: doubtful reaction

1+ Nonvascular erythema, infiltration, possibly papules: weak positive reaction

2+ Vesicular erythema, infiltration, papules: strong positive reaction

3+ Intense erythema and infiltration, coalescing vesicles, bullous reaction: extreme positive reaction

IR Irritant reaction of different types

NT Not tested

The ?+ reaction is meaningful for an allergic reaction in 1-5%, the 1+ reaction in 20-50% (depending upon the allergen), the 2+ reaction in 80-90% and a 3+ reaction is almost always allergic [13].

This method was developed to make the interpretation standard and easy, but the not all types of reactions fit this outline. Irritant reactions are said to be characterized by fine wrinkling ("silk paper"), erythema and papules in follicular distribution, petechiae, pustules, bullae or even necrosis and with minimal infiltration [18].

Irritant reactions are frequent, even to the standard series, because some of the concentrations have been chosen close to the irritancy threshold to diminish the risk of obtaining false-negative reactions. The frequency of irritant reactions is higher at day 2 as compared to later readings [19]. The morphology can differ from mild erythema to bullae. Sometimes it is indistinguishable from an allergic reaction and can be the cause of false-positive test reactions.

Ready-to-use patch test systems are also available, preloaded by the manufacturer. They eliminate some of the variability in patch testing.

The allergens

Different variables can be influenced to obtain an optimal bioavailability of the haptens: intrinsic penetration capacity, concentration, exposure time, vehicle and occlusivity of patch test systems.

The penetration capacity depends upon the salt used. For example, there is a significant difference between the penetration of nickel achieved by nickel sulphate and nickel chloride [20].

It is important to find the ideal test concentration. Too high concentrations can cause irritation; too low concentrations are responsible for the false-negative test reactions. False-negative test reactions can also be due to the failure to duplicate the conditions present in the real dermatitis situation. The presence of the same allergens is often not enough; e.g. friction and sweating is an important contributing factor in shoe dermatitis, which are difficult to mimic during testing. In this case wet patch testing can be performed, keeping the leather patches from the patient's shoes wet with artificial sweat or physiologic solution.

The concentration of an allergen is normally given as a percentage, but in comparative studies with different salts of a substance it is essential to use the same molality [21].

Mostly an exposure time of 48 hrs is chosen and all test strips are removed at the same time.

Each allergen has its own optimal vehicle, but each vehicle has its own drawbacks. White petrolatum is the most widely used. It gives a good occlusion and keeps the allergens stable, but it can retain the allergen and irritate and even sensitize the skin [22]. Liquid vehicles such as water or solvents facilitate the penetration, but they evaporate, which interferes with exact dosing. Most test solutions with liquid vehicles must be freshly made. When using other, more sophisticated vehicles, containing alkalis, anionic detergents, etc., the vehicle itself must also be patch tested to exclude the possibility that the vehicle is irritant. For example we found a high number of irritant patch test reactions to the vehicle DMSO/alc.abs. 50/50 used to patch test hydrocortisone acetate 1% [23]. Buffer solutions can be used for alkaline or acid allergens, to raise the test concentration [24].

Although the history and examination of a patient with suspected allergic contact dermatitis give clues to the allergens responsible, it is not enough to test only the initially suspected sensitizers, because the unsuspected frequently turn out to be the real cause of the dermatitis [25]. There are approximately 3,700 currently known allergens [26]. It's obvious that it is impossible and undesirable to test them all. That's why a small number of substances, considered to account for the majority of delayed hypersensitivity reactions, are grouped into a standard patch test series. Bruze et al. [12] discussed the requirements to be fulfilled by a sensitizer for inclusion in the standard series. Demands on a sensitizer in the standard series are, being common in the environment, contact allergy rate above 0.5-1.0% in routinely tested dermatitis patients, reliable patch test results, high degree of clinical relevance and minimal adverse effects, particularly patch test sensitization [12]. Generally 20-30 test preparations are grouped in a standard test series, which consist of chemically defined compounds, mixes of allergens, both natural and synthetic. These series are revised frequently to adapt to changes in exposure and the introduction of new allergens onto the market and one should always remain critical of the contemporary composition of the standard series. Minor variations are due to differences in culture, industrialization and use in different countries [27-29]. Testing with the test preparations in a standard series is said to detect 70 to 80% of all contact allergies [13, 30]. The European Environmental and Contact Dermatitis Research Group [31] detected by the standard series hypersensitivity with a range from 37 to 73% of all contact allergies. Many international research groups now appreciate that these numbers may be artificially high. Indeed, the more one seeks out the cause of an allergy, the more likely one is to find an allergic cause. Cohen et al. found that, among 732 patients, only 23% had allergies detected solely by the standard series [32]. In another study by Sherertz and Swartz only 36% of positive reactions occurred to allergens in the standard series [33]. Indeed, there is mounting evidence that most standard series will detect much less than 50% of all relevant allergens.

To evaluate the significance of specific exposures, different specific screening series are available. They can be divided into different categories based upon the occupation of the patient (e.g. hairdressing or bakery series), the localization of the dermatitis (e.g. shoe series), series of chemically related compounds (e.g. the acrylates or epoxy series) or functionally related compounds (e.g. corticosteroid series, cosmetics series). Overlapping occurs since many chemicals are present in several unrelated compounds.

Mixes of four or five closely related chemicals are used to save time and space while patch testing. In the past several different mixes of allergens have been tried in patch tests, but at present the most standard series contain mixes of "caine" anesthetics, parabens, fragrances and rubber chemicals. Caution must be observed in the exact composition of the mixes. The concentration of each chemical has to be sufficient. A sensitive patient can show a negative reaction to a mix, but a positive reaction to one of its ingredients tested separately, because the concentration of the ingredient in the mix is insufficient. The combination of the substances in the mix may not cause chemical reactions deactivating one of the ingredients or induce irritation to the skin.

In cross-sensitivity, contact allergy caused by a primary allergen is combined with allergic reactions to other, chemically closely related substances.

Products or materials brought by the patient and suspected of causing dermatitis should be tested with great caution. First of all it should be stated that totally unknown products should never be applied to human skin. Therefore one should always ask the patient to bring as much information on the product as possible, e.g. in the form of lists of ingredients, safety data sheets, etc. or they should be requested from the manufacturer. Usually some of the ingredients can be suspected more than the others, and if they are available from suppliers of patch test allergens, they should be tested in the vehicle and concentration as recommended by the supplier. Sometimes these tests remain negative and one can wonder whether the suspected product really causes an allergic contact dermatitis. Therefore it is recommendable to start with an open test, to minimize the risk of severe irritancy. If this is negative, occlusive patch testing can take place, usually starting at the lowest concentration and rising if the preceding test is negative. Depending on the likely irritant or sensitization potential of the product, it is recommended to start with concentrations of 0.001% or 0.01%. The ideal vehicle and test concentration for each product or chemical compound is difficult to discover, but help can be found in the literature [25]. It is practical to apply it on a site that can be easily reached by the patient, to remove the patches if severe stinging or burning should occur. If the test remains negative after one day, the concentration can be raised for the next test. If the test is found positive in the patient, at least 10 unexposed controls should be tested to prove that the actual test preparation is nonirritant. As an example we mention 3 cases of dermatitis around a tracheostomy, caused by a cleansing tissue, which the patients used to remove glue from the skin around their tracheostomy. Not only the patients themselves
showed severe patch test reactions to the cleaning tissue itself and a 10% concentration of the extract, but 10 (50.0%) of the 20 healthy subjects tested with the extract showed a + reaction to the 10% concentration after 72 hrs, which was clearly an irritant reaction [34].

When solid products such as textiles, paper, rubber, plants or synthetics are suspected, it can be tested as thin, regular-sided, smooth sheets or
extracts can be obtained by placing a sample of the material in water, synthetic sweat, ethanol or ether, and heating it up to 50 °C [35].

For most products intended for use on normal or damaged skin such as cosmetics, detergents, topical medicaments, etc. open tests and use test give probably more information on the pathogenesis of the patient's dermatitis than an occlusive patch test does [14].

Reproducibility, sensitivity and specificity

The reproducibility of patch tests remains controversial. A series of papers has demonstrated good reproducibility of the patch test results, although some mention a high percentage of non-reproducibility with the concomitant method of patch testing (patch testing simultaneously on both sides of the back). In the literature we can find reproducibility percentages varying from 48% up to 96% [36-46]. Brasch J. et al. found that non-reproducibility of patch tests seems to be strongly allergen dependent. In their synchronous left-versus right-sided patch test study, the likelihood of non-reproducible allergic reactions increased when more than four positive reactions were seen at the same time, and with another positive reaction located in close proximity to an allergic reaction. Other factors such as age, sex, atopy, sleeping habits, lipogenic skin activity, systemic medication, inflammatory dermatoses outside the back and internal medication (excluding corticosteroids) were of minor importance for patch test reproducibility [47]. Weaker patch test reactions seem to be less reproducible. Some of the variability is eliminated by the use of ready-to-use patch test systems. Gollhausen et al. found that such a system (TRUE test) eliminated about half of the non-reproducible reactions [48] and Lachapelle et al. found that another preloaded system (Epiquick) was 95% reproducible in a left-to-right comparison [49].

The ideal patch test should give no false-positive or false-negative reactions. A false-positive reaction is an irritant reaction with the same morphology as an allergic patch test reaction and therefore cannot be separated from reactions caused by sensitization. It can be caused by too elevated test concentrations, impure or contaminated test substances, irritant test substance or vehicle, current or recent dermatitis at the test site, current dermatitis at distant skin sites, pressure effects or mechanical irritation.

False-negative reactions in the presence of a contact allergy can be due to too low test concentrations, test substance in insufficient amount or not released from the vehicle, test panels removed too soon, reading taken too early, inappropriate co-medication such as corticosteroids or due to a compound allergy. The latter is the condition in patients patch tested positive to formulated products, usually cosmetic products or topical medications, but tested negative to all the ingredients tested individually [50]. The mechanism can be irritancy of the original formulation, a false-negative reaction to the ingredients or due to reaction products formed by the combination of the ingredients [13, 51, 52]. Irritants in the original formulation may increase the allergic contact dermatitis response and may explain the presence of contact dermatitis in patients with negative patch tests [53].

Pseudo-compound allergy is due to faulty patch testing, illustrated by patch testing the individual ingredients at the usage concentrations in petrolatum, which are too low for many allergens.

Quenching is the term used to describe the phenomenon of inhibition of sensitization. The concept was introduced by the publication of Opdyke a quarter of a century ago [54]. The term quenching was employed to describe the complete abrogation of the sensitizing potential of 3 fragrance chemicals (cinnamal, citral and phenylacetaldehyde) by the presence of certain other fragrance chemicals (notably eugenol and limonene). In a recent publication Basketter concludes that no satisfactory (physico)-chemical or immunobiological basis for a quenching activity can readily be evinced, particularly since it would appear to be required to operate only during induction in man [55]. It is still a hypothesis, lacking substantive proof.

Inappropriate co-medication during patch testing includes topical and oral corticosteroids and immunomodulators. Treatment of the test site with topical corticosteroids can mitigate the responses obtained to a high degree [56].

Testing patients on oral corticosteroids is not recommendable. Comparison of the intensity of the reaction before and during treatment with corticosteroids has suggested that an important allergy cannot be missed under corticosteroids up to 20 mg [57-59]. However, it is usually advisable to defer the test until after corticosteroids have been stopped. The use of antihistamines as a contraindication for patch testing is not universally accepted. Some studies show that it is useless to stop antihistamines before patch testing, since clinical evaluation of tests is not hampered by a potent antihistamine [60]. Based upon echographic evaluation of patch test inhibition by the oral antihistamine loratadine, Motolese states that it is advisable for patients who are to undergo patch testing to be off loratidine (and probably also other oral antihistamines) to enable an evaluation that reflects their true sensitization state [61]. Nevertheless, in a multitude of other studies, the effects of oral antihistamines on the patch test response have been at best variable, and more frequently non-existent. This is particularly important since many patients are unable to avoid scratching the patch test area without the use of oral antihistamines, resulting in a much more confusing picture to interpret clinically.

The influence on patch testing of other immunomodulators such as orally or parenterally administered cytostatic drugs has not yet been clarified. Oral cyclosporin A inhibits expression of delayed contact hypersensitivity reactions in human skin [62]. Cyclosporin, which has a potent clinical impact on eczema, does not seem to modify the clinical score of intense reactions to contact allergens in patch tests [63]. The influence of topically applied cyclosporin A on patch test reactions is not clear [64].

Several studies have reported on the suppressive effect of ultraviolet B (UVB), UVA sunlight and PUVA on contact dermatitis. However, studies that have tested the hypothesis that patch tests reactions have a seasonal variation due to the suppressive influence of sunlight, have had conflicting results [65-68].

The common screening allergens used to patch test patients thought to have contact dermatitis have a sensitivity of 77% and a specificity of 71%, very good for a bioassay [69].

From a statistical point of view, it is crucial to explore the patient's history carefully and exactly before performing patch testing: indiscriminate testing of many patients with a doubtful allergic origin of their skin problem will lead to many cases of wrongly diagnosed contact dermatitis [70].

Patch testing is as much art as it is science; we all are influenced by our clinical experience as well as by the literature [71]. Careful history taking combined with physical examination is extremely important in deciding whether or not to patch test and in deciding which materials should be tested in order to increase the predictive values of patch test results.

Relevance of positive patch test reactions

Patch tests can be used to confirm a suspected allergic contact dermatitis and either to recommend avoidance of particular products or to recommend alternative products in a particular patient.

The true rate of clinically relevant hypersensitivity in positive patch test reactions remains for a great part unknown. To know that a patient has been exposed to a sensitizer is insufficient to conclude that the positive patch test is relevant. There is always the risk of over- or underestimating the significance of positive patch test reactions.

When a positive patch test reaction is found, an attempt must be made to fit it into the information obtained from the history and clinical examination.

The relevance of the positive patch test reaction is not always easy to judge. For practical reasons it can be categorized as definite, probable or possible [72].

The major prerequisites for a contact allergy to be clinically relevant are: 1) exposure to the sensitizer and 2) presence of a dermatitis which is understandable and explainable with regard to the exposure on the one hand and type, localization and course of the dermatitis on the other [73]. The problem is that one sensitizer causing the entire clinical picture of the dermatitis is rare. Mostly the cause is multifactorial, including irritant and constitutional influences. It is important to obtain sufficient information on the exposure to the suspected sensitizer, by questioning the patient's own experience, analyzing data sheets on packages of used products, chemical analyses, etc. [73]. Patient's own suspected products are often tested and in the case of a positive patch test reaction to one of them, the ingredients must be analyzed. Use tests are generally carried out in a standardized manner, but sometimes the diagnostic value may increase when the test simulates ordinary use conditions. In some products the concentration of the sensitizer is too low to elicit a patch test reaction, but sufficient to cause dermatitis in under use conditions with frequent exposures [73].

Often the substances giving rise to positive patch test results represent aggravating factors, but are not the cause of the primary dermatitis. This is especially true for reactions to medications, e.g. eardrops, used to treat a pre-existent skin disease, such as a chronic otitis externa [74, 75].

Just as a positive reaction does not always mean that the primary cause of the dermatitis has been found, so a negative reaction does not always mean that the dermatitis is not caused by contact allergic hypersensitivity. Standard series include only statistically common allergens; one must be constantly alert to the possibility of rare, exotic or new sensitizers [25].

Adverse reactions

The ideal patch test should cause as few adverse reactions as possible.

- Irritancy itself can be considered as an adverse effect, especially the more severe reactions such as a chemical burn. Irritant patch test reactions are usually sharply demarcated, confined to the area covered by the patch.

- The "excited skin syndrome" or "angry back" means that there are many patch test reactions of which some are false-positive [76, 77]. The cytokines released by inflammatory skin may enhance other patch test reactions [78]. It is a regional phenomenon caused by: a) subclinic dermatitis in an atopic patient or b) the presence of a strongly positive patch test reaction, which produces a state of skin hyperreactivity in which other patch test sites become reactive, especially the marginal irritants. To confirm or deny the significance of the individual reactions, each substance should be tested again individually.

- The "edge effect" is often an irritant reaction, with a more intense reaction at the periphery of the patch than in the center, due to an increased concentration of the irritant liquid at the margin. Sometimes a reaction with edge effect can be a false negative or doubtful patch test reaction to a corticosteroid. This is an eczematous reaction only apparent on the edge of the patch test site, particularly at the first reading. Probably the inflammation is still suppressed in the middle of the site where the concentration of the corticosteroid is the highest, while around the edges of the site, the corticosteroid diffuses through the skin, and the low concentrations allow the allergenic effect to prevail [79].

- Pustular patch test reactions are sometimes a manifestation of irritancy. Pruritus is often minimal or absent and the reactions usually disappear promptly, although they can occasionally persist for some days [80].

- Pressure reactions can occur, especially with the use of solid test substances and in patients with a tendency to dermographism.

- A temporary flare of the existing dermatitis elsewhere on the skin can be due to a positive patch test reaction.

- Numerous substances can cause contact urticaria, most frequently patients' own materials brought from home or work to test; especially common are penicillin, balsam of Peru and phenylmercuric compounds. Katsarou states that almost all allergens of the European standard series can produce an immediate contact reaction. They are associated with delayed contact reactions to the same allergen in a minority of cases [81]. It should be considered if the patient complains of itching minutes after application of the patches. The mechanism may be immunologic or nonimmunologic.

- Some sensitizers are known or suspected carcinogens, but the effect from the low doses applied to the back in patch testing is negligible.

- Some metals and metal compounds used in patch testing can have systemic toxic effects [82].

- Hyperpigmentation may result from the inflammation alone, independent of the chemical response in certain patients [83]. Sometimes a severe reaction causes hyperpigmentation or total depigmentation.

- Several chemicals are capable of inducing contact leukoderma. Contact leukoderma can appear following an allergic contact dermatitis and it may be indistinguishable from idiopathic vitiligo [84]. It occasionally appears during patch testing. Another example that one should be careful in testing products brought and used by the patient is the one of the dental acrylic resins tested "as is". Due to the high concentrations the patch test sites can remain vitiliginous for many months up to years [85].

- Most dermatologists will not patch test pregnant women, although the teratogenic capacity of the patch test substances is probably nil. The rationale is to avoid problems if there should be perinatal or congenital abnormalities due to other causes. However, before introducing new compounds into the test series, the teratogenic potential should be considered.

- Patch test sensitization is considered to be the most serious adverse reaction of patch testing. It is detected by a flare-up reaction at the test site at least 10 days after the application [86]. On being repeated, the test is already positive at day 2-4. It is more likely that the flare-up reaction represents patch test sensitization than does the finding of a positive reaction to a substance that has previously been tested negatively. In the latter case, the patient may have been sensitized in the interval between the performances of the patch tests due to environmental exposure to the antigen [87]. Active sensitization is less likely to happen by using the lowest concentration of test substance required to cause a reaction. That is another reason why one should beware of patch testing substances brought by patients. Nevertheless there is a risk of active sensitization from the standard series. Para-phenylenediamine is an example of a compound in the standard series that has a strong capacity to sensitize and therefore it is still subject of continuing debate whether it belongs in a standard series [88].

- Anaphylactoid or anaphylactic reactions can occur from e.g. penicillin. It is therefore not recommended for routine patch testing. Radioallergosorbent tests (RAST) seem the most safe in case of suspected immediate reactions (type I allergic reaction). In the diagnosis of nonimmediate reactions both patch and intradermal tests could be useful [89, 90].

A lot of adverse reactions have been summarized above, but it has to be noted that the overall risk-benefit equation of patch testing is in favor of the benefit, if performed correctly and with the proper indications.

Other tests

- Open tests are recommended as the first step when testing poorly defined or unknown substances, brought by the patient. This concerns especially gels, liquids or creams, suspected of producing irritant reactions if occluded. Cosmetics such as perfumes, aftershave lotions and hairsprays are the prototypes in this kind of testing. It is applied undiluted to the normal skin twice a day for at least two days. The outer aspect of the upper arm or the retroauricular area is the recommended site for open tests. It should be left uncovered and the application has to be discontinued if any irritation arises. The test is read after 15 to 30 min to detect contact urticaria. Otherwise the readings are done as with the closed patch tests. A negative open test indicates that an occlusive patch test can be preformed with the substance without expecting severe irritant reactions.

- The provocative use tests are performed to confirm positive patch test reactions. The suspected agent is used on normal skin as it was before, to evaluate if a relapse occurs. If no reaction occurs, the test may be considered negative [25]. It is important to evaluate the clinical significance of the ingredients of a formulated product found positive by patch testing.

- The repeated open application test (ROAT) is a use test, performed on the outer aspect of the upper arm, the antecubital fossa or the scapular area of the back over an area of approximately 3 centimeters in diameter. The substances are applied twice daily for 7 days. A positive response usually appears on day 2 to 4. The patient is instructed to stop the application of the test substances when a reaction is noticed [91].

- Prognostic patch testing differs from diagnostic patch testing. Clinicians routinely perform diagnostic patch testing, but prognostic patch testing is the province of corporate research and development departments and, in the United States, contract laboratories [92]. The purpose of this type of patch testing is to determine the irritant or allergic contact dermatitis potential of an ingredient or a product.

- "Preventive" or "prophetic" patch testing, for example on potential employees. It is not at all recommendable. It is not a proper indication and it exposes the patient unnecessarily to the different allergens and by that to the adverse effects possible.

- Photopatch testing should be used to investigate patients with clinically suspected photoallergic contact dermatitis (PACD). PACD is caused by photochemical conversion of a certain agent into a contact allergen, mainly induced by UVA. Particularly plant derivatives, fragrances, antiseptics and sunscreen agents are known for photosensitization. The latter have now become the most common photoallergens, due to extensive use during recent decades. Diagnostic procedures for PACD include MED testing and photopatch testing. The differential diagnosis of PACD includes: airborne allergic contact dermatitis, phototoxic reactions, chronic actinic dermatitis, seborrhoeic dermatitis, polymorphic light eruption, variants of systemic lupus erythematosus and cutaneous porphyrias.

The photopatch test procedure can vary between centers, but in general goes as follows. The test materials are applied to the back in a duplicate set for 24 hrs. One test site is irradiated with UVA (320-400 nm) and the other serves as an unirradiated control. The tests are read immediately and 24, 48 and 72 hrs after UVA irradiation.

- In vitro assays:

Presently, there is no in vitro technique available to replace epicutaneous patch testing.

The lymphocyte transformation test: Defibrinated blood for in vitro lymphocyte studies is obtained from the patient. The lymphocytes are incubated and cultivated with various concentrations of allergens. A positive control of proliferation is obtained with a purified protein derivative. Lymphocyte transformation in control and antigen-treated cultures is measured by 3H-thymidine incorporation into lymphocyte DNA after 5d cultivation.

CONCLUSION

We conclude that epicutaneous patch testing is a useful procedure if it is well performed and if it is performed for the right indication.

However, the problems concerning the relevance of these tests remain. A careful history taking and attention for the clinical picture are key actions to facilitate the interpretation of the clinical relevance of the epicutaneous patch test results.

Maybe we can expect an evolution in epicutaneous patch testing in the 21st century. To diminish the problems of the relevance a proper selection of patch test chemicals should be made, adapted to the individual patient's problems.

Article accepted on 2/5/02

REFERENCES

1. Foussereau J. History of epicutaneous testing: the blotting-paper and other methods. Contact Dermatitis 1984; 11: 219-23.

2. Lachapelle JM. First Jadassohn lecture: a century of patch testing. Paper given at the Jadassohn Conference in London, October 1996.

3. Fischer T, et al. Improved, but not perfect, patch testing. Am J Contact Dermat 1990; 1: 73-90.

4. Städeler. Über die eigenthümlichen Bestandtheile der Anacardium Früchte. Ann Chemie Pharmacie 1847; 63: 117-65.

5. Jadassohn J. Zur Kenntnis der medikamentösen Dermatosen, Verhandlungen der Deutschen Dermatologischen Gesellschaft. Fünfter Congress, Graz, 1895. Wien: Braunmuller W., 1896: 106.

6. Neisser. Uber sogennante mercurielle Exantheme, Verhandlungen der Deutschen Dermatologischen Gesellschaft. 15th Congress, Graz, 1895. Wien: Braunmuller W., 1896: 129-43.

7. Sulzberger MB, Wise F. The contact or patch test: its uses, advantages, and limitation. Arch Dermatol 1931; 23: 519-31.

8. Fischer TI, et al. The science of patch test standardization. Immunol Allergy Clin N Am 1989; 9: 417-43.

9. Rajagopalan R, et al. An economic evaluation of patch testing in the diagnosis and management of allergic contact dermatitis. Am J Contact Dermat 1998; 9: 149-54.

10. Zhang L, Tinkle SS. Chemical activation of innate and specific immunity in contact dermatitis. J Invest Dermatol 2000; 115: 168-76.

11. Belsito DV, et al. The rise and the fall of allergic contact dermatitis. Am J Contact Dermatitis 1997; 8: 193-201.

12. Bruze M, et al. Thoughts on sensitizers in a standard patch test series. Contact Dermatitis 1999; 41: 241-50.

13. De Groot AC. Richtlijnen voor het verrichten van epicutaan allergologisch onderzoek in de perifere dermatologische praktijk. Nederlands Tijdschrift voor Dermatologie & Venereologie 1996; 6: 8-15.

14. Wahlberg JE. Patch testing. In: Rycroft RJG, Menné T, Frosch PJ, et al., eds. Contact Dermatitis. Berlin: Springer-Verlag, 1992: 239-68.

15. Bruze M, et al. The development and course of test reactions to gold sodium thiosulfate. Contact Dermatitis 1995; 33: 386-91.

16. Dooms-Goossens A, et al. Sensitization to corticoids. Consequences for anti-inflammatory therapy. Drug Saf 1995; 13: 123-9.

17. Fregert S. Manual of contact dermatitis, 2nd ed. Munksgaard, Copenhagen 1981.

18. Fischer T, et al. Patch testing in allergic contact dermatitis: an update. Semin Dermatol 1986; 5: 214-24.

19. Rietschel RL, et al. The case for patch test reading beyond day 2. J Am Acad Dermatol 1988; 18: 42-5.

20. Fullerton A, et al. Permeation of nickel salts through human skin in vitro. Contact Dermatitis 1986; 15: 173-7.

21. Wahlberg JE. Nickel chloride or nickel sulfate? Irritancy from patch test preparations as assessed by laser Doppler flowmetry. Dermatol Clin 1990; 8: 41-4.

22. Dooms-Goossens A, et al. Contact allergy to petrolatums. I. Sensitizing capacity of different brands of yellow and white petrolatums. Contact Dermatitis 1983; 9: 175-85.

23. Devos SA, van der Valk PG. Relevance and reproducibility of patch-test reactions to corticosteroids. Contact Dermatitis 2001; 44: 362-5.

24. Bruze M. Use of buffer solutions for patch testing. Contact Dermatitis 1984; 10: 267-9.

25. Adams RM. Patch testing: a recapitulation. J Am Acad Dermatol 1981; 5: 629-43.

26. De Groot AC. Patch testing. Test concentrations and vehicles for 3,700 allergens. Amsterdam: Elsevier, 1994.

27. Andersen KE, et al. Allergens from the standard series. In: Rycroft RJG, Menné T, Frosch PJ, eds. Textbook of contact dermatitis. Berlin: Springer-Verlag, 1995: 415-58.

28. Bruynzeel DP, et al. The European standard series. Contact Dermatitis 1995; 33: 145-8.

29. Lachapelle JM, et al. Proposal for revised international standard series of patch tests. Contact Dermatitis 1997; 36: 121-3.

30. Young E, et al. Patch test results with standard allergens over a decade. Contact Dermatitis 1987; 17: 104-7.

31. Menné T, et al. How large a proportion of contact sensitivities are diagnosed with the European Standard Series. Contact Dermatitis 1992; 26: 201-2.

32. Cohen DE, Brancaccio R, Andersen D, Belsito DV. Utility of a standard allergen series alone in the evaluation of allergic contact dermatitis: a retrospective study of 732 patients. J Am Acad Dermatol 1997; 36: 914-8.

33. Sherertz EF, Swartz SM. Is the screening patch test tray still worth using? J Am Acad Dermatol 1993; 29: 1057-8.

34. Devos SA, van der Valk PG. Dermatitis around tracheostomies due to cleansing tissues. Contact Dermatitis 2001; 44: 111-2.

35. Rycroft RJG. False reactions to nonstandard patch tests. Semin Dermatol 1986; 5: 225-30.

36. Basketter D, et al. Visual assessment of human skin irritation: a sensitive and reproducible tool. Contact Dermatitis 1997; 37: 218-20.

37. Belsito DV, et al. Reproducibility of patch tests: a United States multicenter study. Am J Contact Dermatitis 1992; 3: 193-200.

38. Bourke JF, et al. The reproducibility of patch tests. Br J Dermatol 1999; 140: 102-5.

39. Bousema MT, et al. High reproducibility of patch tests. J Am Acad Dermatol 1991; 24: 322-3.

40. Breit R, et al. Qualitätskontrolle der Epikutantestung - Reproduzierbarkeit im Rechts-Links-Vergleich. Hautarzt 1992; 43: 417-21.

41. Lachapelle JM, et al. Problems raised by the simultaneous reproducibility of positive allergic patch test reactions in men. J Am Acad Dermatol 1989; 21: 850-4.

42. Lachapelle JM, et al. A left versus right side comparative study of Epiquick™ patch test results in 100 consecutive patients. Contact Dermatitis 1989; 20: 51-6.

43. Lindelöf B. A left versus right side comparative study of Finn Chamber™ patch tests in 220 consecutive patients. Contact Dermatitis 1990; 22: 288-9.

44. Machackova J, et al. Reproducibility of patch tests. J Am Acad Dermatol 1991; 25: 732-3.

45. Rietschel RL. Reproducibility of patch test results. Lancet 1996; 347: 1202.

46. Stransky L, et al. A left versus right side comparative study of Finn Chamber™ patch tests in consecutive patients with contact sensitization. Dermatosen 1992; 40: 158-9.

47. Brasch J, et al. A multicenter study of synchronous left- versus right-sided patch tests by the German Contact Dermatitis Research Group. J Am Acad Dermatol 1994; 31: 584-91.

48. Gollhausen R, et al. Reproducibility of patch test results: comparison of TRUE test and Finn chamber test results. J Am Acad Dermatol 1989; 21: 843-6, 1196-202.

49. Lachapelle JM, et al. Problems raised by the simultaneous reproducibility of positive allergic patch test reactions in man. J Am Acad Dermatol 1989; 21: 850-4.

50. Kelett JK, et al. Compound allergy to medicaments. Contact Dermatitis 1986; 14: 45-8.

51. Smeenk G, et al. Contact allergy to a reaction product in Hirudoid cream: an example of compound allergy. Br J Dermatol 1987; 116: 223-31.

52. Serpentier-Daude A, et al. Contact dermatitis to topical antiviral drugs. Ann Dermatol Venereol 2000; 127: 191-3.

53. McLelland J, et al. "Irritants" increase the response to an allergen in allergic contact dermatitis. Arch Dermatol 1991; 127: 1016-9.

54. Opdyke DLJ. Inhibition of sensitization reactions induced by certain aldehydes. Fd Cosmet Toxicol 1976; 14: 197-8.

55. Basketter D. Quenching: fact or fiction? Contact Dermatitis 2000; 43: 253-8.

56. Meffert H, et al. Disodium cromoglycate inhibits allergic patch test reactions. Contact Dermatitis 1985; 12: 18-20.

57. O'Quinn SE, et al. Influence of oral prednisone on eczematous patch test reactions. Arch Dermatol 1969; 99: 380-9.

58. Feuerman E, et al. A study of the effect of prednisone and an antihistamine on patch test reactions. Br J Dermatol 1972; 86: 68-71.

59. Condie MW, et al. Influence of oral prednisone on patch-test reactions to Rhus antigen. Arch Dermatol 1973; 107: 540-3.

60. Grob JJ, et al. Antiinflammatory properties of cetirizine in a human contact dermatitis model. Clinical evaluation of patch tests is not hampered by antihistamines. Acta Derm Venereol 1998; 78: 194-7.

61. Motolese A, et al. Echographic evaluation of patch test inhibition by oral antihistamine. Contact Dermatitis 1995; 32: 250-1.

62. Higgins EM, et al. Oral cyclosporin inhibits the expression of contact hypersensitivity in man. J Dermatol Sci 1991; 2: 79-83.

63. Flori ML, et al. Patch test results in patients treated with cyclosporin. J Eur Acad Dermatol Venereol 1996; 7 (suppl. 2): S109.

64. Lembo G, et al. Influence of topically applied cyclosporin A in various vehicles on patch test reactions. Contact Dermatitis 1992; 26: 56-7.

65. Katsarou A, et al. Seasonal influence on patch test results in Greece. Photodermatol Photoimmunol Photomed 1992-1993; 9: 232-4.

66. Dooms-Goossens A, et al. UV sunlight and patch test reactions in humans. Contact Dermatitis 1988; 19: 36-42.

67. Bruze M. Seasonal influence on routine patch test results. Contact Dermatitis 1986; 14: 184.

68. Edman B. Seasonal influence on patch test results. Contact Dermatitis 1989; 20: 226.

69. Nethercott JR. The positive predictive accuracy of patch tests. Immun Allergy Clin N Am 1989; 9: 549-53.

70. Diepgen TL, Coenraads PJ. Sensitivity, specificity and positive predictive value of patch testing: the more you test, the more you get? Contact Dermatitis 2000; 42: 315-7.

71. Brod BA, et al. To patch or not to patch: what is your threshold for patch testing? Am J Contact Dermat 1999; 10: 94-7.

72. Rietschel RM, Fowler JF. Fisher's contact dermatitis. 4th ed. Baltimore, MD: William & Wilkins, 1995: 14.

73. Bruze M. What is a relevant contact allergy? Contact Dermatitis 1990; 23: 224-5.

74. Van Ginkel CJ, et al. Allergy due to topical medications in chronic otitis externa and chronic otitis media. Clin-Otolaryngol 1995; 20: 326-8.

75. Devos SA, Mulder JJ, van der Valk PG. The relevance of positive patch test reactions in chronic otitis externa. Contact Dermatitis 2000; 42: 354-5.

76. Mitchell JC. The angry back syndrome: eczema creates eczema. Contact Dermatitis 1975; 1: 193-4.

77. Bruynzeel DP, et al. The excited skin syndrome (angry back). Arch Dermatol 1986; 122: 323-8.

78. Pasche-Koo F, et al. How to better understand the angry back syndrome. Dermatol 1992; 184: 237-40.

79. Dooms-Goossens A. Sensitisation to corticosteroids. Consequences for anti-inflammatory therapy. Drug Safety 1995; 13: 123-9.

80. Wahlberg JE, et al. Sterile cutaneous pustules: a manifestation of primary irritancy? J Invest Dermatol 1981; 76: 381-3.

81. Katsarou A, et al. Frequency of immediate reactions to the European standard series. Contact Dermatitis 1999; 41: 276-9.

82. Lidén C, et al. In: Goyer RA, Klaassen CD, Waalkes MP, eds. Metal toxicology. San Diego: Academic Press, 1995: 447-64.

83. Rudzki E, et al. Hyperpigmentation from irritant patch tests. Contact Dermatitis 1977; 3: 53.

84. Kumar A, et al. Leukoderma following occupational allergic contact dermatitis. Contact Dermatitis 1999; 41: 94-8.

85. Kanerva L, et al. Contact leukoderma caused by patch testing with dental acrylics. Am J Contact Dermat 1998; 9: 196-8.

86. Cronin E. Contact Dermatitis. Churchill Livingstone, Edinburgh, London and New York, 1980.

87. Bruze M. Simultaneous patch test sensitization to 4 chemically unrelated compounds in a standard test series. Contact Dermatitis 1984; 11: 48-9.

88. Devos SA, van der Valk PG. The risk of active sensitization to PPD. Contact Dermatitis 2001; 44: 273-5.

89. Romano A, et al. A diagnostic protocol for evaluating nonimmediate reactions to aminopenicillins. J Allergy Clin Immunol 1999; 103: 1186-90.

90. Wen ZM, et al. Skin testing in patients with high risk of anaphylactic reactions to penicillin. Asian Pac J Allergy Immunol 1993; 11: 13-8.

91. Hannuksela M, et al. The repeated open application test (ROAT). Contact Dermatitis 1986; 14: 221-7.

92. Jackson EM. Prognostic patch testing: the other kind of patch test. Review. Am J Contact Dermat 1998; 9: 237-9.