Drug-induced urticaria

ARTICLE

Definitions and terminology

Intolerance, allergy or pseudo-allergy?

The definitions proposed are based on modern terminology used in allergology and clinical immunology [3].

- Drug intolerance: this term covers all the secondary effects of medication whether of allergic appearance or not.

- Allergy: we reserve this term for events due to specific immunity triggers, antibodies or T lymphocytes specific for drugs. Allergy is thus synonymous with hypersensitivity reaction (HSR). In the case of urticaria the triggers are principally the IgEs (HSR type I, which is still termed anaphylaxis) but they can also be IgG (HSR type II) or circulating immune complexes (HSR type III).

- Pseudo-allergy: we reserve this term for events which have all the characteristics of allergies but do not have the underlying triggers of immunity. In the case of urticaria, it is the activation of the innate immune system which is in cause: the non-immunological activation of mastocytes, activation of the complement (anaphylatoxines) or the blockage of enzymatic activity (cyclo-oxygenase, bradikinines).

Urticaria, angioedema, Quincke's disease

- Urticaria: cutaneous reaction consisting of a violent eruption of pruriginous papules, oedematous, with clear borders, of very variable shape, size and location, in which each papule lasts only for a few hours and is related to the liberation of chemical mediators by mastocytes in the dermis, of which the prinicpal one is histamine.

- Deep urticaria (synonym of angioedema): urticaria reaching the integumentary zones or the deeper subcutaneous tissue resulting in the development of urticarial lesions at lower skin levels.

- Quincke's disease (syn: Quincke's angioedema): a form of clinical urticaria where the symptoms appear suddenly with swelling of the lips, eyelids, and even of the tongue, the pharynx or the larynx; the lesions consist of white or rose-coloured oedema, slightly or not at all pruriginous, transient, not lasting more than 24 to 48 hrs.

Physiopathology of drug-induced urticaria

To resume the points detailed below, drug-induced urticaria is only rarely of allergic origin, implying IgEs specific to the medication. On the other hand it is very often evidence of a pseudo-allergic reaction and thus mainly encountered in four circumstances: 1) in the child during antibiotic treatment; 2) in patients suffering from "idiopathic" chronic urticaria; 3) in patients suffering from intolerance to aspirin and non-steroid anti-inflammatories; 4) finally, during treatment with inhibitors of angiotensin enzyme conversion.

The difficulty in the physiopathological diagnosis of drug-induced urticaria is the result of different factors, among which are:

- The poor level of knowledge which exists about the physiopathology of urticaria apart from the activation by specific IgEs. Urticaria is considered as the prototype of IgE specific reactions; prick tests to air-borne allergens show, in atopic individuals, the development of an urticarian papule at the site of an allergen prick which leads us to believe, a priori, that all urticaria is due to mastocyte degranulation connected with IgE. In reality urticaria is only rarely the result of a hypersensitivity reaction.

- The fear of many doctors concerning the potential dangers of conducting an immuno-allergic evaluation where there is a risk of shock during the skin tests or during the provocation/re-introduction tests. It is certainly important to recommend the exclusion of the suspect medicine and of molecules of the same family or those having a similar structure [2]. Carrying out an evaluation of the clinico-biological factors specific to drug intolerance is nevertheless recommended, and should include at least cutaneous tests, the determination of specific IgE if possible and current biological tests. This evaluation will lead to more precision regarding the risk of IgE response and the proposal, in the case of uncertainty, of alternative medication.

- The lack of knowledge about the immunology of medicinal haptens. These medications are chemicals which should interact with the amino acids of exogen or endogen proteins to induce an immune cell response (T lymphocyte) and humoral response (specific antibodies) [5-7]. All medicinal drugs work in the same way from an immunological point of view as the patients who develop allergic IgE responses must first of all become sensitised to the molecule, i.e. induce LT and LB specific activation resulting in the production of specific IgEs. The sensitised patients then develop, when in contact with the medicine, an activation of the mastocytes/basophils bearing specific IgEs which induces the anaphylactic shock. Drug-related immuno-allergic accidents constitute a minute percentage of intolerance accidents; the best known example is allergy to muscle relaxants (curare), which is responsible for one case of anaphylactic shock in 10,000 general anaesthesias, with a grade IV shock in one case out of 100,000 and death in one case per million [8]. These figures are nowhere near the overall frequency of intolerance to general anaesthetics which runs at 1%.

- The large number of patients who develop urticaria to different medications has led to the belief that they could be allergic to different molecules which are not structurally related and thus produce specific IgEs against several drugs. In fact this possibility is very unlikely as there is no background which favours the occurrence of anaphylatic shock due to medication; in particular atopy does not predispose to drug-induced anaphylactic shock. The induction of a specific immune response to a medication follows the classic rules for the induction of immune responses. If one considers that a patient has one chance in 10,000 of developing anaphylaxis to a drug, his chances of developing anaphylaxis to several drugs are even smaller. And yet clinical practice shows us that among intolerant patients the majority develop urticarial lesions when using antibiotics from different classes and also with aspirin, NSAIDs and codeine. These patients are by definition pseudo-allergic and the results of immuno-allergological evaluations favour this diagnosis. The medicines re-introduced are well tolerated.

- The use of cutaneous drug tests for the diagnosis of IgE-dependent allergies is not yet completely validated except in the case of penicillin-class antibiotics [9] and the curares [8]. It should also be noted that the positivity of the tests varies with time, with negative results several months or years after the anaphylatic event.

Drug-induced urticarias with an allergic origin

These urticarias are due to specific IgEs and are generally severe and associated with other signs of anaphylaxis. They represent a low proportion, less than 10%, of acute drug-induced urticarias.

The existence of an allergy due to specific drug-induced IgEs is the definition of anaphylaxis, in which the clinical signs are secondary to the liberation of amines by mastocytes and basophils covered with specific IgEs. The medication interacts with the surface IgEs culminating in their bridging and the degranulation of the mastocytes and/or basophils. The activation of cutaneous and mucous mastocytes is responsible for the signs in the tissues while the activation of circulating basophils is responsible for the shock. Urticaria is only one sign of anaphylaxis which includes a collection of cutaneous (erythema, urticaria), digestive (nausea, vomiting), respiratory (cough, dyspnea, bronchospasms, respiratory arrest) and cardio-vascular signs (tachycardia, hypotension, shock, cardio-vascular insufficiency). These symptoms can be isolated or associated and a classification of anaphylactic shock with four degrees of severity enables us to appreciate its seriousness.

Isolated anaphylactic urticaria is possible, but rare [10]. Generally it is accompanied by other serious signs which precede the beginning of the urticaria. It is associated with angio-oedema. The context of its appearance is suggestive. The accident occurs in the minutes following the taking of the medication. It is more or less rapid depending on the method of administration, by mouth, by IV or by IM. It must be emphasised that all anaphylatic shocks are not accompanied by urticaria. The classic example is anaphylactic shock during anaesthesia [8, 11]. This shock, which can be due to anaesthetics in general or to prophylactic antibiotherapy begins with respiratory, digestive and cardio-vascular signs from the moment of the injection of the product. Generalized erythema is often the first cutaneous sign and urticaria with Quincke's angioedema only appears afterwards, if at all.

Drug-induced anaphylactic urticaria can occur during the course of any treatment. It is rarely an isolated event. A certain number of clinical features favour an IgE-dependent mechanism:

- rapid generalized urticaria with mucous involvement responsible for dysphonia and dyspnea, highly agitated state of the patient;

- the association of signs of severe shock necessitating the intervention of emergency procedures and the initiation of the only physiopathological treatment available, the subcutaneous injection of adrenaline;

- the rapidity of the accident after the first exposure to the medication and in the minutes following taking it. In fact IgE mediated urticaria occurs in a sensitized individual who has specific IgEs on his mastocytes and basophils. As soon as a sufficiently high serum level of the drug is reached, cellular activation takes place.

All haptens are capable of inducing an allergic IgE response in a given individual. This is exceptional and depends on complex factors including the state of the immune system of the host during sensitisation, the existence of associated immunosuppression, the path of administration, its chemical nature and its capacity for inducing activation of the innate immune system. All medicines are thus potentially capable of causing anaphylactic shock. The list of molecules at the origin of anaphylactic shock is long and represents practically all classes of medicines, including anti-histamines [12] and contrast iodine products which were considered until recently as exclusively responsible for pseudo-allergic accidents [13]. However, certain molecules appear more immunogenic than others and are most frequently found to be the cause of anaphylactic accidents: these are antibiotics, especially the beta-lactamines, the muscle relaxants (curares) and the hypnotics.

The immuno-allergological evaluation confirms, in the large majority of cases, the existence of an IgE allergy [14]:

- Cutaneous prick tests and/or intra-dermal positive reactions with the medication. When these tests are rigorously undertaken they seem to have a good specificity and sensitivity. The cutaneous tests enable us to study cross-reactivity by testing a number of molecules of the same family as that which gave rise to the accident, in order to establish which molecules are not recognised by the specific IgEs and could be prescribed for the patient in case of need [15].

- Liberation of histamine and expression of CD63 antigen by the patient's basophils when in contact with the drug [16]. Contrary to the usual advice, these tests seem to be useful and are in the process of being re-evaluated.

- Specific IgEs in the cases of penicillins (penicillin G, amoxicillin and ampicillin) [9] and the muscle relaxants [11]. Numerous research projects are attempting to validate a group of active agents capable of analysing the IgE response to all the penicillins and cephalosporines [17]. The kits testing other molecules have not been validated.

Avoidance of the molecule responsible and molecules of the same family or molecules sharing cross-reactivity is obligatory. An allergy card is provided.

In summary and in practice, IgE allergy to medication is a rare occurrence if one compares the number of accidents with the number of patients treated. Urticaria is part of the picture of anaphylactic shock but its occurrence in isolation with no other serious signs is unusual in the case of an IgE allergy.

Pseudo-allergic drug-induced urticaria

Progress in the understanding of IgE-allergic events and the development of systematic immuno-allergological evaluation in certain specialised centres has enabled it to be shown that the majority of urticarias which develop during the taking of medication are pseudo-allergic and do not involve specific IgE.

Several different syndromes illustrate this:

- exanthema and urticaria associated with the taking of antibiotics in children;

- urticaria due to intolerance of aspirin and NSAIDs;

- chronic idiopathic urticaria;

- urticaria and angioedema during the taking of angiotensin conversion enzyme inhibitors.

Drug-induced urticaria of the child

Acute urticaria in the child proves in 80% of cases to be infectious in cause, with or without antibiotic treatment. The difficulty is in dissociating these two factors. In a study of children who experienced events which appeared to be of immediate hypersensitivity with generalised or localised erythema, urticaria with or without Quincke's disease during the taking of beta-lactamines, Ponvert et al. found less than 10% of truly allergic patients, defined as those having immediate cutaneous test reactions to penicillins and/or specific IgEs [9]. In the 90% of children with negative test results, the reintroduction of the molecule responsible for the initial event took place without incident. As penicillin was not given to patients who had positive tests, the figure of 90% of pseudo-allergic reactions is certainly a minimum. Some children develop urticarial reactions to antibiotics from different families, underlining the non IgE nature of the reactions [19].

The physiopathology of pseudo-allergic reactions is not clear. They occur in sick, feverish, infectious children with an inflammatory syndrome. The modifications induced by the infection and the innate inflammatory immune response (cytokines, chemiokines) certainly lead to activation of cutaneous cells (endothelial cells, mastocytes) which will be more sensitive to the toxic effect of the medication. Two factors seem therefore to be necessary for the triggering of the reaction: the inflammatory syndrome and the drug. If only one of these factors exist, there is no event. So taking the medicine is well tolerated when there is no longer an inflammatory syndrome, which enables us to exclude immediate hypersensitivity. The appearance of a new infection is not accompanied by an urticarial eruption if there is no medication. On the other hand the urticaria may (or may not) be repeated during a later infection treated with an antibiotic of the same class as that which led to the original reaction.

Intolerance of aspirin and NSAIDs

The provocation or aggravation of urticaria during the taking of NSAIDs is frequent [2, 20]. The eruptions occur when aspirin or one of the NSAIDs, mainly one of the old generation (anti-COX-1), is taken. All the NSAIDs can be responsible and not the molecule from one particular class. It is therefore a typical pseudo-allergy due to the pharmacological effect of NSAIDs, the inhibition of the production of prostaglandins from arachidonic acid by the effect of anti-cyclo-oxygenase (COX) [21].

Physiopathology and therapeutic consequences

The blockage of the COX pathway by the NSAIDs has the consequence of blocking the production of prostaglandins with intra-cellular accumulation in the first stages and their derivation towards the lipoxygenase pathway resulting in the excessive production of leukotrienes [20] (Fig. 1). Thus the tables for clinical intolerance to NSAIDs are due to excessive production of leukotrienes, in particular the sulphide leukotrienes LTC4, LTD4 and LTE4. The use of anti-leukotrienes (lipoxygenase inhibitors and leukotriene-receptor antagonists) would appear to be interesting in the treatment of clinical manifestations of intolerance to NSAIDs [22, 23].

Two COX isoenzymes exist: COX-1 is a ubiquitous cellular enzyme (expressed in the make-up of the majority of tissues), and the blocking of its functions by classic NSAIDs (anti-COX-1) is at the origin of the usual side effects like kidney failure, bleeding, stomach ulcers and clinical signs of intolerance. The constitutional expression of COX-2 is limited to certain tissues (central nervous system) but it is induced by the pro-inflammatory cytokines and by endotoxins and thus implicated in the production of inflammatory prostaglandins. The preferred anti-COX-2 (nimesulide) or selective anti-COX-2s which consist of the latest of generation NSAIDS (meloxicam, celecoxib, rifecoxib), appear to be particularly interesting NSAIDs as they are directed against the COX induced by the inflammatory reaction. They represent an alternative therapy which could be of great interest for those patients who are intolerant of NSAIDs, as is shown in the results of recent studies [24, 25].

Clinical manifestations

The term "Intolerance to NSAIDs" includes different clinical presentations [20]:

- Widal's syndrome associates intrinsic asthma, nasal polyps, tissular eosinophilia and chronic rhino-conjunctivitis. The symptoms induced by taking aspirin or any other NSAID concern the whole ear, nose and throat (ENT) and respiratory area and can lead to fatal bronchospasms. The cutaneous signs, including urticaria, are less common.

- Patients suffering from "chronic idiopathic urticaria" may break out in eruptions of urticaria or angio-oedema after taking NSAIDs. As a general rule there is no important respiratory reaction.

- Atopy appears to be an important factor of risk for intolerance to NSAIDs, with a prevalence of 80% of atopic patients in any group of intolerant patients. The signs of intolerance are mainly urticarial and rarely respiratory [26].

- Patients with no particular antecedents develop urticarial or anaphylatic reactions to a particular class of NSAIDs like the pyrazoles or diclofenac.

Diagnosis

The diagnosis of NSAID intolerance is clinically straightforward and can be made with provocation tests during investigations for respiratory functioning and in a hospital setting. The sensitivity of patients to NSAIDs is very variable and depends, in each patient, on the type of NSAID used. Certain molecules seem to be more pro-inflammatory than others.

Treatment

The total avoidance of aspirin and the NSAIDs is only mandatory in the clinically serious forms or in the case of ENT and respiratory manifestations, which necessitate special treatment. The selective anti-COX-2s are generally well-tolerated and represent a first therapeutic possiblity. The anti-leukotrienes are efficient in the prevention of adverse events when taking aspirin is absolutely necessary. Finally it is possible to induce tolerance to NSAIDs by gradually increasing doses of the molecules to the patients [27]. These protocols are reserved for patients who need aspirin or NSAIDs continuously.

Chronic idiopathic urticaria aggravated by medication

Chronic idiopathic urticaria (CIU) is very common. Dermographism concerns 5% of the population. The drugs can be responsible for a worsening of the urticaria, inducing urticarial eruptions in patients with active chronic urticaria or in remission for several years, as well as in patients suffering from dermographism. The drugs most commonly responsible are antibiotics, particularly the penicillins, vancomycin, contrast iodine products, the histamine-liberating molecules (morphine and codeine), local anesthetics, muscle relaxants, and, of course, aspirin and the NSAIDs. The urticaria is usually unimportant, beginning in the minutes or hours after taking the medicine and in the majority of cases without any serious signs (shock), even though the mucous membranes are often involved.

The diagnosis of CIU is easy, either by questioning the patient - provided that one thinks to do it - or by using a dermographism test. The diagnosis can be more difficult if the patient has been in remission from CIU for several years. The fact that a patient presents urticarial eruptions after taking different molecules which are not structurally related (e.g. NSAIDs, codeine and penicillin) is definitive for pseudo-allergic reaction and very suggestive for the diagnosis. Another characteristic pointer is the systematic non-reproducibility of the lesions during subsequent taking of the same medication or during re-introduction/provocation tests. The immuno-allergological prick and intra- dermal tests are always negative but can give false positives in patients suffering from dermographism.

The physiopathology of CIU is not known but is probably connected to chronic activation of the mastocytes with a lowering of the degranulation threshold in response to physical and chemical stimuli. The medication would therefore play the role of an additional irritant factor, which would lead to the degranulation of pre-activated mastocytes.

Prevention of adverse events is possible in the majority of cases and depends on the intensity and the frequence of the eruptions. Taking anti-histamines a few hours before starting treatment is often enough. The association of an anti-leukotriene can help to prevent any sort of urticarial crisis in very sensitive patients.

Angioedemas due to inhibitors of the angiotensin converting enzyme (ACE)

The ACE inhibitors can induce several types of cutaneous reaction, in particular captopril which contains a thiol group. Among these are urticarial reactions, bullous lesions and phototoxic reactions [20]. The classic secondary effects are a cough, and angioedemas of the face, mucous membranes, the extremities and the visceral organs, usually without associated urticaria. It concerns 0.5% of patients treated. These adverse events, seen with all the ACE inhibitors, occur classically in the first weeks of treatment, but the start of the problems may be delayed for several months and may even occur after several years of treatment [28]. The eruptions occur at irregular and unpredictable intervals. The presence of a cough or angioedema during treatment necessitates the immediate cessation of treatment due to the risk of possible severe complications (fatal angioedema of the larynx). The drugs used in the treatment of anaphylactic-like angioedema are not efficacious and intubation should be rapidly undertaken in the event of involvement of the larynx.

The adverse events may be due to the blocking of the angiotensin converting enzyme with accumulation of bradikinin, which has vasodilatory properties (Fig. 2). But the angiotensin II receptor antagonists (the sartans), which have no effect on the metabolic functions of bradikinin, are also responsible for adverse events similar to those produced by ACE inhibitors and thus are contra-indicated in patients with a history of adverse reactions. It should also be noted that with patients under ACE treatment, other medication (antibiotics, anaesthetics) as well as trauma can induce urticaria or angioedema [20].

The diagnosis is made on the basis of patient history and clinical evidence. There are no cutaneous or biological tests currently available.

CONCLUSION

The occurrence of urticaria during the course of drug treatment requires a precise diagnostic approach:

- Record the adverse event as "chronic idiopathic urticaria" as long as the patient developed the urticarial eruptions with several groups of drugs, antibiotics, NSAIDs, contrast products, local and general anaesthetics, medicines containing codeine. It remains possible that a patient subject to CIU can one day develop hypersensitivity to a drug. It must therefore not be forgotten that an allergological evaluation should be undertaken in case of severe adverse events.

- Include urticaria in the "intolerance to aspirin and NSAIDs" syndrome and undertake respiratory function tests and an ENT examination to look for Widal's syndrome, where the prognosis is more serious than in an isolated urticaria due to NSAIDs.

- Stop treatment with ACE inhibitors in patients who develop angioedema.

- Above all rule out immediate hypersensitivity, particularly where the urticaria is severe and associated with signs of anaphylaxis and where it occurs in a patient with no antecedents. An immuno-allergolgoical evaluation is essential in this case and should be associated with cutaneous and biological tests. IgE allergy to drugs is rare if we compare the number of adverse events with the number of patients treated. It is serious because it is potentially fatal through anaphylactic shock. There are 500 deaths annually in the USA due to penicillin shocks [18]. This seriousness justifies the investigation of patients in a specialised unit and the carrying-out of an immuno-allergological evaluation. Urticaria is part of the picture of anaphylactic shock but its occurrence in isolation with no other serious signs is unusual in the case of IgE allergy.

Acknowledgements

We are indebted to Jenny Messenger for translating this article.

REFERENCES

2. Grosshans E. Urticaires médicamenteuses. Rev Prat 2000; 50: 1305-9.

3. Demoly P, Messaad D, Benhamed S, et al. Immunoallergic reactions of drug origin: epidemiologic and clinical data. Therapie 2000; 55: 13-9.

4. Civatte J. Dictionnaire de Dermatologie. Éditions CILF, Paris, France, 2000.

5. Pichler W, Yawalkar N. Pathophysiology of drug-elicited exanthems. ACI International 2000; 12: 166-70.

6. Weltzien H, Padovan E. Molecular features of penicillin allergy. J Invest Dermatol 1998; 110: 203-8.

7. Griem P, Wulferink M, Sachs B, Gonzalez J, Gleichmann E. Allergic and autoimmune reactions to xenobiotics: how do they arise? Immunol Today 1998; 19: 133-41.

8. Laxenaire M. Substances responsables de choc anaphylactiques per-anesthésiques. Troisième enquête multicentrique française. Ann Fr Anesth Reanim 1996; 15: 1211-8.

9. Ponvert C, LeClainche L, deBlic J, LeBourgeois M, Scheinmann P, Paupe J. Allergy to beta-lactam antibiotics in children. Pediatrics 1999; 104: F1-9.

10. Torres M, Mayorga C, Pamies R, et al. Immunologic response to different determinants of benzylpenicillin, amoxicillin and ampicillin. Comparison between urticaria and anaphylactic shock. Allergy 1999; 54: 936-43.

11. Guilloux L, Ricard-Blum S, Ville G, Motin J. A new radioimmunoassay using a commercially available solid support for the detection of IgE antibodies against neuromuscular blocking drugs. J Allergy Clin Immunol 1992; 90: 153-9.

12. Demoly P, Messaad D, Sahla H, Bousquet J. Hypersensitivity to H1-antihistamines. Allergy 2000; 55: 679-80.

13. Laroche D, Namour F, Lefrancois C, et al. Anaphylactoid and anaphylactic reactions to iodinated contrast material. Allergy 1999; 54 (suppl. 58): 13-6.

14. Solensky R, Earl H, Gruchalla R. Clinical approach to penicillin allergy: a survey. Ann Allergy Asthma Immunol 2000; 84: 329-33.

15. Harris A, Sauberman L, Kabbash L, Greineder D, Samore M. Penicillin skin testing: a way to optimize antibiotic utilization. Am J Med 1999; 107: 166-8.

16. Monneret G, Benoit Y, Gutowski M, Bienvenu J. Detection of basophil activation by flow cytometry in patients with allergy to muscle relaxant drugs. Anesthesiology 2000; 92: 275-7.

17. Baldo B. Diagnosis of allergy to penicillins and cephalosporins. Structural and immunochemical considerations. ACI International 2000; 12: 206-12.

18. Adkinson N. Betalactam cross-reactivity. Clin Exp Allergy 1998; 28: 37-40.

19. Park J, Matsui D, Rieder M. Multiple antibiotic sensitivity syndrome in children. Can J Clin Pharmacol 2000; 7: 38-41.

20. Bircher A. Drug-induced urticaria and angioedema. Eur J Dermatol 1999; 9: 657-63.

21. Szczeklik A, Sanak M. Molecular mechanisms of aspirin-induced asthma. ACI International 2000; 12: 171-6.

22. Yoshida S, Ishizaki Y, Onuma K, Shoji T, Nakagawa H, Amayasu H. Selective cyclo-oxygenase 2 inhibitor in patients with aspirin-induced asthma. J Allergy Clin Immunol 2000; 106: 1201.

23. Dahlen B, Szczeklik A, Murray J. Celecoxib in patients with asthma and aspirin intolerance. The Celecoxib in Aspirin-Intolerant Asthma Study Group. N Engl J Med 2001; 344: 142.

24. Bavbek S, Celik G, Ediger D, Mungan D, Demirel Y, Misirligil Z. The use of nimesulide in patients with acetylsalicylic acid and nonsteroidal anti-inflammatory drug intolerance. J Asthma 1999; 36: 657-63.

25. Quaratino D, Romano A, Di Fonso M, et al. Tolerability of meloxicam in patients with histories of adverse reactions to nonsteroidal anti-inflammatory drugs. Ann Allergy Asthma Immunol 2000; 84: 613-7.

26. Sanchez-Borges M, Capriles-Hulett A. Atopy is a risk factor for nonsteroidal anti-inflammatory drug sensitivity. Ann Allergy Asthma Immunol 2000; 84: 101-6.

27. Wong J, Nagy C, Krinzman S, Maclean J, Bloch K. Rapid oral challenge desensitization for patients with aspirin-related urticaria-angioedema. J Allergy Clin Immunol 2000; 105: 997-1001.

28. Schiller P, Langauer Messmer S, Haefely W, Schlienger R, Bircher A. Angiotensin-converting enzyme inhibitor induced angioedema. Late onset, irregular course and potential role of triggers. Allergy 1997; 52: 432-5.