Although allergic drug reactions are often reported by numerous patients, in many instances the true nature and classification of these reactions remains undetermined. To assist our understanding of these reactions to drugs, adverse responses to drugs are often categorized as follows: (1) drug intolerance --- A predictable side effect at low to therapeutic doses due to altered drug metabolism or end-organ hyperacuity; (2) idiosyncratic drug reactions, and (3) immunologic drug reactions (AKA drug allergy) --- due to a drug-specific immune response. Examples of each are shown below.

Allergic reactions can be subdivided into four types: Type I (immediate), Type II (cytolytic), Type III (immune-complex associated), and Type IV (delayed)¹.

Symptoms of adverse drug reactions in descending order include: skin reactions (80%), anaphylaxis (9 - 15%), respiratory symptoms (6 - 9%), and drug fever (2 - 6%)³.

A distinction should be drawn between anaphylaxis and anaphylactoid (or pseudo-allergic) reactions. Anaphylaxis refers to a systemic, immediate hypersensitivity reaction due to the IgE-mediated release of mediators from mast cells and basophils. An anaphylactoid event refers to a clinically similar event not mediated by IgE. They cause, via an unknown mechanism, a degranulation of mast cells and/or basophils. Examples of anaphylactoid reactions include urticaria associated with ASA or other NSAIDS or urticaria and angioedema after use of radiologic contrast media⁴. Syndromes such as Stevens-Johnson syndrome and toxic epidermal necrolysis are often drug-associated and thought to be immune mediated although the specific mechanism remains unclear.

Pathophysiologic Classification of Anaphylaxis and Anaphylactoid Reactions

Anaphylaxis; 1gE-mediated reaction
    Food
    Drugs
    Insect bites and stings
    Perhaps some cases of exercise
Anaphylactoid
    Direct release of mediators from mast cells and basophils
        Drugs
        Idiopathic
        Exercise
        Physical factors, such as cold, sunlight
    Disturbances in arachidonic acid metabolism
        Aspirin
        Nonsteriodal anti-inflammatory drugs
    Immune aggregates
        Gammaglobulin
        IgG-anti-lgA
        Possibly dextran and albumin
    Cytotoxic
    Transfusion reactions to cellular elements
    Miscellaneous
        Non-antigen-antibody-mediated complement activation
            Radiocontrast material
            Possibly some cases of protamine reactions
            Dialysis membranes
        Activation of contact system
            Dialysis membranes
            Radiocontrast material

Drugs alone are poor stimulators of immune responses due to their simple structure and low molecular weight. Generally, an antigen must be presented to the immune system in a multivalent form to elicit a specific immune response. Valency refers to the number of binding sites available to bind to antibody. Multivalency is necessary to ensure cross-linking of receptors on the surface of cells, which then causes transduction of the signal within the cell and the initiation of an immune response. Drugs can fulfill the requirement for multivalency and cause an immune response in two ways. The first way is for drugs to form multivalent hapten-carrier complexes. A hapten in this case would be a particular drug, which would be immunogenic in protein-conjugated but not free form. For example, penicillins and other beta-lactam antibiotics bind covalently to proteins. Other drugs may become noncovalently associated with proteins. A second way for drugs to elicit an immune response is to be converted to reactive intermediates during drug metabolism in the liver or elsewhere. This is the case with sulfonamides, which are acetylated and oxidized to yield the predominant N4-sulfonamidoyl hapten.

Evaluation of drug allergy in a patient receiving multiple medications should be done systematically. First, obtain a complete drug reaction history, atopic history, complete medication list, chronology of all symptoms and signs (including non-specific signs such as eosinophilia, sedimentation rate, tryptase, etc.). Second, narrow the list of medications suspected based on the temporal association between drug starts and stops and changes in dose. Next, stop and/or substitute all drugs with known allergic potential begun on the day of or several days prior to the reaction. If the suspected drug cannot be substituted, skin testing can be used to assess IgE response. Negative prick-puncture tests may be confirmed by more sensitive intradermal techniques. Currently, skin testing is accepted to test only for penicillin allergy. Radioallergosorbent (RAST) testing and the lymphocyte transformation test can also be used to evaluate for drug allergy.

In RAST testing, a given allergen is covalently bound to polydextran beads; serum is added and antigen specific IgE if present will bind to the immobilized antigen. After washing off unbound antibodies, radiolabeled anti-IgE is added. The amount of bead-bound radioactivity after washing off unbound labeled antibody is directly related to the concentration of antigen-specific IgE present in the serum. Although this test is highly sensitive and accurate, it is expensive and the range of antigens for which there are available tests is limited. Skin testing is preferred for the diagnosis of drug allergy since it allows the testing of a wider array of antigens and also the results of skin testing appear to correlate better with symptoms. For instance, a patient could be positive by RAST to several allergens but be asymptomatic⁵.

The lymphocyte transformation test can also be used to evaluate for drug allergies but cannot evaluate reactions to drugs that are metabolized in vivo, such as sulfonamide metabolism in the liver, or detect antibodies to the hapten-carrier complex. Just how useful the LTT is to detect drug allergy remains to be seen.

Finally, readminister the suspected drug if necessary with gradual escalation of the dose (if skin test negative) or desensitize (if reaction is IgE - dependent or skin test positive). Rechallenge should generally begin at 1% of the desired therapeutic dose and increased incrementally (three-fold) at intervals determined by the half-life of the drug and the patient’s prior experiences with it. This is done to ensure that very mild reactions will be noticed if the patient remains sensitive. Once the full therapeutic dose is achieved without complications, continuous therapy should begin. Rechallenge of drugs should not be performed in patients with Stevens-Johnson syndrome, toxic epidermal necrolysis, or with any mucous membrane involvement.

Drug desensitization is effective in the treatment of type I allergic reactions and may be effective for other reactions that are delayed in onset but are not IgE-mediated. This procedure should be distinguished from allergen immunotherapy. Antigen-specific mast cell desensitization appears to be responsible for the tolerant state. Both IgE- and IgG-specific antibodies increase during postdesensitization therapy. The rising IgG titer may neutralize epitopes and serve a "blocking" function similar to that observed with allergen immunotherapy. Skin test reactivity, however, diminishes as a result of desensitization despite a large increase in drug-specific IgE.

Successful antibiotic desensitizations⁶

Successful desensitizations to other agents

Specific mast cell desensitization is poorly understood, but it may involve low-level subthreshold stimulation by hapten-carrier conjugates.

Desensitization protocols have been applied successfully to many types of drugs. Oral or parenteral administration can be used. Pretreatment with antihistamines or corticosteroids should not be used so that minimal, easily treated reactions may be identified early and to avoid having serious reactions occur later. The starting dose for the drug can be determined by performing intradermal skin tests with the native drug at a dose that does not cause a non-specific irritant reaction. For example, if a 0.02 ml intradermal injection of a drug at 1mg/ml concentration does not cause a local or systemic reaction, oral desensitization may be started at the dose injected (i.e. the tolerated dose, 20 micrograms). Parenteral desensitization should be initiated using 1/10 or 1/100 of the dose that was administered intradermally.

A Generic Protocol for Drug Desensitization for IgE-dependent Allergy

Preparation
1.    Skin test patient to determine degree of sensitivity:
    a.    Dilute available drug solutions/suspension to 1/3 mg/ml.
    b.    Prepare three tenfold dilutions.
    c.    Perform prick-puncture testing with 1:1000 dilution
    d.    If negative, serial intradermal tests (0.02 ml [2-4mm bleb] in duplicate) up to and including 3 mg/ml stock; discontinue testing when >8 m wheal 
           is observed.  Test is positive if both duplicate wheals increase significantly (>2-3 mm) 20 min after placement compared with diluent control.
2.    Prepare sufficient quantities of drug solution/suspension for desensitization regimen in half-log 10 dilutions (threefold and tenfold dilutions from
       concentrate [1-3 mg/ml]).
Procedure
1.    Establish baseline monitoring of patient in medical setting appropriate for patient's clinical conditions and the nature and severity of the prior reaction.
       Start a secure intravenous infusion.
2.    Starting dose: If skin test negative and test is unvalidated, begin with 0.1 ml of 1/3 microgram/ml solution/suspension; if skin test positive, begin
       100-fold below the dose producing a midpint (5-8 mm wheal) reaction.
3.    Route; oral by ingestion or ng tube in 30 ml water; parenteral by intradermal (<0.2 ml), SC (0.2-0.6 ml), or intramuscular (>0.6 ml) injection.
4.    Dosing interval: 15-20 min for parenteral doses; 20-30 min for oral dosing.  Repeat dose for mild systemic reaction: drop back two doses (tenfold) for
       moderate reactions, further for any reactions producing hemodynamic changes.
5.    Dose escalation; half-log 10 (-threefold) increments; e.g., 1 mg, 3 mg, 10 mg, 30 mg, 100 mg, etc.
6.    If IV therapy is indicated, begin infusion to deliver a dose equivalent to the last oral/parenteral dose slowly over 1 hour. Double the infusion rate every
       hour until target therapeutic dosing is achieved.
Follow-up
1.    If IV therapy follows desensitization, a continuous 24-hour drug infusion is preferable if feasible.  If not, avoid rapid infusion of intermittent drug doses.
2.    If drug skin test was positive, repeat postdesensitization to document shift in skin sensitivity.
3.    Avoid lapses in therapeutic doses.
4.    Treat through selected mild-to-moderate reactions (e.g., urticaria) to avoid need to repeat desensitization.
5.    Desensitization therapy is dose- and drug-dependent; it must be repeated before subsequent courses of therapy if skin tests remain positive.

Desensitization is a reversible process that is dependent on the continuous presence of the drug. After the drug is stopped, the desensitization state disappears and repeat desensitization is usually required for subsequent treatment courses. It is also drug-dose dependent in that a substantial dose increase may result in breakthrough allergic symptoms.

Allergic symptoms include a maculopapular, erythematous and usually pruritic rash and urticaria. Other less common symptoms include angioedema, serum sickness, arthralgias, bronchospasm, laryngeal edema, and anaphylaxis. It is interesting to note that most deaths resulting from penicillin allergy occur in patients with no history of penicillin allergy⁷. Allergic reactions due to beta-lactams can be classified as immediate if they appear within 1 hr of administration, accelerated if they appear 1-72 hrs later, and delayed if the reactions appear after 72 hrs. Overlapping between the classifications sometimes exists. Other types of reactions, such as vasculitis, eosinophilic pneumonitis, and other organ-specific reactions or fixed drug eruptions, are difficult to classify since the immunological mechanisms have not yet been clearly defined. An example of a delayed reaction would be maculopapular or morbilliform rashes associated with treatment with aminopenicillins, particularly ampicillin. The incidence of rash associated with ampicillin has been estimated at 9.5%⁸. Since the rash typically appears 2-3 days or more after drug administration, it is thought that these often represent type IV (cell-mediated) hypersensitivity. Patch test and delayed (>6 hrs after administration) intradermal test positivity in such patients also suggest a type IV mechanism. Additional evidence comes from skin biopsies, which demonstrate an infiltration of mononuclear cells with a predominance of T cells⁹. Finally, oral challenges with the suspect drug provoke delayed-onset skin rashes^9,10.

Sogn et al¹¹
A prospective study using patients with infectious diseases for which penicillin or related compounds was the drug of choice. Patients were skin tested with major and minor determinant antigens. Only if skin tests were negative, patients were given penicillin or a semisynthetic penicillin. 9 skin test positive patients were accidentally given penicillin and 2 had allergic reactions. The allergic reactions noted in the table below were urticaria, generalized erythema, and pruritis.

                                    Skin test positive             Skin Test Negative         Allergic Reactions

Skin testing is widely used to evaluate for penicillin allergy. The reagents used in testing for allergy include penicilloyl-polylysine and benzylpenicillin as a minor determinant reagent. More than 80% of history-positive patients will have negative skin tests^11-12. Allergic reactions observed in the retreatment of history-positive, skin-test negative patients have virtually all been mild and self-limited; no life-threatening false-negative reactions have been reported. The incidence of cutaneous reactions when penicillin is administered to patients with histories of penicillin allergy, but with negative skin tests, is approximately 1%. Up to 67% of patients with positive skin tests have experienced clinical allergic reactions when given therapeutic doses of penicillin¹³. For patients with a history of rash with the aminopenicillins, skin testing with extended observation for late reactions and also patch testing is recommended. Negative results with both tests appear to be a good indicator that the rash was not due to an immunological reaction but more likely due to an underlying disease state. Since recurrent rather than continuous therapy can resensitize patients to penicillin, skin tests should be repeated before subsequent courses of therapy.

Summary of Studies of Administration of Cephalosporiin to Patients with
Histories of Penicillin Allergy and Penicillin Skin Test Evaluations²¹

Allergic reactions to cephalosporins do not appear to correlate with positive penicillin skin test reactions. Currently there are no reliable cephalosporin allergens available for skin testing. The overall incidence of reported adverse reactions to cephalosporin antibiotics ranges from 1 to 10%. These reactions include rashes (1-5%), eosinophilia (3-10%), fever (<1%), gastrointestinal symptoms (3%), hematologic manifestations (1-2%), and phlebitis (2%)^14-18. The majority of allergic reactions are rashes that occur 24 hrs after drug exposure, the mechanism of which is unknown. Anaphylactic reactions are extremely rare (<0.02%). It has been reported that there is an 8.1% incidence of allergic reactions to various cephalosporins in patients with histories of penicillin allergy compared to a reaction rate of 1.9% in patients without such histories¹⁹. The overall incidence of cephalosporin allergy in the general population is about 4%, so the 8% incidence of reactions in the penicillin-allergic group is only a 2-fold increase. In fact, when penicillin-allergic patients receive any drug (unrelated to penicillin), the incidence of adverse drug reactions is 3 times higher compared to those with no history of penicillin allergy²⁰. Therefore, a 2 to 4- fold increase in the incidence of cephalosporin reactions in penicillin-allergic patients might be expected, even if there is no immunologic cross-reactivity. In conclusion, skin tests are not useful prior to administering cephalosporins to patients with histories of penicillin allergy since they do not identify patients who may have anaphylactic reactions.

Sullivan et al⁴⁸
30 patients with histories of allergic reactions to PCN, positive skin tests, and life threatening infections (bacterial endocarditis, Pseudomonas sepsis or pneumonia) were desensitized. Skin test reactions disappeared or diminished in all 23 subjects who were retested after desensitization. Full courses of antibiotic therapy and cure of the infections were accomplished in 30 of 30 patients.No deaths, anaphylaxis, or severe acute allergic reactions occurred. Pruritic cutaneous eruptions appeared in 9 patients (30%) 6 to 48 hrs after the onset of therapy. One patient developed reversible nephritis 3 weeks into therapy with PCN G.

Sulfonamides were first introduced in the 1930’s but their use was not widespread due to the availability of penicillin and the emergence of resistant bacteria. The formulation of co-trimazole, or sulfamethazole-trimethoprim, in the late 1960’s and the HIV epidemic in the 1980’s led to renewed interest in sulfa drugs clinically. Co-trimazole is used extensively in the HIV population as prophylaxis against Pneumocystis carinii pneumonia. Unfortunately, there is an unusually high incidence of adverse reactions to sulfonamides in HIV patients. It is estimated that the overall prevalence of sulfa hypersensitivity in the general population is approximately 3.3%, while in the HIV population it is in the range of 17-20% ^22,23. Up to 80% of these reactions has been reported in HIV-infected patients compared with less than 5% in HIV-seronegative patients^23,24. The incidence of adverse events to TMP/SMX is greater than 50% in patients receiving the medication for treatment of active PCP²⁵. A recent retrospective case-control study evaluating factors contributing to sulfa hypersensitivity found an association between the number of opportunistic infections and the occurrence of TMP/SMX hypersensitivity reactions²⁶. However, Hennessy et al²⁷ found no correlation between a low CD4 counts and an increased risk of hypersensitivity reactions to sulfa.

Hennessy et al²⁷
A retrospective cohort study involving patients in outpatient HIV clinic and university-affiliated internal medicine and infectious disease practices receiving cotrimazole for primary PCP prophylaxis. The outcome measured was the occurrence of a cutaneous hypersensitivity reaction (rash, fever, or pruritis) per chart review that resulted in the discontinuation of cotrimazole

Shafer et al²⁹
34 homosexual men given IV cotrimazole for PCP with development of hypersensitivity reactions in 21 of these patients (erythematous macular or maculopapular rashes, fever). All 31 survivors were started on oral cotrimazole for PCP prophylaxis but only 4 developed reactions which necessitated discontinuation of the drug (desquamative rash, fever, etc.)

The classical sulfonamide hypersensitivity reaction is a multiorgan, systemic disease characterized by fever, skin rash, and toxicity in one or more internal organs starting 7 to 14 days after initiation of therapy. Organ toxicities can also occur without fever or skin rash. The incidence of severe, life-threatening hypersensitivity reactions is less than 1/1000. Skin reactions occur in about 1.5 to 3.0% of exposed individuals. Maculopapular eruptions and urticarial rash occur most frequently within the first 1-3 days after administration of the drug, usually not accompanied by fever, and resolve spontaneously on withdrawal. Non-urticarial drug eruptions leading to more serious outcomes (Stevens-Johnson syndrome, toxic epidermal necrolysis) typically occur 7 to 14 days after initiation of a primary course of therapy. The incidence of these reactions is between 1/10000 and 3/100000. Any patients with either disease should be removed from the drug immediately and rechallenge avoided since rapid worsening of the disease may occur as a result. Other disease states associated with the hypersensitivity reaction include: hepatoxicity, eosinophlic pneumonitis, aseptic meningitis, acute interstitial nephritis, serum sickness polyarthritis, blood dyscrasias.

MECHANISMS

The mechanism of hypersensitivity to trimethoprim-sulfamethazole is poorly understood. Several hypotheses have been put forward to explain the predominance of reactions in HIV patients. These include: (1) slow acetylation in HIV patients; (2) polypharmacy with drugs such as INH and rifampin which compete for microsomal enzyme systems in the liver; and (3) reduced availability of cellular glutathione.

Sulfa drugs are metabolized in the liver by two pathways: oxidative metabolism by the cytochrome p450 system and acetylation by N-acetyltransferase. SMX is predominantly cleared by acetylation and excreted by active tubular excretion. The alternative pathway yields a reactive metabolite, sulfamethazole hydroxylamine, which can generate an immune response. Generally, slow acetylators develop more adverse reactions, whereas rapid acetylators are more prone to show an inadequate response to a standard dose³⁰. The ratio of rapid versus slow acetylators varies widely among ethnic groups throughout the world. For example, approximately 50% of Caucasians and African-Americans are "slow" acetylators while this is rare among Asians.

Lee et al³¹
Caffeine ingested and then urine was collected to determine the ratio of various caffeine metabolites.

The cause of the increased prevalence of slow acetylation in acutely ill AIDS patients is not known. Due to slow acetylation, more of the drug is shunted to alternative oxidative pathways. These pathways form toxic metabolites which are normally detoxified by scavengers, such as glutathione. A few studies have reported decreased levels of glutathione in HIV patients but this has not been confirmed in other studies^32,33. The accumulated metabolites can then cause cellular injury which may be expressed clinically as an adverse reaction.

Some investigators have reported positive skin tests in approximately 25% of patients with immediate hypersensitivity reactions to SMX, but others have not found it useful in predicting the recurrence of SMX related adverse events³⁴. A new antigen for skin testing, sulfonamidoyl-poytyrosine, has been created recently and has been shown to elicit positive type I skin tests in a small minority of patients with sulfonamide allergy by history³⁵. The negative predictive value of this reagent remains to be determined. Currently, skin testing cannot be recommended.

The earlier reports of desensitization to sulfonamides had limited success rates, ranging from 45% to 82% ^36-38. The success rates of desensitization have improved significantly over the years³⁹.

Kalanadhabhatta et al³⁹
A prospective study with 13 patients with AIDS (CD4<200) with PCP and allergy to sulfonamides who failed alternative therapy (dapsone, pentamidine). The allergic reactions noted were a generalized, pruritic maculopapular rash, urticaria, angioedema, and pruritis. All patients had tolerated oral desensitization to cotrimazole without any adverse reaction including three patients who were critically ill and on mechanical ventilation. Total follow up ranged from 4 to 84 weeks.

Shortly after aspirin was introduced over 100 years ago, ASA was found to contribute to violent attacks of asthma. In 1922, the association of ASA sensitivity, asthma, and nasal polyposis was described by Widal et al⁴⁰ and was subsequently coined as the "aspirin triad." Aspirin-induced asthma (AIA) affects 10% of adults with asthma⁴¹. After ingestion of ASA or an NSAID, an acute asthma attack occurs within 3 hours, usually accompanied by profuse rhinorrhea, conjunctival injection, and periorbital edema. NSAID intolerance prevalence in asthma and nasal polyposis or rhinosinusitis is 30-40%. In chronic urticaria, the prevalence of NSAID intolerance is 20-30%.

Clinical symptoms associated with AIA resemble immediate hypersensitivity reactions. However the search for an underlying antigen-antibody mechanism has not been successful. Skin tests with ASA-lysine have been negative, and numerous attempts to demonstrate specific antibodies against ASA or its derivatives have been unsuccessful. In patients with AIA, asthmatic attacks can be caused by ASA and other NSAIDS. After ASA desensitization, cross desensitization to other NSAIDS that inhibit COX also occurs. Szczeklik et al⁴² reported in 1975 that drug cross-reactivity could be predicted on the basis of each NSAID’s in vitro inhibition of COX. This has been consistently reaffirmed over the years^43,44. Nimesulide and meloxicam, drugs which inhibit COX-2 substantially more than COX-1, are well tolerated by ASA-sensitive asthmatics when given average therapeutic doses but cause rhinorrhea and mild asthmatic attacks when ingested in higher doses.

Three hypotheses have been proposed to explain AIA: (1) cyclooxygenase inhibition; (2) overproduction of leukotrienes; and (3) chronic inflammation of the airways. During inflammatory respiratory disease cys-leukotrienes, histamine, and eosinophilic cationic protein are formed and released with subsequent increase in vascular permeability, mucus secretion, and bronchial hyperreactivity. 5-lipoxygenase and COX catalyze the production of leukotrienes and prostaglandins, respectively. Prostaglandin E2 has several immunoregulatory effects, including inhibition of 5-lipoxygenase and preventing the release of mediators from mast cells. When ASA is given, COX-1 and COX-2 are disabled, PGE2 synthesis stops and its modulating effects on mast cells and 5-LO are removed, and mediators are released or synthesized. Leukotrienes are continuously and aggressively synthesized in patients with AIA before any exposure to NSAIDS/ASA, and during ASA-induced reactions, marked acceleration of synthesis occurs. It is not known why interruption of PGE2 synthesis does not induce respiratory reactions in all humans.

Eosinophil infiltration of airway tissue appears to be a central feature of AIA. The large numbers of eosinophils, loaded with leukotriene enzymes, may be responsible for the overproduction of leukotrienes, a differentiating feature of AIA.

Delayed reactions to NSAIDS appear to have an immunologic basis since they recur on re-exposure with a shorter latency after re-exposure. On the other hand, the rashes may be due to a direct toxic effect on the skin.

The clinical manifestations of NSAID intolerance include urticaria, angioedema, rhinoconjunctivitis, bronchial asthma, and occasionally anaphylactoid reactions. Many patients with chronic urticaria experience an increase in cutaneous lesions after ingesting ASA. Stevens Johnson syndrome and toxic epidermal necrolysis have been associated with NSAIDS in 14% and 19% of cases, respectively⁴⁵. The most frequent delayed cutaneous reaction is a morbilliform exanthem, which usually occurs 1 week after therapy is begun and can last 1 – 2 weeks. Vasculitis and photosensitivity have also been observed.

Numerous attempts to demonstrate specific antibodies against ASA or its derivatives were unsuccessful. Skin test responses with ASA-lysine have been negative. Currently, the only way to test NSAID intolerance is with provocation tests. There are three types of provocation tests: oral (most common), inhaled and nasal. Only oral ASA challenges are available in the US. A standard 3 day protocol is described below².

Patients with a history of severe or very rapid ASA or NSAID-induced reactions may be started with a dose lower than 30 mg. As soon as signs and symptoms of reactions occur (20% decrease in FEV1, rhinorrhea, ocular injection, periorbital edema, stridor and rarely flushing, urticaria, cramps, or explosive diarrhea), the ASA challenge is stopped. Oral challenges to detect anaphylaxis should not be done; history of anaphylaxis should be relied upon instead.

NSAIDS that cross react with ASA in respiratory and cutaneous reactions:

Piroxicam mefenamic acid

Flurbiprofen diflunisal

Ketoprofen diclofenac

Ketorolac etodolac

Nabutemone oxaprozin

Indomethacin Sulindac

Tolmetin Zomepirac

Ibuprofen Naproxen

naproxen sodium fenoprofen

meclofenamate

Patients should avoid ASA and other analgesics that inhibit COX.

Acetaminophen is usually safe for these patients to take. However, high-dose acetaminophen (1000 mg followed by 1500 mg) has been shown to have a cross-reaction prevalence of 34% ⁴⁶. Animal models have demonstrated that acetaminophen may have COX inhibitory activity. Patients can also take sodium salicylate, salicylamide, choline magnesium trisalicylate, benzydamine, chloroquine, azapropazone, and dextropropoxyphene. The drugs just mentioned are all without anti-COX activity or are weak anti-COX-2 inhibitors. Selective inhibitors of COX-2, in theory, should be safe in patients with AIA due to continued synthesis of the protective prostanoid, PGE2, by COX-1. However, COX-2 inhibitors have not been studied yet in patients with AIA and are currently contraindicated.

NSAID allergic patients can also be desensitized. Small incremental doses of ASA are taken over 2 to 3 days until 400 to 650 mg of ASA is tolerated. ASA should then be given daily, with doses ranging from 80 to 325 mg to maintain desensitization. After each dose of ASA, there is a refractory period of 2 to 5 days, during which ASA and other COX inhibitors can be taken without any risk of an allergic reaction. Nasal inflammatory disease appears to respond best to desensitization treatment. The patient most likely to benefit from desensitization is one with AIA who has just had sinus/polyp surgery. ASA desensitization has been shown to delay recurrence of nasal polyp formation by an average of 6 years⁴⁷. The mechanism behind ASA desensitization is poorly understood but may be associated with downregulation of leukotriene receptors.

Stevenson et al⁴⁷
Between 1988 and 1994, 78 patients with asthma and documented ASA sensitivity per oral challenge (decrease of 20% or more in FEV1 and/or naso-ocular reactions within 3 hours of incremental oral challenges) were enrolled in a prospective study investigating ASA desensitization. 10 patients discontinued ASA desensitization treatment due to gastritis. 3 patients were lost to follow up. After desensitization, all patients began a treatment program of 650 mg of ASA twice daily, which was continued from 1 to 6 years (mean 3.1 years). Data was compared for 65 patients from the year prior to ASA desensitization and one year before re-evaluation (January to March 1995).

BASELINE AFTER ASPIRIN TREATMENT

4 BASELINE AFTER TREATMENT

Patients with a history of drug allergy are often encountered in medicine. One should consider the particular drug in question before deciding how to proceed. Patients with a history of penicillin allergy should be skin tested. More than 80% of patients with a history of penicillin allergy will have negative skin tests. Allergic reactions observed in the re-treatment of history-positive, skin-test negative patients have virtually all been mild and self-limited; no life threatening false-negative reactions have been reported. Up to 67% of patients with positive skin tests have had allergic reactions when given therapeutic doses of penicillin. There are no skin tests available to evaluate for cephalosporins. Positive penicillin skin tests do not predict allergic reactions to cephalosporins. If no other alternatives are available, cephalosporins can be administered cautiously to patients with a history of penicillin allergy.

There is an increased frequency of adverse reactions to sulfa drugs in HIV patients; the reason for which is not known at this time. A history of rash is not necessarily a contraindication to retreatment since less than 20% may have a recurrence on rechallenge. Skin testing cannot be recommended to evaluate sulfa allergy. Desensitization to sulfa drugs has had variable success.

NSAID intolerance prevalence in asthma and nasal polyposis or rhinosinusitis is 30-40%. Aspirin-induced asthma affects 10% of adults with asthma. The only way to test for NSAID intolerance in the US is with an oral provocation test. High dose acetaminophen has been shown to have a cross-reaction prevalence of 34%. After ASA desensitization, cross desensitization to other NSAIDS than inhibit cyclooxygenase also occurs. The patient most likely to benefit from desensitization is one with AIA who has just had sinus/polyp surgery.

1. Adkinson N. Franklin. Drug Allergy. In Allergy: Principles and Practice, St. Louis, 1998, Mosby, 1212 – 1224.
2. Stevenson DD, Simon RA. Sensitivity to aspirin and nonsteroidal inflammatory drugs. In Allergy: Principles and Practice, St. Louis, 1998, Mosby, 1225 – 1234.
3. Barranco P, Lopez-Serrano MC. General and epidemiologic aspects of allergic drug reactions. Clinical and Experimental Allergy 28, supplement 4: 61-62, 1998.
4. Lieberman P. Anaphylaxis and Anaphylactoid Reactions. In Allergy: Principles and Practice, St. Louis, 1998, 1079-1090.
5. Virella Gabriel, ed. Introduction to Medical Immunology, New York, 1993, Marcel Dekker, Inc., 346.
6. Gruchalla RS. Acute drug desensitization. Clinical and Experimental Allergy 28, supplement 4: 63 – 64, 1998.
7. Lin RY. A perspective on penicillin allergy. Archives of Internal Medicine 152: 930 – 937, 1992.
8. Saxon A. Immediate hypersensitivity reactions to beta-lactam antibiotics. Annals of Internal Medicine 107: 204 – 215, 1987.
9. Aihara M, Ikezawa Z. Evaluation of skin test reactions with delayed type rash induced by penicillins and cephalosporins. Journal of Dermatology 14: 440 – 448, 1987.
10. Vega JM, Blanca M, Carmona MJ, et al. Delayed allergic reactions to betalactams. Allergy 46: 154 – 157, 1991.
11. Sogn Dorothy et al. Results of the National Institute of Allergy and Infectious Disease collaborative clinical trial to test the predictive value of skin testing with major and minor penicillin derivatives in hospitalized patients. Archives of Internal Medicine 152: 1025 – 1032, 1992.
12. Gadde J, Spence M, Wheeler B et al. Clinical Experience with penicillin skin testing in a large inner-city STD clinic. JAMA 270: 2456 – 2463, 1993.
13. Stark et al. Acute and chronic desensitization of penicllin-allergic patients using oral penicillin. Journal of Allergy and Clinical Immunology 79 (3): 523 - 532, 1987.
14. Smith CR. Considerations regarding clinical safety and tolerability of antibiotics in serious and nosocomial infections. Clinical Therapeutics 4 (Supplement A): 133 – 145, 1981.
15. Neu HC. The new beta-lactamas-stable cephalosporins. Annals of Internal Medicine 97: 408 – 419, 1982.
16. Moskovitz BL. Clinical adverse effects during ceftriaoxone therapy. American Journal of Medicine 84 – 88, 1984.
17. Sanders CV, Greenberg RN, Marier RL. Cefamandole and cefoxitin. Annals of Internal Medicine 103: 70 – 80, 1985.
18. Norby SR. Side effects of cephalosporins. Drugs 34 (supplement 2): 105 – 120, 1987.
19. Petz LD. Immunologic cross-reactivity between penicillins and cephalosporins: a review. Journal of Infectious Disease 137 (supplement): 74 – 79, 1978.
20. Smith JW, Johnson JE, Cluff LE. Studies on the epidemiology of adverse drug reactions: II. An evaluation of penicillin allergy. New England Journal of Medicine 274: 998 – 1002, 1966.
21. Anne S, Reisman RE. Risk of administering cephalosporin antibiotics to patients with histories of penicillin allergy. Annals of Allergy, Asthma, and Immunology 74: 167 – 170, 1995.
22. Jick H. Adverse reactions to trimethoprim-sulfamethazole in hospitalized patients. Review of Infectious Diseases 4: 426 – 428. 1982.
23. Gordin FM, Simon GL, Wofsky CB, Mills J. Adverse reactions to trimethoprim-sulfamethazole in patients with acquired immune deficiency syndrome. Annals of Internal Medicine 100: 495- 499, 1984.
24. Carr A, Cooper DA, Penny R. Allergic manifestations of HIV infection. Journal of Clinical Immunology 11: 55 –64, 1991.
25. Hughes WT, Lafon SW, Scott JD, Masur H. Adverse events associated with trimethoprim-sulfamethoxazole and atovaquone during the treatment of AIDS-related Pneumocystis carinii pneumonia. Journal of Infectious Disease 171: 1295 –1301, 1995.
26. Lehmann DF et al. The association of opportunistic infections with the occurrence of trimethoprim/sulfamethoxazole hypersensitivity in patients infected with HIV. Journal of Clinical Pharmacology 39: 533 – 537, 1999.
27. Hennessy S et al. Predicting cutaneous hypersensitivity reactions to cotrimazole in HIV-infected individuals receiving primary Pneumocystis carinii prophylaxis. Journal of General Internal Medicine 10: 380 – 386, 1995.
28. Roudier, C et al. Adverse cutaneous reactions to trimethoprim-sulfamethoxazole in patients with AIDS and PCP. Archives of Dermatology 130: 1383 –1386, 1994.
29. Shafer RW et al. Successful prophylaxis of Pneumocystis carinii pneumonia with trimethoprim-sulfamethoxazole in AIDS patients with previous allergic reactions. Journal of Acquired Immune Deficiency Syndromes 2 (4): 389 – 393, 1989.
30. Rieder MJ et al. Prominence of slow acetylator phenotype among patients with sulfonamide hypersensitivity reactions. Clinical Pharmacology and Therapeutics 49 (1): 13 – 17, 1991.
31. Lee BL et al. Altered patterns of drug metabolism in patients with acquired immunodeficiency syndrome. Clinical Pharmacology and Therapeutics 53 (5): 529 – 535, 1993.
32. Buhl R et al. Systemic glutathione deficiency in symptom-free HIV-seropositive individuals. Lancet 2: 1294 – 1298, 1989.
33. Aukrust P, et al. Increased levels of oxidized glutathione in CD 4 + lymphocytes associated with disturbed intracellular redox balance in human immunodeficiency virus type Ic infections. Blood 86: 258 – 267, 1995.
34. Simonart T, et al. Skin testing in AIDS patients with cutaneous trimethoprim-sulfamethoxazole hypersensitivity. Clinical and Experimental Dermatology 23: 47, 1998.
35. Gruchalla RS, Sullivan TJ. Detection of human IgE to sulfamethoxazole by skin testing with sulfamethoxazoly poly-L-tyrosine. Journal of Allergy and Clinical Immunology 88: 784 –792, 1991.
36. Tenant-Flowers et al. Sulfa desensitization in patients with AIDS and cerebral toxoplasmosis. AIDS 5: 311- 315, 1991.
37. White MW et al. Desensitization to trimethoprim-sulfamethoxazole in patients with AIDS and PCP pneumonia. Annals of Allergy 62: 177 – 179, 1989.
38. Absar N et al. Desensitization to trimethoprim-sulfamethoxazole in HIV-infected patients. Journal of Allergy and Clinical Immunology 93: 1001 – 1005, 1994.
39. Kalanadhabhatta V et al. Successful oral desensitization to trimethoprim-sulfamethoxazole in AIDS. Annals of Allergy, Asthma, and Immunology 77: 394 – 400, 1996.
40. Widal MF et al. Anaphylaxie et idisyncrasie. Presse Med 30: 189 - 192, 1922.
41. Szczeklik A, Sanak M. Genetics of aspirin-induced asthma. American Journal of Respiratory and Critical Care Medicine, In Press.
42. Szczeklik A et al. Relationship of inhibition of prostaglandin biosynthesis by analgesics to asthma attacks in aspirin-sensitive patients. British Medical Journal 1: 67 - 69, 1975.
43. Stevenson DD, Lewis RA. Proposed mechanism of aspirin sensitivity reactions. Journal of Allergy and Clinical Immunology 80: 788 - 790, 1987.
44. Lee TH. Mechanism of aspirin sensitivity. American Review of Respiratory Disease 145: 34 - 36, 1992.
45. Schopf E, Stuhmer A, Rzany B et al. Toxic epidermal necrolysis and Stevens-Johnson syndrome: an epidemiologic study from West Germany. Archives of Dermatology 127, 839 - 842, 1991.
46. Settipane RA et al. Prevalence of cross-sensitivity with acetaminophen in aspirin-sensitive asthmatics. Journal of Allergy and Clinical Immunology 96: 480 - 485, 1995.
47. Stevenson DD et al. Aspirin desensitization/treatment of aspirin-sensitive rhinosinusitic-asthmatic patients: long term outcomes. Journal of Allergy and Clinical Immunology 98: 751 -758, 1996.
48. Sullivan TJ et al. Desensitization of patients allergic to penicillin using orally administered betalactam antibiotics. Journal of Allergy and Clinical Immunology 69 (3): 275 - 282, 1982.