Aspirin-exacerbated respiratory disease: NSAID challenge and desensitization

INTRODUCTION — Patients with aspirin-exacerbated respiratory disease (AERD) present with one or more of the following disorders:

●Asthma

●Chronic rhinosinusitis with nasal polyposis

●Reactions to aspirin (acetylsalicylic acid [ASA]) and other cyclooxygenase 1 (COX-1) inhibiting nonsteroidal antiinflammatory drugs (NSAIDs), in which symptoms begin 30 minutes to three hours after ingestion and characteristically involve the upper and lower airways (eg, nasal congestion and bronchospasm)

AERD is also called NSAID-exacerbated respiratory disease (NERD). NSAID challenge protocols involve cautiously administering gradually increasing doses of an NSAID, usually aspirin, in an appropriately monitored medical setting. These procedures are used to diagnose NSAID pseudoallergy and also to desensitize AERD patients to NSAIDs when indicated. The safety and technical aspects of these protocols, as well as the efficacy of aspirin therapy for AERD, are discussed here. The clinical manifestations and management of AERD and a general discussion of allergic and pseudoallergic reactions to NSAIDs are reviewed separately. (See "Aspirin-exacerbated respiratory disease" and "NSAIDs (including aspirin): Allergic and pseudoallergic reactions".)

OVERVIEW OF NSAID CHALLENGE — NSAID challenges are structured protocols in which the patient is given incrementally increasing doses of an NSAID under medical supervision.

Referral — NSAID challenges are usually performed by allergy specialists with expertise in drug allergy and the training, medications, equipment, and support staff needed to manage any resultant symptoms whenever possible. In addition, there are several types of NSAID reactions, and determining which type is likely present can require some expertise. In some centers (especially in Europe), pulmonary and otolaryngology specialists also perform NSAID challenges. (See "NSAIDs (including aspirin): Allergic and pseudoallergic reactions".)

Safety — Challenges in patients with suspected type 1 pseudoallergy are provocative challenges, meaning that they are likely to induce symptoms, albeit in a controlled and limited manner. These protocols differ from most other types of drug challenges (particularly those for immunoglobulin E-mediated allergic reactions, such as anaphylaxis), which are performed only on patients who are not believed to be allergic in order to confirm lack of allergy.

Because NSAID challenges in AERD are likely to provoke symptoms, the following precautions are necessary:

●Challenges should be performed in a medically monitored setting. In most cases, these protocols may be safely performed in specially equipped outpatient allergy clinics, although the decision about where to perform challenges requires clinical judgment, and each case must be considered individually.

●The patient's asthma should be well controlled because NSAID challenges may induce wheezing and bronchospasm.

●Many published drug challenge protocols call for obtaining intravenous access prior to challenge as a precaution in case intravenous medications or fluids are required. However, while this is the common practice, neither the author nor editors of this topic have needed to administer intravenous fluids or medications to a patient with AERD undergoing aspirin challenge.

●Informed consent should be obtained.

Indications for challenge — NSAID challenge protocols are performed either to diagnose NSAID pseudoallergy or to desensitize the patient to NSAIDs in order to administer NSAIDs on a regular basis.

●Diagnosis of NSAID pseudoallergy – There are no in vitro or skin testing methods available to diagnose NSAID pseudoallergy, and a provocative challenge procedure is the only means of obtaining a definitive diagnosis. In patients with AERD, an NSAID challenge that results in characteristic nasal, ocular, and/or respiratory symptoms is diagnostic of type 1 pseudoallergy. However, outside of research settings, challenges are uncommonly performed solely for the purpose of diagnosis as they are not without risk to the patient.

●Desensitization to NSAIDs – Desensitization is performed by extending the challenge procedure beyond the initial induction of symptoms. Desensitization is commonly (although not exclusively) performed with oral aspirin. Successful aspirin desensitization induces a state of temporary tolerance to NSAIDs, which can be maintained indefinitely as long as the patient continues to take aspirin or another NSAID on a daily basis. This can be useful in several clinical scenarios. (See 'Desensitization' below.)

Contraindications to challenge — It is essential that any concurrent cardiopulmonary conditions are under optimal control before a patient is challenged.

In the author's clinic, we perform oral challenges on patients with asthma only if the prebronchodilator forced expiratory volume in one second (FEV₁) is ≥70 percent of the patient's best and ≥1.5 L. The criteria for inhalation challenges are less stringent as inhalational challenges induce milder reactions.

TYPES OF CHALLENGES — Challenges to NSAIDs may be performed orally or via inhalation. Both types of challenges may be used in diagnosis and desensitization.

Oral — Aspirin (acetylsalicylic acid [ASA]) is the drug most often utilized in oral challenges. There is more published experience with aspirin than any other NSAID, and aspirin has the additional advantage of not having been implicated in any confirmed cases of immunoglobulin E-mediated anaphylaxis.

Inhalational — Both bronchial and nasal inhalational protocols have been developed. Inhalational challenges result in milder symptoms than those provoked by oral challenge, presumably because there is less systemic absorption of the drug (although this has not been directly demonstrated).

●In other countries, aspirin-lysine (ASA-lysine) solutions are available, which may be administered by bronchial or nasal inhalation for the diagnosis of AERD (although not for desensitization to aspirin) [1,2].

●In the United States, parenteral ketorolac can be used because ASA-lysine solutions are not available [3,4]. This is diluted in saline and administered intranasally (table 1).

CHALLENGE PROTOCOLS AND PROCEDURES

Choice of protocol — Provocative aspirin challenges in patients with AERD can be performed using oral aspirin or using solutions of aspirin-lysine (ASA-lysine; not available in the United States) that are inhaled into the lungs or solutions of ketorolac that are sprayed into the nose. Both oral and inhalational challenges with aspirin may be used to diagnosis AERD, but only oral challenges have been used as a means of desensitizing both the upper and lower airways for patients who require long-term oral aspirin therapy.

All aspirin challenges for AERD involve inducing symptoms. Because the symptoms most often begin 40 to 60 minutes after oral ingestion, the time interval between doses must take this delay into account. An oral protocol using three-hour intervals between doses is the best-studied approach, but protocols using 90- and 60-minute intervals, which take less time to complete, have also been published [5-9].

The choice of protocol should take into account the patient's history of past reactions to NSAIDs, if this information is available. Although most patients react 40 to 60 minutes after aspirin ingestion, some will react later (up to two hours), which is why the interval of three hours was chosen when these protocols were first developed. Specialty centers and large allergy groups generally choose one protocol that is consistently used at that institution and is familiar to the staff. For clinicians who perform challenges and desensitizations infrequently, we suggest use of the oral protocol with three-hour intervals because it has an excellent safety record and requires no special solutions or equipment. The main disadvantage of this protocol is the time required to achieve desensitization, which is typically two days (assuming the initial day of placebos, which is most relevant to research settings, is not needed).

If a faster protocol is desired, we would suggest using the one with 90-minute intervals as most patients will react within 90 minutes of ingestion. However, those patients who react after 90 minutes might have a more severe reaction as they will have already received another dose when they begin reacting to the provoking dose.

Premedication for oral protocols — Patients with suspected or known AERD are usually pretreated with leukotriene-modifying drugs (LTMDs) prior to undergoing an oral challenge or desensitization to NSAIDs if the patient is not already taking one of these medications. If the patient is already taking an LTMD, then it is continued through the procedure.

We suggest that one of the following medications be administered prior to the procedure:

●Either montelukast (10 mg once daily in adults) or zafirlukast (20 mg twice daily in adults) – The author's practice is to administer montelukast, either 10 mg daily beginning three days before or 10 mg twice a day for one day before and then 10 mg daily on each day of the challenge/desensitization.

●Extended-release zileuton (1200 mg twice daily in adults) – We administer this for at least three days prior to challenge.

LTMDs significantly impact the results of oral NSAID challenge by blunting the bronchospastic component in particular [10-12]. Before LTMDs became available, NSAID challenges were routinely performed in intensive care units because of the severe asthmatic and sometimes systemic symptoms (eg, hypotension) that could develop in highly sensitive patients, but the recognition that LTMDs attenuated the severity of the reactions allowed aspirin desensitization to be performed safely in properly equipped outpatient settings [5].

Large observational studies comparing patients challenged with LTMD premedication with historical controls challenged without LTMD premedication show that patients undergoing oral aspirin (acetylsalicylic acid [ASA]) challenge while taking LTMDs demonstrate fewer and less severe episodes of bronchospasm, while nasal and ocular symptoms remain [10,11]. One study found that premedication with montelukast resulted in a 90 percent reduction in asthmatic reactions to challenge [13]. The symptoms are essentially shifted from the lower airway to the upper airway. Montelukast and zafirlukast do not mask the reactions in oral challenges altogether, and symptoms are still detectable [10]. There is less experience with zileuton, and this agent may suppress symptoms to a greater degree. The effects of montelukast (10 mg and 40 mg) on nasal challenge were studied in a small, randomized trial, and both doses reduced nasal symptoms by approximately one-third compared with placebo, with no difference between the two doses [12].

Topical corticosteroids taken regularly by the patient (including inhaled, intranasal, and/or in combination with inhaled long-acting beta-agonists [LABAs]) should be continued throughout the procedure. In fact, it is our preference that patients with asthma are taking a combination inhaled corticosteroid-LABA medication at the time of the challenge. Systemic glucocorticoids, which have minimal effect on the outcome of challenge [10], are sometimes administered for one week or more before the challenge to optimize asthma control in patients with moderate-to-severe asthma.

Omalizumab did not appear to prevent or significantly alter symptoms during aspirin desensitization in a study that included nine patients receiving this agent [14].

Premedication with antihistamines or cromolyn is not given. H1 antihistamines were shown to blunt reactions to aspirin in one study [15], and cromolyn was shown to delay the onset of symptoms in another [16].

Oral aspirin challenge

Three-hour intervals — At the author's center, over 1000 patients with AERD have been successfully challenged and desensitized to oral aspirin [17]. It is rare that a patient cannot complete the protocol because of persistent respiratory or gastrointestinal symptoms. The approach reviewed below is based upon that extensive clinical experience.

Use of placebo doses — The author does not usually include placebos when a challenge is performed solely for the purposes of patient management. Particularly for patients suspected of having AERD, the index of suspicion for the condition is usually highly based on history alone, and placebos are rarely needed.

In contrast, placebo doses are essential in research studies in which challenge is performed for the purposes of confirming the diagnosis. Many early diagnostic protocols include an initial day in which only placebos are administered to control for nonspecific changes in pulmonary function that can occur as the day progresses.

There are some situations in which the use of placebo doses is clinically useful. For example, placebo doses are helpful in the evaluation of the patient who is suspected of not having NSAID sensitivity at all, in whom challenge is indicated to prove that the patient tolerates NSAIDs. We sometimes give such patients an initial dose of placebos and also end the challenge with one or more placebo doses (so that the patient receives placebo when expecting the largest dose of the day). Informed consent should be obtained for the general inclusion of placebos, although the patient is obviously not told when in the protocol the placebo will be given.

Procedure — A sample oral challenge protocol used at the author's institution is provided (table 2). This protocol uses commercially available aspirin tablets, with lower doses obtained by cutting an 81 mg tablet with a pill cutter [6]. Doses are given at three-hour intervals, up to a maximum dose of 325 mg. These protocols are carried out over one to two consecutive days. It is rare that a patient cannot be desensitized using this approach.

A starting dose of 40.5 mg is suggested in most cases. This is based on a study in which 420 patients with AERD were challenged with aspirin, and 75 percent reacted after the administration of either the 45 or 60 mg dose [18].

Provocation of symptoms — The criteria used to determine a positive oral challenge are shown in the table (table 3). Most patients with AERD and pseudoallergic reactions to NSAIDs develop symptoms in response to a dose between 60 and 100 mg. Premedication with an LTMD dramatically reduces the pulmonary symptoms seen during challenge, although nasal and ocular symptoms remain. (See 'Premedication for oral protocols' above.)

Interpretation — Patients with positive respiratory reactions to oral challenge are confirmed to have type 1 pseudoallergy. Patients with negative aspirin challenges either do not have type 1 pseudoallergy or symptoms have been sufficiently blocked by LTMDs as to not be apparent [19]. If the challenge was performed to diagnose type 1 pseudoallergy, then the procedure should be repeated without LTMDs. If the procedure was done to both diagnose and desensitize the patient to aspirin, a practical approach is to continue the patient on aspirin (650 mg twice daily) for up to three months. If the patient's symptoms have not substantially improved, then either the patient does not have AERD or the patient is a nonresponder to aspirin treatment and aspirin can be discontinued.

Patients who develop cutaneous as well as respiratory symptoms may have type 4 pseudoallergy. (See "NSAIDs (including aspirin): Allergic and pseudoallergic reactions".)

Treatment of symptoms during challenge — Pretreatment with LTMDs dramatically reduces the severity of symptoms. However, if symptoms do develop, they may last for several hours. Suggestions for management are based on clinical experience [6]:

●Bronchospasm can be treated with nebulized albuterol, as needed, to relieve symptoms.

●Nasal obstruction can be treated with topical nasal decongestant sprays (eg, oxymetazoline). Untreated nasal obstruction may lead to entrapment of secretions and severe headache. Oral antihistamines may also be helpful for nasal symptoms.

●Ocular symptoms may be treated with an oral antihistamine.

●Laryngospasm may be managed with nebulized racemic epinephrine and/or intramuscular epinephrine.

●Gastrointestinal symptoms can be treated with antacids or histamine-2 receptor antagonists. If symptoms persist, we administer a proton pump inhibitor and sometimes misoprostol.

Repeating the provoking dose — Once the patient has developed symptoms and these have been treated, the next step in the protocol is to repeat the provoking dose. For example, if the patient developed symptoms after the 81 mg dose, then 81 mg is given again. Many patients will not develop symptoms to this repeat dose, but some will. When the 81 mg dose is tolerated, the protocol is then continued until the patient reaches the 325 mg dose. It is rare (but possible) for patients to react to two different doses at or below 325 mg. (See 'Desensitization' below.)

Faster oral protocols — Protocols using shorter intervals between doses have been developed.

Ninety-minute intervals — Protocols using 90-minute intervals between doses have also been published. One such protocol was reported to be successful in desensitizing 43 of 44 subjects [6,8]. The authors describe the protocol as a "one-day" aspirin challenge, although the average time to complete the desensitization was 9.5 hours. That may be longer than many clinics and offices can remain open in a single day. Three hours of total observation was required after the onset of symptoms. Patients were only included if their forced expiratory volume in one second (FEV₁) was 70 percent of predicted or greater, whereas other aspirin challenge protocols include patients with FEV₁ of 70 percent or greater of the patient's prior best. Thus, patients with severe asthma may have been excluded, and patients with AERD not uncommonly have severe asthma. Also, the authors defined desensitization somewhat differently than previous studies. The patient was considered desensitized once they tolerated the dose after the provoking dose and had also consumed a total dose of at least 325 mg. Those patients were then sent home to take 325 mg twice a day. Thus, 14 patients took the first 325 mg dose at home without monitoring. After a week, they advanced to full-dose aspirin treatment (650 mg twice a day). Both doses were tolerated by all subjects without incident. This study presents a promising approach that requires further study, although, at present, we still recommend that oral challenges be performed with three-hour intervals between doses because of the extensive safety record of this protocol.

Shorter intervals — Studies have also examined 60-minute intervals, although the author believes these protocols require additional study in larger numbers of patients [7,9]. One group compared protocols using 60- and 90-minute time intervals performed at a single center and concluded that safety was comparable, and the 60-minute interval could reduce the total challenge time by three to four hours [9]. However, this study included only 54 patients: 16 undergoing the 90-minute protocol and 38 receiving the 60-minute protocol. Nineteen percent of the 90-minute subjects and 13.3 percent of the 60-minute subjects could not complete the challenges (15 percent of the total subjects studied) due to gastrointestinal symptoms or persistent bronchospasm, which is a much higher failure rate than that of the three-hour oral protocols. In addition, AERD was not confirmed in this patient population prior to challenge, and the author's experience suggests that many AERD patients react after 60 minutes, such that higher rates of more severe reactions would be observed if this protocol were applied to larger numbers of patients. Thus, this protocol may be reasonable if the patient's history clearly suggests that their reaction occurs in less than 60 minutes, but, for patients who cannot provide this detail, we favor longer time periods between doses.

Inhalational challenge and desensitization — In countries where ASA-lysine is available, the inhalational and nasal routes are preferred at most centers. In the United States, ASA-lysine is not available, but parenteral ketorolac may be used by inhalation instead [4]. This approach is preferred at the author's institution because it is both safe and relatively quick.

The benefits of nasal/bronchial over oral provocation techniques include:

●Nasal/bronchial challenges can be performed more quickly, increasing patient and clinician convenience.

●The symptoms induced by nasal/bronchial challenges are more limited in nature and more quickly reversible relative to those induced by oral challenges.

●Nasal administration is safe in patients who have a FEV₁ <1.5 L and therefore are not good candidates for oral or bronchial provocative challenge. However, nasal administration cannot be performed in patients with severe or complete nasal obstruction due to severe polyposis.

The dose, means of aspirin instillation, duration of observation period, and criteria for positivity are different with these protocols [1,20,21]. Note that patients undergoing bronchial provocation challenges should not be pretreated with leukotriene modifiers, as this could blunt symptoms enough to render the procedure nondiagnostic [22].

Ketorolac nasal provocation — AERD can also be diagnosed using challenge with an intranasal solution of ketorolac instead of oral aspirin. This approach is the inhalation protocol currently favored at the author's institution. It was studied in 100 patients with pseudoallergic reactions that were previously confirmed by oral aspirin challenge [4]. Patients received incrementally increasing doses of intranasal ketorolac at 30-minute intervals and were then switched to oral aspirin at three-hour intervals to complete the challenge and reach the full dose of 325 mg (table 1).

In a series of 167 patients desensitized in this manner who reacted during the procedure (thus confirming aspirin intolerance), the process was completed in an average of 1.67 days [14]. Patients with gastrointestinal symptoms required more time (average 2.29 days) to achieve desensitization and reach a dose of 325 mg of aspirin. Baseline scores on the Sino-Nasal Outcome Test (SNOT) were higher in the subgroup of severe reactors compared with nonsevere reactors, suggesting that this might be useful in predicting which patients are more likely to react.

Definition of a positive nasal challenge — A positive nasal challenge was defined as nasal or ocular symptoms consistent with rhinitis or conjunctivitis, a decrease in nasal flow rate ≥20 percent, or any lower airways signs and symptoms, including bronchospasm with a decrease in FEV₁ ≥15 percent or laryngospasm [4]. In the study of 100 patients mentioned above, the average provoking dose of ketorolac was a cumulative dose of 12.1 mg, and patients most often reacted after the third or fourth dose of the intranasal portion of the procedure [4].

Aspirin-lysine bronchial provocation — Bronchial provocation with ASA-lysine (not licensed in the United States) involves inhaling increasing doses of ASA-lysine at 30-minute intervals. As with other challenges, the patient's asthma should be controlled before beginning the challenge (ie, FEV₁ should be ≥70 percent of the patient's best value when feeling well and ≥1.5 L).

Definition of a positive inhalational challenge — A positive bronchospastic response is defined as a drop in FEV₁ ≥20 percent [23-25]. Bronchospasm is usually the only type of response elicited, occurs fairly quickly (approximately 20 minutes), and is easily reversible with inhaled beta-agonists, such as albuterol [23-25].

Comparison of oral and inhalational challenge — Both oral and inhalational challenges with aspirin may be used to diagnosis AERD, but only oral challenges have been used as a means of desensitizing patients who require long-term oral aspirin therapy.

For the purpose of diagnosis, oral challenges are slightly more sensitive, although they induce more severe bronchospasm. These pulmonary symptoms may be reduced or eliminated by pretreatment with LTMDs. Inhalational challenges can fail to detect patients with very mild symptoms.

There have been a small number of well-designed studies comparing oral aspirin challenge with bronchial ASA-lysine provocation [24,25]:

●One study performed both oral aspirin and bronchial ASA-lysine provocation in 22 patients suspected of having AERD [24]. Both protocols detected bronchospastic reactions in the same subset of patients, although bronchial inhalation produced limited and shorter bronchospastic reactions that were more easily treated. However, 2 of the 22 patients who underwent oral aspirin challenge had solely naso-ocular symptoms, and neither of these individuals reacted to the bronchial ASA-lysine challenge, which is not unexpected since the bronchial challenge provides no direct contact to the naso-ocular surfaces.

●Another study evaluated 35 patients who were already diagnosed with AERD (by clinical history and past positive oral provocation test by the FEV₁ ≥20 percent criterion) and rechallenged them with either oral or bronchial provocation [25]. This study utilized a higher cumulative dose of inhaled ASA-lysine than the previous study and a modified oral aspirin challenge protocol. They demonstrated a higher sensitivity (89 percent versus 77 percent) and similar specificity (93 percent in both groups) in the oral aspirin challenge versus the bronchial ASA-lysine provocation. Three patients had a positive oral aspirin challenge (defined by a decrease in FEV₁ ≥20 percent and/or extrabronchial symptoms) but a negative bronchial provocation challenge. One of these individuals had a sudden 40 percent drop in FEV₁, and the other two had only extrabronchial symptoms during the oral aspirin challenge.

●A third study evaluated 100 patients known to have AERD (by clinical history and past positive oral aspirin challenge) and rechallenged them with intranasal ketorolac (at 30-minute intervals) followed by oral aspirin with a traditional oral aspirin protocol [4]. Patients undergoing nasal ketorolac challenge experienced a smaller decrease in FEV₁ (9 versus 13 percent drop) and a reduced incidence of extrapulmonary symptoms (23 versus 45 percent), compared with those undergoing standard oral aspirin challenge. In addition, nasal challenge patients achieved desensitization to NSAIDs more rapidly (1.9 versus 2.6 days).

DESENSITIZATION — NSAID desensitization is a process of inducing tolerance to NSAIDs in a patient with pseudoallergic reactions. The drug must be taken regularly following desensitization since tolerance is lost if exposure is interrupted.

Indications — In patients with AERD, the indications for aspirin desensitization and daily aspirin therapy include:

●Chronic rhinosinusitis with nasal polyposis that is refractory to topical corticosteroids and other therapies, since chronic aspirin therapy can slow the rate of polyp formation, reduce symptoms, and improve quality of life [26]. The impact of aspirin (acetylsalicylic acid [ASA]) therapy on asthma is less clear. Note that patients with asthma and/or chronic rhinosinusitis but without aspirin intolerance do not appear to benefit from aspirin therapy [27]. (See "Aspirin-exacerbated respiratory disease", section on 'Efficacy'.)

●The need for aspirin antiplatelet therapy in patients with cardiovascular risk factors or confirmed cardiovascular disease.

●The need for NSAIDs to treat chronic inflammatory conditions, such as arthritis.

The utility and limitations of NSAID desensitization are reviewed here. A more general discussion of the management of the different types of NSAID reactions, including treatment with highly selective cyclooxygenase (COX) 2 inhibitors or with very weak COX-1 inhibitors, is found separately. (See "NSAIDs (including aspirin): Allergic and pseudoallergic reactions".)

Protocol — Desensitization of patients with pseudoallergic NSAID reactions involves extending the challenge procedure used for diagnosis. Once the patient has developed symptoms and these have been treated, the challenge is resumed by repeating the same dose that elicited the reaction. If there is another reaction, the symptoms are again treated, and the same dose is given again until there is no further reaction. Once the patient no longer reacts to that dose, the protocol is continued until the patient has ingested a full 325 mg dose without developing symptoms. At this point, the patient is desensitized.

AERD patients who have been successfully desensitized to 325 mg of aspirin can subsequently increase the dose to 650 mg twice daily at home. We believe this approach to be safe because, in over 600 consecutive desensitizations to aspirin, we have not seen a reaction to 650 mg of aspirin once a patient is desensitized to 325 mg [28].

The issue of whether to repeat the provoking dose in patients with mild reactions rather than simply continuing the protocol with the next higher dose has not been investigated. Thus, we recommend always repeating the provoking dose.

Mechanism and outcomes — The mechanism through which aspirin desensitization alters a patient's response to NSAIDs is not completely understood. One theory proposes that desensitization and subsequent daily aspirin treatment reduces interleukin (IL) 4-induced expression of leukotrienes by inhibiting the transcription factor, signal transducer and activator of transcription (STAT) 6 [29,30]. Uncommon exceptions include patients with type 4 or blended reactions with a history of urticaria occurring as part of a respiratory reaction to NSAIDs (particularly more than one NSAID). Patients with type 4 reactions are less likely to experience successful desensitization. A minority of patients will not be able to complete the aspirin desensitization procedure, because of refractory nausea, abdominal pain, diarrhea, or the appearance of an erythematous, pruritic, macular dermatitis (picture 1 and picture 2) [31,32]. Rarely, a patient may develop lower respiratory symptoms that do not resolve between doses, and we may abort the protocol if the patient's symptoms cannot be controlled in the course of a few hours.

Issues following desensitization — Nearly all patients with AERD can be desensitized to NSAIDs. In response to chronic aspirin therapy, patients with AERD can experience improvements in both asthma and chronic rhinosinusitis.

Cross-desensitization to other NSAIDs — Patients with AERD who successfully complete an aspirin desensitization will also tolerate other NSAIDs and can use them as needed, as long as they continue to take at least 325 mg of aspirin or the equivalent of another COX-1-inhibiting NSAID daily to maintain the desensitized state [33]. As an example, a patient with AERD who has been desensitized to 325 mg of aspirin twice daily should be able to take 600 mg of ibuprofen for a headache if needed without developing symptoms. A desensitized patient can also change from one NSAID to another without losing tolerance. For example, a patient taking 650 mg of aspirin twice daily could discontinue this and immediately begin taking naproxen sodium for an inflammatory condition instead. This applies to all patients with pseudoallergic reactions who have been desensitized.

Low-dose aspirin treatment — Patients with AERD who wish to take 81 mg of aspirin daily for cardioprotection will remain tolerant of this low dose of aspirin, although they will not be able to tolerate larger doses of aspirin or be cross-desensitized to other NSAIDs.

Missed doses — When aspirin therapy is interrupted in a patient with AERD who was previously desensitized, there is a "refractory period" of at least two to three days before the patient becomes reactive again. During this period, aspirin or the desired NSAID can be reintroduced without a reaction. Also, the return to sensitivity is gradual, so even if three to five days have elapsed, one can reintroduce the 325 mg dose (or 81 mg if desired as treatment) in the office/clinic without the patient experiencing the severity of the reaction to the provoking dose during the initial challenge. Beyond five days, the full challenge and desensitization process must be repeated before the patient can resume NSAID therapy.

Breakthrough symptoms — Occasionally, a patient reports recurrent symptoms and appears to have lost tolerance. It is our clinical experience that symptoms in these situations are almost always due to some other asthma trigger, such as an upper respiratory tract infection, gastroesophageal reflux disease, or irritant exposure. The patient may inappropriately attribute this to the aspirin and stop taking it. Patients should be counseled that they will remain tolerant to aspirin as long as they are compliant with the daily dose. They may still experience flares in their asthma and rhinitis symptoms from time to time, but, provided they have been compliant, they should not attribute these changes to the aspirin or discontinue it without speaking to their clinician first.

Prevention of gastrointestinal toxicity from NSAIDs — Clinicians prescribing chronic aspirin therapy should discuss the risk of gastrointestinal toxicity with patients and assess the patients' baseline risk of this complication before initiating therapy. The author does not empirically start patients on proton pump inhibitors or other preventative medications unless they are at increased risk. Primary prevention of gastrointestinal toxicity is reviewed separately. (See "NSAIDs (including aspirin): Primary prevention of gastroduodenal toxicity".)

Reasons for discontinuation — The development of gastric irritation from daily aspirin is reported by up to one-half of patients, although not all of these patients have symptoms severe enough to cause them to stop therapy [26,34]. Still, gastric irritation is the most common reason for discontinuing therapy. In the author's experience, this appears to occur at similar rates in patients taking 325 mg twice daily and 650 mg twice daily. However, published data relating NSAID dose to gastritis and gastrointestinal bleeding are inconsistent, and other risk factors may be more important than dose. One retrospective series of approximately 100 patients with AERD found an incidence of gastrointestinal bleeding of <1 percent, with no relation to dose [22,35].

A small number of patients presenting with either acute or chronic epigastric pain in association with aspirin therapy, with or without nausea, have been found to have pancreatitis, as evidenced by elevations in serum lipase and amylase [36,37]. However, it is not entirely clear that aspirin was the cause in these cases.

Other reasons for discontinuation include pregnancy (in the third trimester) and elective surgery [26]. The use of aspirin and NSAIDs during pregnancy is reviewed separately. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'NSAIDs'.)

EFFICACY — The efficacy of daily aspirin therapy for the treatment of AERD is discussed separately. (See "Aspirin-exacerbated respiratory disease", section on 'Efficacy'.)

OPTIMAL ASPIRIN DOSE FOR AERD — The available evidence suggests that both 325 mg twice daily and 650 mg twice daily are effective in reducing nasal/sinus and asthma symptoms in patients with AERD. At the higher dose, benefit is seen in 65 to 87 percent of patients treated for 6 to 12 months [26]. An analysis of 137 patients with AERD referred to the Scripps Clinic and treated with aspirin for at least one year found that approximately one-half could successfully decrease the dose from 650 mg twice daily to 325 mg twice daily without a relapse of symptoms [34]. However, the other one-half required the higher dose to maintain improvement long term. Only occasional patients derived lasting benefit from 325 mg once daily.

The lowest effective dose is not precisely defined [38-40]:

●In the small, randomized trial already discussed, most subjects began to develop recurrence of their nasal congestion when the aspirin dose was reduced below 325 mg twice daily [38].

●A subsequent unblinded study sought to determine if even lower doses could be effective by comparing 100 mg and 300 mg of aspirin daily [39]. After one year, there was no improvement in asthma, polyps recurred in the group receiving the lower dose, and all patients were changed to 300 mg of aspirin daily. A subsequent cohort of 39 patients was evaluated, and each completed at least one year of aspirin therapy. Nasal polyposis significantly decreased in all patients. Asthma symptoms were not evaluated. Thus, 300 mg daily appears to be helpful for the management of nasal polyposis, although the benefit on asthma symptoms is not as clear. In contrast, a dose of 100 mg daily is clearly not sufficient.

In summary, both 325 mg twice daily and 650 mg twice daily are effective in reducing nasal/sinus and asthma symptoms in patients with AERD. Because approximately equal percentages of patients tolerate each of these doses but one-half clearly benefit only from the higher dose, it is logical to begin therapy at the higher dose. The effects should be apparent after three to six months of therapy, after which patients who have responded well could have their doses reduced to the lowest dose that maintains symptom control [41]. If symptoms begin to escalate again on the lower dose, then patients should resume 650 mg twice daily.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Drug allergy and hypersensitivity".)

SUMMARY AND RECOMMENDATIONS

●Indications for NSAID challenge and desensitization – Patients with aspirin-exacerbated respiratory disease (AERD) may develop type 1 pseudoallergic reactions to nonsteroidal antiinflammatory drugs (NSAIDs; including aspirin [acetylsalicylic acid; ASA]). NSAID challenges are used to confirm the diagnosis of pseudoallergy. These same challenges can be extended to achieve desensitization to NSAIDs. NSAID desensitization allows patients with AERD and type 1 pseudoallergy to safely take any NSAID, provided it is continued on a daily basis. Daily NSAID therapy may be indicated for several reasons, including treatment of the upper and lower respiratory manifestations of AERD, treatment of rheumatologic disorders, pain management, or antiplatelet therapy. (See 'Indications for challenge' above.)

●Referral and safety – Whenever possible, NSAID challenge and/or desensitization should be performed by an allergy specialist with experience in diagnosing and managing drug adverse reactions. NSAID challenges are expected to provoke symptoms, and clinicians must be prepared to manage these safely. (See 'Referral' above and 'Safety' above.)

●Oral versus inhalational protocols – Both oral and inhalational (either nasal or bronchial) NSAID challenge procedures have been developed for diagnosing pseudoallergy reactions to NSAIDs. However, only oral protocols have been used for desensitization (see 'Choice of protocol' above):

•Oral aspirin challenge/desensitization has been the approach in the United States (table 2), although a hybrid protocol that begins with nasal administration of liquid ketorolac and concludes with oral aspirin has been developed and offers some safety advantages (table 1). This is the author's preferred approach. (See 'Oral aspirin challenge' above.)

•In Europe, a liquid preparation of aspirin-lysine (ASA-lysine) has been available for decades, and most challenges are performed by inhalation. (See 'Inhalational challenge and desensitization' above.)

●Challenges are expected to provoke symptoms – NSAID challenge procedures are expected to provoke symptoms, and the development of characteristic signs and symptoms confirms the diagnosis of a type 1 pseudoallergic reaction (table 3). (See 'Provocation of symptoms' above.)

●Premedication – We recommend that all patients with AERD be premedicated with a leukotriene-modifying drug (LTMD) before oral aspirin challenge/desensitization to reduce the severity of provoked symptoms (Grade 1B). Montelukast is commonly given in this setting. (See 'Premedication for oral protocols' above.)

●Desensitization – Desensitization is accomplished by continuing the diagnostic challenge procedure. We prefer a dosing interval of three hours over faster protocols for safety reasons. Symptoms are provoked and treated, and the provoking dose is then repeated until the patient no longer reacts to it. The patient is considered desensitized when 325 mg of aspirin is tolerated. Aspirin desensitization is nearly always successful in patients with AERD. (See 'Desensitization' above.)

●Maintaining the desensitized state – Once desensitized, patients must take at least 325 mg of aspirin daily to maintain the desensitized state. They are also cross-desensitized to other NSAIDs and can take these medications as well. However, if the dose of aspirin is reduced to 81 mg daily, cross-desensitization is usually lost. (See 'Issues following desensitization' above.)