Aspirin-exacerbated respiratory disease (AERD) is a clinical tetrad of nasal polyps, chronic hypertrophic eosinophilic sinusitis, asthma and sensitivity to any medication that inhibits cyclooxygenase-1 (COX-1) enzymes, namely aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) (Table 1).1,2 Ingestion of aspirin, and most NSAIDs, results in a spectrum of upper and/or lower respiratory reactions, to include rhinitis, conjunctivitis, laryngospasm and bronchospasm.1,2 AERD affects 0.3-0.9% of the general population, but the prevalence rises to 10-20% of asthmatics and up to 30-40% in those asthmatics with nasal polyposis.3-7 The average age of onset is 34 years in a US study and is thought to be acquired between teenage to middle adulthood years with no ethnic predilection and rare familial associations.3-7 AERD is more commonly reported in females (57% vs. 43%).7,8

AERD has been described as "Samter's triad, aspirin induced asthma, aspirin sensitive asthma and aspirin hypersensitivity".1,9 Despite the fact that it was first described in 1922 by Widal et al.10, the pathophysiology of AERD is only partially understood and is not the focus of this review; however two reviews on this topic were recently published by Stevenson.2,11

Clinical features of AERD include onset of nasal congestion with anosmia and progression to chronic pansinusitis and nasal polyps which re-grow rapidly after surgery.2,7 Nocturnal nasal obstruction with sleep deprivation fatigue occurs routinely in these patients. Asthma may precede the upper airway disease or develop later. CT or plain radiographs of the sinuses reveal complete opacification in nearly all AERD patients.2,12 Normal imaging of the sinuses essentially rules out the diagnosis of AERD.

Aspirin challenge is the gold standard for diagnosing AERD.1,2,13 There are four routes of provocation challenges: oral, bronchial inhalation, nasal inhalation and intravenous.13-17 In the United States oral aspirin challenges are performed. The protocol used at the Scripps Clinic is found in Table 2.

History of an asthma attack following ingestion of aspirin or other NSAIDs is suggestive and sometimes diagnostic. However, 16% of patients who believed they had AERD, based upon a historical asthma attack after ingesting aspirin/NSAID, underwent negative oral aspirin challenges and therefore, did not meet criteria for AERD.18 On the other hand, in that same study, of patients who had nasal polyps, chronic sinusitis, asthma and were avoiding aspirin and NSAIDs, only 43% had positive oral aspirin challenges.18 Thus, AERD is both over diagnosed (coincidental history) or under diagnosed (no prior exposure to aspirin or COX-1 inhibitor during the time of the inflammatory respiratory disease).

Once patients are diagnosed, management options are limited to either complete avoidance of COX-1 inhibiting drugs or aspirin desensitization and continuous aspirin therapy.2,13 In order to completely avoid COX-1 inhibitors, it is essential that clinicians are familiar with cross-reacting NSAIDS that inhibit COX-1 (Table 1) and that they educate patients on avoidance of all the various cross-reacting NSAIDS. On the other hand, it is important to note that there are other NSAIDs that either partially inhibit COX-1 or are selective for COX-2, such as celecoxib, which lacks cross-reactivity with COX-1 inhibitors (Table 1). Well designed studies have shown that therapeutic doses of selective COX-2 inhibitors do not cross react with aspirin or other NSAIDs and therefore, can safely be given to patients with AERD.19-21 Due to the rare possibility of other types of reaction to COX-2 inhibitors, we recommend giving the first full dose in the physician's office.

Avoidance of aspirin is not always possible due to need for aspirin in the management cardiovascular diseases. In addition, even with avoidance of aspirin and NSAIDs, AERD patients usually experience progressive airways disease, despite aggressive surgical intervention and topical or systemic anti-inflammatory treatment with corticosteroids and leukotriene modifiers.1,2,7,8 Therefore, the purpose of this review will be to discuss the specific issues related to aspirin desensitization, the clinical benefits, patient selection, aspirin dosing, potential adverse reactions and newer methods that may improve on current protocols for aspirin desensitization.

Multiple studies have shown that desensitization and daily treatment with aspirin can significantly improve overall symptoms and quality of life, decrease formation of nasal polyps and sinus infections, reduce the need for oral corticosteroids and sinus surgery and improve nasal and asthma scores in patient with AERD at both 6 months and after one year of therapy (P<0.0001).22-25 In a study from an otolaryngology clinic comparing outcomes of AERD patients who had surgery with aspirin desensitization versus surgery alone, none of the aspirin desensitized patients needed revision of their surgery, while 80% of the patients who did not undergo desensitization required additional surgery over a 2 year period (P=0.003).26

Significant clinical improvement is seen in as little as 4 weeks after treatment with nasal scores, sense of smell and asthma scores improving significantly (P<0.0001) and for those taking systemic corticosteroids, prednisone doses decrease from an average of 10.7 mg to 5.9 mg daily (P=0.0003).27 From personal experience, there is a marked flattening of nasal membranes immediately after aspirin desensitization and many patients often note improvement in congestion and sense of smell within 24-48 hours after completing aspirin desensitization. Improvement in smell is more complicated, since the mechanism of compression of olfactory nerves by inflamed nasal tissues leads to nerve damage. Even when mucosal tissues return to normal, olfactory nerves may not regenerate.

In addition to the clinical effectiveness of aspirin therapy, Shaker et al.28 performed a healthcare cost analysis to model the costs of aspirin desensitization for therapeutic and prophylactic use and found it to be cost-effective, even when factoring in the up front cost of aspirin desensitization. From an economic perspective, there is a substantial reduction of medical and surgical costs in the years following aspirin desensitization and daily aspirin treatment. Aspirin is also much less expensive for both treatment of primary and secondary prophylaxis in cardiovascular disease compared to other anti-platelet agents.28

Most patients with AERD will benefit clinically from desensitization as described earlier, however this treatment is particularly helpful in patients who have suboptimal control with currently available pharmacotherapy or if they require multiple operations due to re-growth of nasal polyps or intractable sinus disease.29,30 Also, this procedure is indicated in AERD patients who require aspirin or NSAIDs for concomitant cardiovascular diseases, arthritis, or other medical indications (Table 3).29,30

It is important to note that AERD patients often have co-existing and provoking factors for upper and lower airway inflammation, including allergic rhinosinusitis, infections, gastroesophageal reflux disease, exercise-induced asthma, etc.2 Therefore, clinicians must recognize these additional provoking factors and aggressively treat them in order to optimize management. Patients with concomitant allergic rhinitis and asthma should be maintained and stable on antihistamines, topical steroids, leukotriene modifiers and immunotherapy. Sometimes omalizumab needs to be added to control this important co-morbid condition. Consider aggressive treatment of underlying gastroesophageal reflux disease with proton pump inhibitors or even Nissen fundoplication in severe cases recalcitrant to medical management. In addition, if patients have significant polyp or sinus disease, we routinely recommend surgical debulking 2-4 weeks prior to aspirin challenge and desensitization; we find that aspirin treatment works best in preventing new polyp formation. Although some patients experience marked regression of previously formed nasal polyps (Table 3), it is not possible to predict prospectively which patients might enjoy that benefit and to what extent that benefit might occur.

Most AERD patients synthesize excessive leukotrienes and severity of respiratory disease corresponds with the baseline production of leukotrienes.31,32 In addition, AERD patients express more cysteinyl LT1 receptors on their inflammatory cells.33 Aspirin and inhibitors of COX-1 divert arachidonic acid toward the lipoxygenase pathway, hence causing additional overproduction of inflammatory mediators.34 Interestingly, despite the role of leukotrienes in AERD, medications that modify this pathway, such as zileuton, a 5-lipoxygenase inhibitor, or inhibit cysteinyl leukotriene receptor (CysLT₁), such as montelukast or zafirlukast, do not prevent upper airway reactions during aspirin challenges.35 Rather, CysLT₁ inhibitors primarily prevent or attenuate bronchospasm.35 Studies have demonstrated that leukotriene modifier drugs positively impact treatment of AERD and, as we will discuss, also improve safety of aspirin challenges.36-38

We recommend that all patients have baseline spirometry showing FEV1 values to be more than 60% of predicted and at least 1.5 L. If they are not already taking a leukotriene modifier, we start either montelukast, zileuton or both prior to aspirin challenge. Patients are instructed to continue oral and topical corticosteroids and long-acting bronchodilators. We do ask that patients discontinue their antihistamines, decongestants and short-acting inhaled beta-agonists prior to aspirin challenge. These medications may mask a potential response and thus prevent diagnosis (Table 2). As an aspirin challenge and provocative reaction is the gold standard for diagnosis, it is necessary to measure a naso-ocular or bronchial reaction and thus confirm a diagnosis of AERD.

Over 1,400 patients have successfully undergone aspirin challenge and desensitizations at the Scripps Clinic without any deaths or serious complications. In the past, the assumption was that the historical reactions to aspirin/NSAIDs should dictate the location of the aspirin desensitization procedure. The rational was that a bad reaction to aspirin 650 mg (or equivalent doses of NSAID) that required hospital intervention predicted a severe asthma attack during oral aspirin challenges and therefore should be conducted in a hospital ICU. However, Williams et al.39 at our institution compared the severity of prior aspirin/NSAID reactions and the degree of asthma induced during oral aspirin challenges and found that there was no correlation. A small number of severe reactions occurred during oral ASA challenges but they were just as likely to occur in patients with mild historical reactions. Furthermore, none of the "severe" reactions (FEV1 dropping >30% from baseline) required any more treatment than is available in all allergy outpatient offices.39 None required intubation or transfer to an ICU. Our explanation for the above is linked to dose response reactions. Historical reactions were induced by full therapeutic doses of aspirin or NSAIDs, compared to the average provoking dose of aspirin of 60 mg during oral aspirin challenges. Furthermore, the historical reactions may have occurred many years previously, usually without anti-leukotriene coverage. Also, the inflammatory respiratory disease may have changed over the years. In conclusion, historical reaction severity is not predictive of the degree of bronchospasm which might occur during controlled oral aspirin challenges.39

With experienced providers and staff, outpatient oral challenges are as safe as inpatient challenges and much more cost effective. At Scripps Clinic, beginning in 2003, we switched from inpatient (General Clinical Research Center) to outpatient Allergy Division offices and have conducted all oral aspirin challenges in our outpatient with one exception - inpatient desensitization should be strongly considered in patients receiving beta-blockers, recent myocardial infarction, and/or with severe or uncontrolled asthma. The only setting where we have encountered this need for ICU aspirin desensitization is with asthmatic patients admitted for an acute coronary syndrome urgently requiring aspirin therapy for either coronary stent or bypass surgery.

Although aspirin desensitization is beneficial in most patients with AERD, contraindications to aspirin desensitization include pregnancy, unstable asthma, gastric ulcers or bleeding disorders. We screen out patients with the above conditions and require that all patients sign informed consents prior to challenge and desensitization.

Many variations of the oral aspirin challenge have been published.2,30 Hope et al.40 recently published a rational approach to aspirin challenge dosing which reviewed 420 aspirin desensitizations at the Scripps Clinic and used historical data to enhance the efficiency of the desensitization protocol. We found that most reactions occurred at 45 to 100 mg of aspirin and no patients reacted after the 650 mg dose.40 The average time to reaction was 102 minutes after last aspirin dose.40 Based on this rationale, we start with 20 to 40 mg of aspirin and advance every three hours (Table 2), and often complete the protocol in 2.5 days, 3 hours after the 325 mg asprin dose.40 If at any time a reaction occurs in the larynx or the bronchi, reactions are treated before continuation and the provocation dose is then repeated. If the reaction is limited to the nasal membranes, oxymetazoline nasal spray is used and the challenge is then continued with the next dose of aspirin.

In a group of 172 patients, who had undergone aspirin desensitization and treated with daily aspirin 650 mg twice daily for a year, 24 (14%) discontinued ASA treatment due to known side effects (i.e. gastric pain or bleeding, non-gastrointestinal bleeding, such as epistaxis, and urticaria).22 Eleven percent of patients stopped due to poor response and other reasons (i.e. planned surgeries, rash, death from other causes, etc).22 Of the remaining 126 patients who were continued on aspirin for a year, 110 (87%) benefitted from aspirin therapy.22

In the group of 16 "non responders", 12 had concomitant severe allergic respiratory disease to dust mutes and pets in the house. None were on immunotherapy and the study preceded the availability of omalizumab. The study design did not anticipate or account for this variable which may have negated the therapeutic effect of aspirin desensitization treatment.

There is a refractory period of about 48 to 72 hours following aspirin desensitization; this is maintained by continued aspirin therapy. Therefore, if there are brief interruptions to therapy (<48 hours) due to surgery or illness, aspirin may be restarted. However, if there is greater than a 48 hour disruption from treatment, we recommend repeat desensitization. The authors are aware of one patient, from another center, who lost the desensitized state after 24 hours.