Introduction

Aspirin-exacerbated respiratory disease (AERD), or NSAID-exacerbated respiratory disease (NERD), is a heterogeneous inflammatory syndrome characterized by Samter’s clinical triad of chronic rhinosinusitis with nasal polyposis (CRSwNP), moderate to severe asthma, and nonallergic hypersensitivity to all cyclooxygenase-1 (COX-1) inhibiting NSAIDs (pseudoallergy). Throughout this paper, the terms AERD and NERD are used interchangeably. AERD has a prevalence of 0.6-2.5% in the general population and affects 7% of asthmatics, 15% of severe asthmatics, and 10-16% of patients with CRSwNP [1•, 2]. Women are affected twice as often as men are. This condition usually occurs in the fourth decade of life [3], although rare pediatric cases have been reported [4••, 5].

The pathophysiology of AERD involves dysregulated arachidonic acid-derived lipid mediator production, chronic mast cell degranulation, and type 2 (T2) inflammation of the upper and lower respiratory tract. These immunologic derangements result in elevated cysteinyl leukotrienes and proinflammatory prostaglandin D2, as well as decreased anti-inflammatory prostaglandin E2 [6]. Prior drug sensitization is not required for this competitive inhibition of the COX-1 enzyme, which causes NSAID hypersensitivity reactions after the first exposure to all NSAIDs except for selective COX-2 inhibitors.

At baseline, AERD patients have severe refractory sinusitis as well as aggressive nasal polyposis. In addition, AERD patients have more severe upper and lower respiratory tract disease than CRSwNP patients who are ASA tolerant, regardless of the presence of asthma. Flares of sinusitis are potent drivers of asthma exacerbations requiring systemic steroids, immediate care or emergency department visits, and hospitalizations. This inflammatory disease is characterized by heterogeneous endotypes, with most patients in the US and Europe displaying type 2 (T2) inflammation, although inflammatory profiles, including T2-low and a mixture of endotypes, have been described [6].

In AERD, this underlying inflammatory process is further exacerbated by the use of NSAIDs which inhibit COX-1. NSAID-exacerbated symptoms are not IgE-mediated and involve both the upper and lower respiratory tract. Studies have shown that following NSAID use, respiratory (48%-88%) and naso-ocular (8%-20%) symptoms, including nasal congestion, rhinorrhea, ocular erythema, and bronchospasm, may be exacerbated [7, 8]. NSAID hypersensitivity occurs 30-180 minutes after NSAID use and is independent of the route (oral, topical, intravenous); however, there have been reports of delayed symptoms associated with the use of transdermal patches or slow-releasing NSAID tablets [9].

Standard medical treatments include corticosteroids via various delivery methods (such as topical nasal corticosteroids, sinus rinses with corticosteroids, exhalation delivery systems of topical nasal corticosteroids, and corticosteroid eluting stents), as well as leukotriene modifiers, biologics targeting T2 inflammation, and ASA desensitization followed by ASA therapy after desensitization (ATAD). Additionally, surgical options include functional endoscopic sinus surgery (FESS) with either a focused or full-house Draf III frontal sinusotomy approach to debulk nasal polyps and improve penetration of topical agents [1•].

Unfortunately, many AERD patients fail first-line therapies for asthma and CRSwNP and often need repeated sinus surgeries and frequent courses of oral corticosteroids to control symptoms. Interestingly, while COX-1-inhibiting NSAIDs are known to cause exacerbations in AERD, aspirin desensitization followed by maintenance oral high-dose aspirin therapy has been proven to modulate the inflammatory cascade and has become a well-established treatment for most AERD patients. ASA desensitization followed by high-dose ASA is the only disease-modifying therapy shown to improve clinical outcomes for both upper and lower airway disease. In this review, we highlight all randomized controlled trials (RCTs) and meta-analyses published on the clinical efficacy of ASA therapy in AERD, as well as recent works published between January 1, 2018, and January 1, 2024, to explore current outpatient ASA desensitization protocols.

Methodology

A complete search was performed based on key words and MeSH terms related to ‘aspirin desensitization’, ‘Asthma, Aspirin-Induced/drug therapy’, ‘aspirin-exacerbated respiratory disease’, ‘AERD’, ‘NSAID-exacerbated respiratory disease’, ‘NERD’, ‘Asthma, Aspirin-Induced', and ‘Clinical Protocols’. All qualified aspirin desensitization protocols published up to Jan 1, 2024 were included in this review. For clinical trials, a complete search in PubMed was performed based on key words and MeSH terms related to ‘aspirin desensitization’, ‘ASA treatment’, ‘Asthma, Aspirin-Induced/drug therapy’, ‘ASA-sensitive’, ‘aspirin exacerbated respiratory disease’, ‘AERD’, ‘NSAID-exacerbated respiratory disease’, ‘NERD’, ‘Asthma, Aspirin-Induced', ‘double-blind randomized clinical trial’, ‘Randomized Controlled Trial’, ‘Meta-Analysis’, and ‘double-blind crossover study’.

The evidence supporting daily maintenance aspirin therapy after desensitization in treatment of AERD/NERD

In 1899, Bayer patented aspirin and sold it as a pain reliever worldwide. In 1902, the first case of ASA hypersensitivity in an asthma patient was reported in Germany by Hirschberg. However, it was not until 1922 that AERD was defined by M. Fernand Widal, who described a rare constellation of symptoms, including asthma exacerbation, sinus symptoms, and urticaria, after ingestion of oral ASA and antipyrine; this disease was later known as “Widal’s syndrome.” Widal et al. also published the earliest documented ASA desensitization as a case report in 1922, which described how a 37-year-old woman was desensitized to ASA and subsequently treated with daily ASA therapy without adverse effects [1•, 2, 10].

In 1968, Samter and Beers characterized the “aspirin triad” of asthma, ASA sensitivity, and nasal polyps as Samter’s triad. In 1980, Stevenson et al. desensitized two AERD patients to ASA, followed by maintenance ASA 325 milligrams (mg) twice daily; they reported improvements in nasal symptoms and sinusitis with a reduced need for systemic corticosteroids over the next year. This was followed up in 1996 with a 6-year longitudinal study of 65 patients with ATAD, which showed significant improvement in olfaction and sinus infections, as well as reduced FESS and the need for intranasal/systemic corticosteroids. Notably, adverse effects of gastritis and bleeding were also observed in that study [1•].

Multiple published case reports and case series have corroborated that AERD patients with ATAD have improved upper and lower airway symptoms and a reduced need for oral corticosteroid use. However, the early case reports were limited by small sample sizes, while the observational studies were limited by variable design (prospective, retrospective, or longitudinal) and study duration (ranging from 2 to 180 months). Additionally, not all patients had FESS prior to ASA desensitization. Finally, the daily maintenance dose of ASA varied from 100 to 2600 mg. The limitations of these early studies included the lack of a control arm or comparison group, selection bias, difficulty in comparing different patients, and inability to generalize these findings. Despite these limitations, these early observational studies described the benefits of desensitizing AERD patients and continuing ATAD to improve asthma and sinonasal disease. However, these studies also demonstrated that ATAD is not universally beneficial for all AERD patients, with some patients discontinuing ATAD due to adverse effects or a lack of clinical benefit [1•].

Since then, numerous case reports, case series, and nonrandomized controlled trials have shown the clinical efficacy of ATAD [1•]. In particular, the Scripps Clinic in La Jolla, California has the distinction of publishing many notable works on AERD and ATAD. Nevertheless, there is a paucity of RCTs in the area of AERD and on the beneficial effects of ATAD. As of this publication, there are 7 RCTs [10,11,12,13,14,15,16] and 5 meta-analyses evaluating the clinical efficacy of ATAD [17,18,19,20,21]. A summary of the studies is shown in Table 1.

Table 1 Summary of Randomized Controlled Trials and Meta-analyses Evaluating the Clinical Efficacy of ATAD in AERD/NERD Patients
Full size table

Additionally, Lee et al conducted a longitudinal study of AERD patients who had already undergone aspirin desensitization. While this was not an RCT, patients were randomized to daily maintenance ASA 650 mg versus 1300 mg. Patients were not required to have FESS in the 4-6 weeks preceding ATAD. The results showed that daily administration of 650 mg ATAD or 1300 mg ATAD produced a fairly equivalent but significant decrease in nasal, sinus, and asthma symptoms and improved olfaction. Both doses of aspirin showed improvement in nasal symptoms and need for sinus surgery, improvement in asthma symptoms and hospitalizations. Of the 137 patients, 32 (23.3%) had adverse effects or dropped out of the study. The dropout rate was 16% (22 patients total) due to dyspepsia (8), urticaria/angioedema (4), asthma exacerbation (3), ecchymosis (2), incarceration (1), tinnitus (1), pregnancy (1), myalgias (1), or unknown reasons (1). Interestingly, the incidence of adverse effects in patients taking 650 mg twice daily was not significantly different than that in patients taking 325 mg twice daily. In fact, patients discontinued treatment due to adverse effects more frequently in the 325 mg twice daily group than in the 650 mg twice daily group (12.9% vs. 4.5% dyspepsia reports, respectively) [1, 22].

Lastly, a recent retrospective multicenter study by Celik et al evaluating 125 patients over the span of three years found that 300 mg of daily aspirin was similar in efficacy to 600 mg of daily aspirin for ATAD, and thus suggested using 300 mg daily as the dose of ATAD to minimize adverse effects [23].

Current protocols for aspirin desensitization in AERD/NERD patients available in the US

Although the exact mechanism of tolerance induction in ASA desensitization is unclear, there are many published protocols for successful ASA challenge and desensitization in AERD/NERD patients. However, there is no standardized protocol for ASA desensitization. Typically, patients are recommended to undergo sinus debulking surgery 2-6 weeks before desensitization. Leukotriene modifiers should be started at least 3 days before desensitization to reduce the risk of lower airway symptoms. In general, the procedure involves gradually increasing doses of ASA; desensitization is completed when the patient has reached 325 mg. This is followed by a lifelong daily dose of ASA ranging from 650-1300 mg, and a desensitized state is lost if the dose is discontinued for more than 48 hours. These protocols vary in terms of the route of administration, starting dose (usually 20.25 mg or 40.5 mg), dosing interval (60 minutes, 90 minutes, 180 minutes), and total duration of the procedure.

In the US, the most commonly utilized protocols involve oral ASA, intranasal ketorolac, or a combination of both. The protocols for ASA challenge and desensitization are the same, and patients are expected to have upper and lower airway symptoms in both procedures. In AERD, protocols during challenges differ from desensitization when clinical reactivity occurs. In aspirin challenges, reactions result in discontinuation of the procedure and treatment of the patient. In contrast, following clinical reactivity during desensitization, the patient is treated, the provoking ASA dose is repeated, and the procedure is continued to a maximum dose of 325 mg. In this process, reactions become increasingly mild until they disappear. FEV1 is monitored every hour during the procedure [1•].

Definitive diagnosis of AERD requires a formal ASA challenge, but the diagnosis is usually made clinically. In patients who are unsure of having a strong respiratory reaction to COX-1 inhibitors, an ASA challenge can be considered to confirm the diagnosis of AERD unless a challenge is contraindicated. The contraindications included poorly controlled asthma with an FEV1 < 70% of the predicted value, exacerbation of asthma in the previous month, a large nasal polyp burden, eosinophilic esophagitis, gastric or peptic ulcers, bleeding disorders, current anticoagulation, chronic renal failure, and pregnancy. Additionally, aspirin challenge is contraindicated in children younger than 14 years of age due to the possibility of Reye syndrome with concurrent viral infections and salicylate use. The incidence of Reye syndrome is highest at ages 1, 5, and 13. Treatment with ASA should be avoided in children younger than 14 years of age and considered on a case-by-case basis in children older than 14 years of age. Additionally, ASA treatment should be considered on an individualized basis in elderly patients with multiple comorbidities, as most studies included patients up to 70 years of age [1•].

Over the years, ASA desensitization has evolved to emphasize outpatient protocols with the goal of shortening desensitization duration and minimizing patient side effects. Allergists expect symptomatic reactions to occur at doses ranging from 40-160 mg, and patients are recommended to use a maintenance dose of ASA ranging from 650-1300 mg daily to achieve clinical improvement. In general, prior to 2012, most allergists followed the oral ASA desensitization protocol as described by Stevenson et al. [10, 24]. After 2012, the modified intranasal ketorolac and ASA challenge protocol was most frequently followed, as described by Lee et al. [22].

Table 2 describes the most widely used outpatient protocols for ASA desensitization. While protocols that utilize both oral ASA and intranasal ketorolac exist, this review focuses on protocols utilizing only oral ASA—in the U.S., intranasal ketorolac is more costly and not as widely available. Current desensitization protocols require 1-3 days to complete and can be cost- and time-intensive for both patients and medical facilities [29]. As stated above, leukotriene modifiers should be started at least 3 days before desensitization and continued during desensitization. For patients with asthma, inhaled corticosteroids should be continued. The data on whether patients should continue antihistamines are mixed [1•].

Table 2 Oral aspirin desensitization protocols for AERD/NERD patients
Full size table

There are also several considerations in determining the optimal setting for desensitization. Desensitization can be safely performed in the outpatient setting if the FEV1 is greater than or equal to 70% of the predicted value (>1.5 L absolute), if the patient has no other underlying comorbidities, or if the patient is not taking oral beta blockers [1•]. Patients with hemodynamic instability or uncontrolled asthma should undergo desensitization in the inpatient setting. Inpatient desensitization should also be considered for patients who cannot discontinue oral beta blockers.

Factors to consider when selecting the appropriate ASA desensitization protocol include patient availability, patient and physician comfort level with the procedure, and cost. DeGregorio et al. reported that patients with AERD on a stable asthma regimen and an FEV1 > 70% could be desensitized in 1 day with a 90-minute dose escalation protocol [29]. Similarly, a recent QI project involving 15 patients undergoing ASA desensitization demonstrated that there were similar rates of completion and reactions in both the 1-day and 2-day protocols, with lower costs associated with the 1-day protocol [31].

Conclusion

The safety and efficacy of ATAD for improving chronic rhinosinusitis symptoms, long-term asthma control, and quality of life is evident in the studies presented in aggregate above. Additionally, ATAD reduced the need for revision FESS as well as inhaled and systemic corticosteroids. However, ATAD is a long-term therapy; its benefits, potential adverse effects, patient comorbidities, patient preference, and other available treatment options must be considered. While the vast majority of patients find ATAD to be helpful, there are subsets of patients who are not able to tolerate ASA desensitization and ATAD. The most common adverse effects include gastritis, gastrointestinal bleeding, asthma exacerbation, and rash [17, 32, 33]. Additionally, there are case reports of intranasal ketorolac (in modified ASA desensitization protocols) and oral ASA causing pancreatitis in patients with AERD [34]. Due to these adverse effects, patients may also discontinue ATAD therapy, with discontinuation rates as high as 63% in AERD patients [35].

Jerschow et al showed that among patients with worsening respiratory symptoms during aspirin treatment, 75% were Black and 25% were Latinx. Among those who failed to complete the initial desensitization, 85.7% were Black and 14.3% were Latinx, and none were white or non-Latinx patients [36]. Previous studies have shown disparities in asthma and AERD prevalence, with increased severity of sinus disease and asthma in minorities, as well as a higher prevalence of AERD/NERD in Black patients [36,37,38,39]. However, further studies are necessary to delineate the reasons for these disparities and how they can be rectified.

There are currently no studies identifying standardized predictors of who will tolerate ATAD. Additionally, there are no studies identifying specific clinical features associated with individual AERD endotypes, appropriate standardized biomarkers for identification of endotypes, or appropriate treatment plans based on endotypes. AERD is a heterogeneous disease process characterized by variability in inflammatory patterns, nasal polyposis severity, and time to polyp recurrence after surgery. Thus, beneficial ATAD outcomes are not universal. The impact of this heterogeneity in inflammatory patterns likely impacts patient selection and response to therapy, although there is a scarcity of related research [40].

However, long-term safety studies of ATAD in AERD patients are lacking, as are studies demonstrating the minimum effective dose for maintenance therapy or standardized treatment protocols for ASA desensitization. For patients who are not candidates for ASA desensitization, the use of leukotriene modifiers and biologics (biomarker inhibitors) can be considered. Studies show that the use of biomarker inhibitors can confer tolerance to ASA [41]. Unfortunately, there are currently no studies directly comparing ATAD to biologics for the treatment of asthma and CRSwNP in patients with AERD; further studies evaluating ATAD in combination with biologic therapy are also necessary [42]. A large meta-analysis comparing treatment of CRSwNP with different biologics (dupilumab, omalizumab, mepolizumab, and benralizumab), which target biomarkers (interleukin (IL)-4Rα, immunoglobulin E, IL-5, and IL-5Rα, respectively), and ASA desensitization revealed that dupilumab was the most beneficial for all seven outcomes analyzed [43]. Thus, the utility of long-term ASA therapy in combination with biologics and other medical and surgical treatments must be considered on an individualized basis with multidisciplinary specialists for a truly equitable and comprehensive approach.