Ethical Approval and Trial Registration Statements
The study was approved by the Clinical Research Ethics Committee (CEIC) of the University Hospital Fundación Alcorcón in 2021.
To the Editor,
Non-steroidal anti-inflammatory drugs (NSAIDs) are the second cause (11%) of drug-related allergic reactions after antibiotics1, with ibuprofen being the most frequently involved2. NSAIDs are widely used in the pediatric population for their antipyretic, analgesic, and anti-inflammatory effects3, consumed by 16.9% of children in Europe4.
In patients diagnosed with hypersensitivity to NSAIDs, attempts are being made to find other NSAID-tolerant alternatives, such as weak COX-1 inhibitors or selective COX-2 inhibitors. However, many of these NSAIDs are contraindicated in children, leaving patients with few therapeutic options. Studies about the natural history of cross- intolerance to NSAIDs in the pediatric population are scarce.
The study design was observational and retrospective, focusing on patients under 18 years old with cross intolerance to NSAIDs by drug provocation tests (DPT), between 1999 and 2019 [visit 1 (V1)].
These patients were prospectively re-assessed in 2021-2022 [visit 2 (V2)] by DPT with acetylsalicylic acid (ASA) and the NSAID involved in their initial reaction to determine whether they had developed tolerance to NSAIDs over time.
The aim of this study was to assess the clinical characteristics of these patients and identify potential factors that may have contributed to the development of tolerance. The study was conducted by the ethical principles of the Declaration of Helsinki and was approved by the Clinical Research Ethics Committee (CEIC) of the University Hospital Fundación Alcorcon. All patients and/or legal representatives were informed about the study and signed the corresponding informed consent.
Patients were categorized into different phenotypes based on their clinical symptoms5. These included NSAID-exacerbated respiratory disease (NERD) for patients with rhinitis and/or asthma, NSAID-exacerbated cutaneous disease (NECD) for patients with chronic urticaria, NSAID-induced urticaria/angioedema (NIUA) for those without baseline pathology, and mixed or blended reactions for patients showing a combination of urticaria/angioedema and respiratory symptoms.
Several socio-demographic variables, such as gender and age at diagnosis, were collected from patient records during the V1. Clinical variables included the patient’s atopy, rhinitis or asthma, sensitization to environmental allergens, food allergy, atopic dermatitis and/or baseline chronic urticaria.
The main variable of the study was the development of tolerance or intolerance to NSAIDs, which was determined based on the patient’s reaction during the oral challenge test at V2. Patients who exhibited symptoms during DPT at V2 were classified as intolerant to NSAIDs, maintaining their original diagnosis of cross intolerance.
DPT were based on ENDA recommendations6 and modified based on patient’s weight. Patients remained under medical observation for up to two hours after administration of the last dose of the drug. If they showed no signs of hypersensitivity reaction to the drug, another DPT was administered with the NSAID involved. DPT were considered positive if symptoms or objective signs, such as urticaria, angioedema, rhinorrhoea, sneezing, bronchospasms, dyspnoea, hypotension, anaphylaxis or a >15% decrease in the forced expiratory volume in one second (FEV1), were observed during the test. In case of a positive reaction, the study was discontinued, and patients received symptomatic treatment.
Between 1999 and 2019, 46 patients aged 1-17 years old with a diagnosis of cross intolerance to NSAIDs were recruited. Of these, 19 were re-assessed in 2021-2022 (V2).
Of the 46 patients who were initially diagnosed (V1), 29 (63%) were male, and median age was 10 (IR 7-13) years. Clinical and demographic data based on the phenotype of the reaction are shown in Table 1 .
Half of patients had a history of atopy, mainly to pollens. Regarding history of allergies, the most frequent corresponded to rhinitis and asthma, followed by food allergy, atopic dermatitis, and urticaria. Of the six patients with food allergy, one half was allergic to seafood and the other half, to egg proteins, fruit, and dried fruit.
Angioedema was the most occurring symptom in reactions. The most frequent phenotype was NIUA, followed by blended reactions.
Mean time of diagnosis was 9.4 months. Ibuprofen was the most involved NSAID in reactions.
Regarding the initial oral challenge, 26 patients were challenged with ASA, 10 with ibuprofen and 15 with metamizole. The mean cumulative dose at which patients had a reaction at V1 were 545.4 (150.0 – 875.0), 288.4 (11.0 – 400.0) and 707.8 (318.0 – 962.0) respectively.
For the 19 patients who were re-assessed (V2), the median (IR) time between V1 and V2 was of 116 (57-153) months. In this study, 15 (78.9%) patients were tolerant to ASA, whereas four (21.0%) showed symptoms. The cumulative dose at which they had a reaction at V2 was higher than at V1 (Table 2 ).
Table 3 shows the differences between patients who developed tolerance and those who maintain their diagnosis of cross intolerance to NSAIDs. The group that remained classified as hypersensitive to NSAIDs had a higher percentage of atopy, history of rhinitis and asthma, and sensitization to mites, animal hair, or fungi. Patients who initially presented with asthma maintained their diagnosis at V2 (26.6% vs. 100%; 6.6% vs. 50%, respectively), whereas 93% of patients who initially presented with angioedema had developed tolerance at V2.
No differences were observed regarding gender, age at diagnosis, the number of previous reactions, the interval between taking the drug and having a reaction, or the NSAID involved.
The natural history of NSAID hypersensitivity in the pediatric population is currently unknown. Our study confirmed that 78.9% of patients were tolerant to NSAIDs at least 2 years after diagnosis. In addition, greater tolerance to cumulative doses of ASA was observed in patients with consistent positive challenge test results, which could be considered a sign of probable subsequent tolerance over the years. Tolerance to NSAIDs at V2 was found to be higher in patients with a baseline history of angioedema compared to patients with other baseline reactions such as respiratory conditions and/or history of asthma and/or rhinitis.
Studies about the natural history of drug allergy in the pediatric population are scarce. In 2017, one study similar to ours was published7, with the difference that the study population was older (14-60 years) and only the NIUA phenotype was assessed. In this study, 63.15% tolerated NSAIDs and as in our study it was observed that patients who initially presented isolated angioedema and/or were not atopic were more likely to develop tolerance after a few years; or patients who remained intolerant to ASA in V2 showed symptoms with a higher dose than in V1.
Several theories have been proposed as potential causes for later tolerance in children8, such as the effect of the patient’s age on the drug metabolism and the effects of certain co-factors, such as infections9, physical exercise, or food allergies10. In our study, no differences have been found regarding patients’ age or food allergies. However, the involvement of infections or exercise has not been analysed, since one condition for the performance of controlled challenge tests was that patients did not present with an active infection and without the possibility of doing exercise.
In conclusion, our results suggest that the study of later tolerance in children with cross intolerance to NSAIDs is safe and useful. For this reason, we propose that children diagnosed with cross intolerance to NSAIDs are subjected to a re-assessment after their initial diagnosis over the years. Furthermore, we consider that the period of re-assessment might be shorter for non-atopic patients and those who initially presented with isolated angioedema.