Patients with suspected allergic reactions to COVID‐19 vaccines can be safely revaccinated after diagnostic work‐up

Abstract Background When initiating the Danish vaccination program against COVID‐19, the incidence of anaphylaxis was estimated to be 10 times higher compared to other virus‐based vaccines. In this study, we present data on patients referred with suspected allergic reactions to COVID‐19 vaccines. The main purpose of the study is to investigate the incidence and severity of the allergic reactions, and to evaluate the safety of revaccination. Methods All patients in the region of Southern Denmark with case histories of allergic reactions to COVID‐19 vaccines in a defined period are included in this study. Diagnostic work up consisted of a detailed case history, evaluation of Brighton level of diagnostic certainty and World Allergy Organization grade of anaphylaxis and skin prick testing‐ and basophil histamine release testing with COVID‐19 vaccines and relevant drug excipients. Patients were revaccinated at the Allergy Center when possible. Results Sixty‐one patients are included in this study. In 199,377 doses administered, nine patients fulfilled the criteria of anaphylaxis when using the Brighton Criteria (incidence being 45 per million). Of 55 patients with reactions to the first dose, 52 patients were revaccinated without adverse reactions. We found no proven cases of immediate anaphylaxis due to COVID‐19 vaccines. By skin prick test, we diagnosed three patients with drug excipient allergy and further a patient with mastocytosis was found. Conclusions Anaphylactic reactions to COVID‐19 vaccines are rare and the incidence is similar to what is seen with other virus‐based vaccines. Revaccination is safe in the majority of patients; however, allergological evaluation is important since some prove to have drug excipient allergy.


| INTRODUCTION
Virus-based vaccines are expected to elicit anaphylactic reactions with a frequency of 1.3:1,000,000. 1 Based on data from the United States, 2 the COVID-19 vaccines are suspected to elicit reactions more frequently, estimated by the Danish authorities to 1:100,000. 3 For safety reasons, we therefore set up a diagnostic routine prior to vaccination in order to investigate reactions to COVID-19 vaccines and to diminish reactions to COVID-19 vaccines by identifying patients with high risk of reacting to vaccines. We thus decided to classify citizens into four groups, the first being citizens experiencing an allergic reaction to the COVID-19 vaccine. These patients are described here. The second group consists of all Danish patients with a diagnosis of systemic mastocytosis, where an even higher frequency of anaphylaxis may be expected, although solid data are missing. 4 All Danish patients with mastocytosis will be vaccinated in our department, where full anaphylaxis staff and equipment are available. The third group consists of approximately 25 patients with an already established diagnosis of allergy to drug excipients, mostly macrogols/polyethylene glycols (PEG), or polysorbates. Finally, patients with a previous reaction to a virus-based vaccine, or to parenteral drugs, containing PEG or polysorbate, are referred to the Allergy Center for evaluation prior to COVID-19 vaccination. Data from the three latter groups will be presented elsewhere.

The vaccination program for citizens in the region of Southern
Denmark is organized in seven centers, supplemented with mobile vaccination clinics for nonambulant people. Health professionals are vaccinated in the hospitals. Our department is responsible for vaccinating the staff of Odense University Hospital (8000 persons) and the citizens from the major part of the island Funen (500,000 citizens). All serious immediate reactions to a COVID-19 vaccine are treated initially in the vaccination center and afterwards in the nearest acute ward. Delayed reactions are treated either in the acute ward, other medical departments, or by the patient's general practitioner (GP).
All adverse reactions to COVID- 19  Histamine solution (10 mg/ml) and saline were used as positive and negative control, respectively. 9,10 The size of the resulting wheals was recorded after 15 min and wheal size was measured on the longest and shortest perpendicular axis, the numbers were added and divided by two (mean wheal diameter). Wheals ≥3 mm larger than the negative control were considered positive. Blood was drawn for measurement of specific IgE to latex protein and chlorhexidine (Thermo Fisher Scientific), and for basophil histamine release (BaHR) (www.Reflab.dk), using the same allergens as in SPT in six dilutions. 11 BaHR was only considered significantly positive, when a bell shaped curve with at least two positive values above baseline was obtained.
BaHR with release above 15 ng/ml, not fulfilling these criteria, were considered marginally positive, but treated as negative when evaluating possibility for revaccination. S-tryptase level (Thermo Fisher Inc) and c-KIT mutation 12 were measured. c-KIT mutation in blood was detected by using real-time qPCR assay. 12 Baseline level of stryptase >12 μg/L was considered elevated, and a s-tryptase level, measured following the acute allergic reaction, was considered

| Ethics
In this study, we report the results of our standard diagnostic work up for patients with suspected allergic reactions to a COVID-19 Vaccine. Written informed consent was obtained from all patients.

| RESULTS
Sixty-one patients were referred to the Allergy Center after a case history of an allergic reaction to a COVID-19 vaccine: 30, 6, and 25 patients who were vaccinated with the PB-, M-and AZ-vaccine, respectively. This cohort includes 54 females and 7 males: age ranging from 18 to 88 years (median 46 years).
In Table 2, patients are arranged according to Brighton level, 6 time of onset of the adverse reaction, the patient's primary treatment place, and the treatment administered. Nine patients were meeting the criteria of anaphylaxis according to the Brighton criteria (level 1 through 3).
In Table 3, the patients are arranged according to onset and severity of reactions using the WAO criteria of anaphylaxis. 8  Thirty-five patients had late onset adverse reactions 6 h or more after vaccination, including patient (N60) with severe anaphylaxis (Tables 2 and 3). Most patients suffered from skin symptoms: Urticaria, rash, itch, angioedema, and localized injection site reactions, for some coupled with subjective respiratory symptoms or gastrointestinal symptoms. In addition, a case of asthma attack in an asthma patient was seen. In the vast majority of cases, urticaria, angioedema, and rash were photo documented by the patients.
No patients reported a prior history of allergic reactions to vaccines, and the cohort had in average received 1.6 virus-based vaccines during the last 5 years, Table 3. Thirty-nine patients (64%) were or had been suffering from at least one atopic disease: Hay fever 36%, eczema 23%, and/or asthma 20%, and 20% had chronic spontaneous urticaria/angioedema. We did not find a correlation between atopy in the patient and severity nor timing of the reaction to COVID-19 vaccine, Table 3.  Table 4. He was successfully revaccinated after discontinuation of the ACE inhibitor.

Symptoms and signs in patients
Note: Concomitant atopy, nonatopic drug allergy and chronic spontaneous urticaria/angioedema, presence of cofactors during the reaction, and history of previous vaccinations are also presented. WAO grade 1-2 constitute nonanaphylaxis, grades 3-5 is consistent with anaphylaxis when meeting the criteria of anaphylaxis: Criteria 1 or 2 as given in Table 1.   15 These discrepancies remain to be explained.
In a large population based study, vaccines are generally thought to induce anaphylaxis at a rate of 1.31 cases per million doses. 1  Another explanation could be that evaluation by the authorities based on written information without assistance of supplemental information from the hospital nor from the patient (preferentially with photos taken by the patient or by relatives) overestimates the incidence as we previously have reported for cases of anaphylaxis in the acute ward. 16 The majority of the reactions, fulfilling the Brighton level 1 through 3 criteria, was immediate, elicited within 30 min after injection. Same pattern is seen in other studies. 17 We were able to successfully revaccinate the vast majority of patients, including the patients from Brighton level 2 through 3, indicating, that the majority of patients with reported severe reactions, did not had a true allergic reaction, which is in line with recently published data. 18  In this cohort, both diagnostic scoring systems were able to identify the one patient suffering from true anaphylaxis. The Brighton classification is widely used by the authorities in connection with allergic reactions to drugs including vaccines, whereas the WAO criteria are applied in allergologist settings. While the Brighton criteria focus on the level of diagnostic certainty demanding multiorgan involvement in cases of anaphylaxis, 6 the WAO criteria focus on the severity of symptoms not demanding multiorgan involvement. 8 In this study, the concordance between the tools is in line with previous correlations found between different tools for scoring anaphylaxis. 22 Similar differences have been obtained between the Brighton, Ring and Messmer, and NIAID/FAAN validated scales. 18 Maybe in the future both scoring systems should be applied for evaluating anaphylaxis to vaccines. The main take home message is however, that the value of diagnostic work up including interview and testing remains superior to written reports.
In conclusion, we showed that anaphylactic reactions in connection with COVID-19 vaccination are rare. After proper diagnostic work up, it is safe to revaccinate the vast majority of patients with an adverse event to a COVID-19 vaccine, as most patients with an immediate adverse reaction did not have true allergic reactions. All patients, with late onset hypersensitivity reactions, tolerated revaccination. We, however, diagnosed three patients with drug excipient allergy, stressing the importance of proper evaluation of patients with suspected allergic reactions to vaccines in order to avoid future adverse reactions in these patients. Thus in the majority of cases allergic reactions to the first COVID-19 vaccine should not prevent the citizen from receiving the second dose, but should prompt allergological testing prior to revaccination. Furthermore, the incidence of allergic reactions to COVID-19 vaccines seems similar to other virus-based vaccines.

F I G U R E 2
Correlation between Brighton and WAO scoring systems for anaphylaxis. A significant correlation is presented. Spearman correlation coefficient −0.70; p < 0.01