To the Editor,
Severe immediate adverse events following immunization (AEFI) with COVID‐19 vaccines have been reported in up to 2.5 per 10,000 vaccinations. 1 Differentiating allergic reactions from immunization stress‐related responses (ISRR) 2 at the point of vaccination can be difficult, presenting an added burden to allergy clinics. Recent reports suggest that most individuals will tolerate mRNA re‐vaccination in a supervised setting. 3 , 4 More information is needed to help predict those most likely to tolerate future vaccine doses, as excipient skin testing has provided a limited role. 5
METHODS
This is a retrospective multi‐centre cohort study of adults referred to two allergy centres with immediate AEFI to the BNT162b2 or ChAdOx1s vaccines between April and August 2021. Anaphylaxis was defined as per Brighton collaboration criteria. 6 Skin testing and challenge procedures are in the eMethods in Appendix S1. The primary outcome was challenge positivity, stratified into subjective or objective signs or symptoms. To identify predictors of challenge positivity, univariable logistic regression then multivariable analysis via stepwise backward method was performed using variables with p < 0.2 on univariable analysis. Results are reported as odds ratios (OR) with 95% confidence intervals (CI).
RESULTS
A final cohort of 116 challenge patients were identified after excluding delayed AEFI and those who declined testing (Figure S1). The cohort is described in Table 1. Of these, 58.6% received at least 1 dose of adrenaline. Patients met case definition for anaphylaxis: Level One 11.2%, Level Two 34.5%, Level Three 2.6%. Sixty (51.7%) patients did not meet case definition (Table 1).
TABLE 1.
Baseline and index reaction characteristics
| Factor | Challenged | ChAdOx1s (AstraZeneca) | BNT162b2 (Pfizer) |
|---|---|---|---|
| N | 116 | 50 (43.1%) | 66 (56.9%) |
| Sex | |||
| Female | 108 (93.1%) | 47 (94%) | 61 (92%) |
| Age at first review, median (IQR) | 45.2 (35.75, 56.23) | 54.8 (45.7, 62.7) | 40.8 (32.8, 46.2) |
| Ethnicity | |||
| African | 0 (0.0%) | ||
| Asian | 1 (0.9%) | 1 (1.5%) | |
| Caucasian | 45 (38.8%) | 22 (44%) | 23 (34.9%) |
| Not recorded | 70 (60.3%) | 28 (56%) | 42 (63.6%) |
| Psychiatric history | |||
| None | 54 (46.6%) | 29 (58%) | 25 (38%) |
| Unknown | 35 (30.1%) | 11 (22%) | 24 (36%) |
| Anxiety | 16 (13.8%) | 6 (12%) | 10 (15%) |
| Bipolar | 1 (0.9%) | 0 (0%) | 1 (2%) |
| Depression | 10 (8.6%) | 4 (8%) | 6 (9%) |
| Prior COVID‐19 infection | |||
| Yes | 2 (1.7%) | 0 (0%) | 2 (3%) |
| Mastocytosis | 0 (0.0%) | 0 (0%) | 0 (0%) |
| Idiopathic anaphylaxis | 2 (1.7%) | 0 (0%) | 2 (3%) |
| Chronic spontaneous urticaria | 10 (8.6%) | 4 (8%) | 6 (9%) |
| Prior history of atopy | 23 (19.8%) | 13 (26%) | 10 (15%) |
| Allergic Rhinitis | 11 (9.5%) | 6 (12%) | 5 (8%) |
| Atopic Dermatitis | 2 (1.7%) | 1 (2%) | 1 (2%) |
| Asthma | 15 (12.9%) | 9 (18%) | 6 (9%) |
| Prior history of PEG or polysorbate allergy, suspected | 2 (1.7%) | 1 (2%) | 1 (2%) |
| Prior history of immune mediated food allergy history | 27 (23.3%) | 10 (20%) | 17 (26%) |
| History of food‐related anaphylaxis | 19 (16.4%) | 7 (14%) | 12 (18%) |
| Prior history of immune mediated drug allergy history | 33 (28.4%) | 16 (32%) | 17 (26%) |
| Penicillin allergy | 14 (12.1%) | 6 (12%) | 8 (12%) |
| Sulfa allergy | 4 (3.4%) | 2 (4%) | 2 (3%) |
| Other antibiotic allergy | 6 (5.2%) | 2 (4%) | 4 (6%) |
| NSAID allergy | 3 (2.6%) | 1 (2%) | 2 (3%) |
| Flu vaccine allergy | 1 (0.9%) | 1 (2%) | 0 (0%) |
| Other drug allergy | 23 (19.8%) | 10 (20%) | 13 (20%) |
| History of drug related anaphylaxis | 6 (5.2%) | 2 (4%) | 4 (6%) |
| Time since first vaccination (days), median (IQR) | 41 (27, 69) | 69 (42, 92) (n = 49) | 30 (22, 42) (n = 64) |
| Vaccine dose 1 | 115 (99.1%) | 50 (100%) | 65 (98.5%) |
| Vaccine dose 2 | 1 (0.9%) | 0 (0%) | 1 (1.5%) |
| Brighton criteria level of certainty | |||
| Level 1 | 13 (11.2%) | 5 (10%) | 8 (12.1%) |
| Level 2 | 40 (34.5%) | 18 (36%) | 22 (33.3%) |
| Level 3 | 3 (2.6%) | 2 (4%) | 1 (1.5%) |
| Did not meet case definition | 60 (51.7%) | 25 (50%) | 35 (53%) |
| Treatment | |||
| Unknown | 1 (0.9%) | 1 (2%) | 2 (3%) |
| No | 18 (15.5%) | 9 (18%) | 9 (14%) |
| Yes | 95 (81.9%) | 40 (80%) | 55 (83%) |
| Treatment at vaccination centre | 64 (55.2%) | 26 (52%) | 38 (58%) |
| Treatment at hospital or medical centre | 53 (45.7%) | 21 (42%) | 32 (48%) |
| Treatments received | |||
| Prednisolone | 16 (13.8%) | 6 (12%) | 10 (15%) |
| Antihistamine | 45 (38.8%) | 19 (38%) | 26 (39%) |
| Adrenaline | 68 (58.6%) | 24 (48%) | 44 (67%) |
| Unknown/other | 11 (9.5%) | 6 (12%) | 5 (8%) |
| Adrenaline administration site | |||
| Vaccination centre | 45 (38.8%) | 15 (30%) | 30 (45%) |
| Community medical centre | 6 (5.2%) | 5 (10%) | 1 (2%) |
| Ambulance | 8 (6.9%) | 2 (4%) | 6 (9%) |
| Hospital | 20 (17.2%) | 8 (16%) | 12 (18%) |
| Adrenaline doses total (IM) | |||
| 1 | 34 (29.3%) | 11 (22%) | 23 (34.9%) |
| 2 | 17 (14.7%) | 6 (12%) | 11 (16.7%) |
| 3 | 10 (8.6%) | 4 (8%) | 6 (9.1%) |
| 4 | 5 (4.3%) | 2 (4%) | 3 (4.6%) |
| 5 | 1 (0.9%) | 1 (2%) | 0 (0%) |
| Adrenaline infusion | 8 (6.9%) | 4 (8%) | 4 (6.1%) |
| Hospitalization | 41 (35.3%) | 18 (36%) | 23 (35%) |
| ICU admission | 5 (4.3%) | 2 (4%) | 3 (5%) |
Skin testing with PEG and Polysorbate80 panels was performed on 23 patients (19.7%), all were negative (Table S1). Forty‐five patients (38.7%) reported a positive challenge, of which 18 (40%) were to ChAdOx1s and 27 (60%) to BNT162b2. Forty (34% of the 116) were mild and subjective versus 5 (4%) with objective signs. Two (1.7%) patients received adrenaline for a combination of rash and throat tightness with normal vital signs. Of the five with objective signs, four did not meet Brighton criteria for their index reaction and one was Level 1.
Univariable analysis of factors that predict positive challenge is in Table S2. Predictors of positive challenge on multivariable analysis included absence of a history of atopy (OR 8.13 [95% CI, 1.66–39.70]), absence of hospitalization (OR 2.89 [95% CI, 0.97–8.66]) and any treatment received for the index reaction (OR 6.06 [95% CI, 1.46–25.16]). Those with a Brighton level 1 (OR 7.27 [95% CI, 1.52–34.66]) or not meeting case definition (OR 3.06 [95% CI, 1.01–9.34]) were more likely to have a positive challenge, compared with a Brighton level 2/3. (Table 2).
TABLE 2.
Predictors of positive challenge
| Variable | OR (95% CI) | p value |
|---|---|---|
| Absence of hospitalization | 2.89 (0.97, 8.66) | 0.057 |
| Any treatment | 6.06 (1.46, 25.16) | 0.013 |
| No history of atopy | 8.13 (1.66, 39.70) | 0.01 |
| Brighton diagnostic certainty | ||
| Level 1 | 7.27 (1.52, 34.66) | 0.049 |
| Level 2/3 | Ref | |
| No level | 3.06 (1.01, 9.34) | 0.049 |
DISCUSSION
In this cohort study of patients reporting immediate AEFI post‐COVID vaccination, 61.2% of the 116 had negative vaccine challenge with a further 34.4% developing mild, subjective symptoms post‐vaccination. Thus, 95.6% of the cohort were able to be re‐vaccinated safely with 4.3% developing objective signs of a possible immune AEFI, consistent with international experience. 2 , 3 , 4 Our study is one of the first to report vaccine challenge outcomes in a cohort including the ChAdOx1s vaccine.
Skin testing was not a useful predictor of challenge positivity (Table S1). Atopy and hospitalization were not associated with an increased risk of positive challenge, suggesting those at risk of positive challenge lack classical allergy phenotypes. As the 51.7% of patients who did not meet Brighton criteria had a higher risk of positive challenge, it's possible that these reactions are not immune‐mediated which may explain why those with severe reactions tolerated challenge. Based on our data, re‐vaccination is possible in those with immediate AEFI post‐COVID vaccination if selected carefully and challenged in a supervised environment.
CONFLICT OF INTEREST
None.
Supporting information
Appendix S1
Figure S1
ACKNOWLEDGEMENTS
We thank the Infectious Diseases Department of Austin Health, the Victorian Specialist Immunisation Service and the Victorian Department of Health for supporting this program. We also thank the dedicated nursing staff of both participating hospitals.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
REFERENCES
- 1. Blumenthal KG, Robinson LB, Camargo CA, et al. Acute allergic reactions to mRNA COVID‐19 vaccines. JAMA. 2021;325(15):1562‐1565. doi: 10.1001/jama.2021.3976 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. World Health Organization . Immunization stress‐related responses: a synopsis of the manual for program managers and health professionals to prevent, identify and respond to stress‐related responses following immunization. 2019. Accessed July 24, 2022. https://apps.who.int/iris/handle/10665/330276
- 3. Krantz MS, Kwah JH, Stone CA, et al. Safety evaluation of the second dose of messenger RNA COVID‐19 vaccines in patients with immediate reactions to the first dose. JAMA Intern Med. 2021;181(11):1530‐1533. doi: 10.1001/jamainternmed.2021.3779 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Tuong L‐AC, Capucilli P, Staicu M, Ramsey A, Walsh EE, Mustafa SS. Graded administration of second dose of Moderna and Pfizer‐BioNTech COVID‐19 mRNA vaccine in patients with hypersensitivity to first dose. Open Forum Infect Dis. 2021;8:ofab507. doi: 10.1093/ofid/ofab507 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Wolfson AR, Robinson LB, Li L, et al. First‐dose mRNA COVID‐19 vaccine allergic reactions: limited role for excipient skin testing. J Allergy Clin Immunol Pract. 2021;9(9):3308‐3320.e3. doi: 10.1016/j.jaip.2021.06.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Rüggeberg JU, Gold MS, Bayas J‐M, et al. Anaphylaxis: case definition and guidelines for data collection, analysis, and presentation of immunization safety data. Vaccine. 2007;25(31):5675‐5684. doi: 10.1016/j.vaccine.2007.02.064 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Appendix S1
Figure S1
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.