COVID-19 vaccine-related presumed allergic reactions and second dose administration by using a two-step graded protocol

Abstract

Background:

Acute allergic reactions to messenger RNA (mRNA) vaccines are rare but may limit public health immunization efforts.

Objectives:

To characterize suspected allergic reactions to the first dose of coronavirus disease 2019 (COVID-19) mRNA vaccine and to assess the safety and utility of a two-step graded-dose protocol for the second dose of the Pfizer-BioNTech vaccine in patients with a history of low suspicion of anaphylaxis to their first dose.

Methods:

This was a retrospective evaluation of referrals to the allergy and immunology clinic for a presumed allergic reaction to the first dose of the COVID-19 mRNA vaccine (Pfizer-BioNTech or Moderna) between December 17, 2020, and February 28, 2021. Recommendations for the second dose and outcomes were evaluated by trained board-certified allergists.

Results:

Seventy-seven patients presented with a Pfizer-BioNTech reaction (56 [72.7%]) or with a Moderna reaction (21 [27.3%]). Most patients (69.7%) had symptom onset within 4 hours. Most commonly reported symptoms were cutaneous (51.9%), cardiovascular (48.1%), and respiratory (33.8%) symptoms. Recommendations included to proceed with the single dose (70.1%), two-step graded dose (19.5%), or deferral (10.4%). Twelve of 15 patients completed the second dose with a graded-dose protocol. Of these patients, five reported at least one or more similar symptoms as experienced with their first dose.

Conclusion:

Of the patients with presumed allergic reactions to their first dose of COVID-19 mRNA vaccine, most were able to safely receive the second dose. For those with a low suspicion of anaphylaxis, the two-step graded protocol with the Pfizer-BioNTech vaccine was well tolerated. A graded-dose protocol could be an effective strategy for second-dose vaccination in those who may otherwise defer the second dose.

Pfizer-BioNTech (BNT162b2) (Pfizer Inc, New York, New York, BioNTech SE, Mainz, Germany) and Moderna (mRNA-1273) (Moderna, Inc, Cambridge, Massachusetts) are messenger RNA (mRNA) vaccines available in the United States to prevent coronavirus disease 2019 (COVID-19). Suspected anaphylaxis has been reported with both vaccines, at an estimated rate of 2.5–4.7 cases/million doses administered.¹ Although risk-stratification pathways exist for assessing both prevaccine and first-dose reactions, more data are needed to better evaluate individuals with a presumed first-dose allergic reaction, especially because some cases of vaccine deferral may be due to patients' concern for allergic reactions.^2,3 We report our clinical experience in evaluating first-dose reactions and recommendations for second-dose administration, including the use of a graded-dose protocol.

METHODS

We retrospectively evaluated patients referred to the Northwestern Medicine Allergy and Immunology Service for a presumed allergic reaction to the first dose of either of the mRNA vaccines between December 17, 2020, and February 28, 2021. The patients were followed up until May 4, 2021, to assess if they received a second dose and if a subsequent reaction was documented. We used a standardized questionnaire for evaluating the vaccine reactions to maintain consistency in the assessment (Supplemental figure S1). Given the novelty of these vaccines, the patients were discussed twice a week at allergy and immunology division rounds to develop a consensus decision on recommendations for each patient. Final provider recommendations for the second dose were based on the managing physician's clinical judgment and included proceeding with the full dose of the vaccine with one injection, proceeding with administration by using a two-step graded-dose protocol (limited to Pfizer-BioNTech recipients due to vaccine availability and access at the time of these reactions), or deferral of vaccination.

We did not perform a skin test to polyethylene glycol due to the lack of predictive values at the time of evaluation and the unknown role in identifying anaphylaxis to mRNA vaccines.⁴ For patients who were recommended to undergo a graded-dose protocol, we applied the Brighton Collaboration and the Consortium for Food Allergy Research (CoFAR) criteria to assess the severity of the reaction. However, this was not used as part of clinical decision-making at the time of evaluation. The Brighton Collaboration case definition of anaphylaxis is based on the level of certainty, with the highest certainty starting at level 1, followed by level 2 and level 3; the CoFAR grading scale ranges from grades 1 to 5 to reflect the severity of the allergic reaction with increasing grades ^5,6

Our two-step graded-dose protocol for the second dose of the Pfizer-BioNTech vaccine included a 0.05-mL followed by a 0.25-mL intramuscular injection in the same arm, with 30 minutes in between doses. All the patients remained under the allergist's (G.P., A.P., or C.S.) observation for a minimum of 60 minutes after vaccination. Vital signs were taken and physical examinations were performed for each patient before vaccine administration, after the initial dose, and after the final dose. Additional physical examinations were performed as needed throughout the graded-dose protocol based on symptoms. This study was approved by Northwestern University Feinberg School of Medicine's institutional review board.

RESULTS

A total of 77 patients had presumed allergic reactions to the first dose of the COVID-19 vaccines (56 with Pfizer-BioNTech; 21 with Moderna). Patient demographics and allergic history are shown in Table 1. Nearly all the subjects were women. Most (69.7%) developed symptoms within 4 hours of the vaccination, with an increased frequency in those who received the Pfizer-BioNTech (80%) compared with the Moderna (43%) vaccine (Fig. 1 A). Cutaneous (51.9%), cardiovascular (48.1%), and respiratory (33.8%) symptoms were commonly reported (Fig. 1 B). Cutaneous symptoms occurred more frequently in those who received the Moderna vaccine (local injection site, 33.3%; non-local, 66.7%) compared with the Pfizer-BioNTech vaccine (local injection site, 17.9%; non-local, 46.4%). Three patients (3.9%) received epinephrine for their initial first-dose reaction (Fig. 1 C). Fifty-four patients (70.1%) were recommended to proceed with the second injection of the full-dose vaccine, followed by a 15-to-30-minute observation period. A two-step graded-dose protocol was recommended to 15 patients (19.5%): 1 who had a history of delayed (>4 hours) rash and 14 who had immediate symptoms (≤4 hours) after their first dose. Eight (10.4%) were recommended to defer the second dose (Fig. 2).

Table 1.

Characteristics and demographics of the patients who presented to the allergy and immunology clinic with presumed allergic reactions to the first dose of COVID-19 messenger RNA vaccine (N = 77)

COVID-19 = Coronavirus disease 2019.

*A COVID-19 diagnosis based on laboratory confirmed results or self-report.

#Hydroxychloroquine (n = 2), rituximab and prednisone (n = 1), tofacitinib (n = 1), ustekinumab (n = 1), adalimumab (n = 1).

Figure 1.

Time until sign and symptom onset, and categories of signs and symptoms reported in all the patients evaluated for presumed allergic reaction to the first dose of their messenger RNA (mRNA) coronavirus disease 2019 (COVID-19) vaccine. (A) Initial reaction onset was ≤ 4 hours for 69.7% of the cohort. (B) Common reactions reported by the patients after the first dose. (C) Approximately 42.9% sought medical care and received one or more doses of an antihistamine for their reaction; 3.9% of the patients required epinephrine administration, and 2.6% required hospitalization. P = Pfizer-BioNTech; M =Moderna; T = total; local cutaneous = redness or swelling at site of injection; generalized cutaneous = facial swelling, generalized itch, flushing, hives, other rash; upper airway = throat swelling or tightness, mouth tingling, or trouble swallowing or talking; respiratory = shortness of breath, wheezing, cough, chest pain, or tightness; CV = cardiovascular: tachycardia, dizziness, or syncope; GI = gastrointestinal: nausea, vomiting, abdominal pain, or diarrhea; constitutional = fever, fatigue, sweats, headache, or weakness.

Figure 2.

Second-dose recommendations and the rate of complete vaccinations for patients with a coronavirus disease 2019 (COVID-19) first-dose vaccine reaction. Physician recommendations included (1) proceed with full dose in one injection, (2) proceed with a two-step graded dose, or (3) defer the vaccination.

Of the group recommended to proceed with the standard full dose, 41 of 54 (75.9%) received the second dose without any evidence of an allergic reaction. Of the remaining 13 patients, 5 chose not to proceed with the second dose, and 8 were lost to follow-up. Eight patients were recommended to defer the second dose. Five of these eight patients had reactions to the first dose that were concerning for anaphylaxis (n = 4, Pfizer-BioNTech vaccine; n = 1, Moderna vaccine). All five fulfilled the Brighton Collaboration Criteria of level 1 for the highest level of diagnostic certainty for anaphylaxis and the CoFAR grade 3 (n = 5) severity.^5,6 Three additional patients were deferred for the following reasons: uncontrolled urticaria (n = 1, Moderna vaccine), neurologist recommended deferral (n = 1, Pfizer-BioNTech vaccine; the patient had a history of sequelae from concussions, with transient episodes of tongue and arm heaviness after the first dose), and delayed, significant uvular angioedema (n = 1, Pfizer-BioNTech vaccine).

Among the 15 patients recommended for the graded-dose protocol to the Pfizer-BioNTech vaccine, 14 had immediate symptoms (≤4 hours) and 1 had delayed symptoms (>4 hours). Provided in Table 2 are the details of all 15 patients who were recommended for the two-step graded-dose protocol and compared with the other two groups (to proceed with full dose and deferral). Twelve completed the graded-dose vaccine protocol. Of these 12 patients, 6 reported symptoms within an hour of administration of a vaccine dose, whereas 6 reported no symptoms. Patient no. 1 developed tightness of her fingers 45 minutes after the second dose (which did not meet CoFAR or Brighton criteria). Her physical examination was normal. She was observed for 45 minutes after symptom developed and then was discharged. Patient no. 2 developed pruritus of the face, neck, and upper back 30 minutes after the 0.05-mL dose (CoFAR grade 1; did not meet the Brighton criteria). A physical examination revealed linear wheals in areas where she scratched, suggestive of dermographism. Levocetirizine 5 mg was administered orally and then the patient was monitored for an additional 10 minutes while the dermographism resolved. The patient received the remaining 0.25-mL dose without an adverse reaction.

Table 2.

A comparison of demographics, allergic history, and self-reported symptoms after the first dose of the Pfizer-BioNTech vaccine in the patients recommended for full-dose, graded-dose protocol or deferral^*

^*Two-step graded-dose protocol for the second dose of the Pfizer-BioNTech vaccine, defined as the administration of 0.05 mL, followed by 0.25 mL intramuscular injection in the same arm, with 30 minutes in between doses; all patients who underwent the graded-dose protocol remained under an allergist's observation for a minimum of 60 min after vaccination.

#Symptoms included pruritus, dermatographism, shortness of breath, throat swelling sensation, dizziness, palpitations, or tightness in the fingers. One patient underwent a three-step graded protocol with 0.05 mL, 0.125 mL, and 0.125 mL doses.

§Value represents a sum of level of care for observation by a physician, primary care physician, urgent care, and emergency department.

Patient no. 3 developed subjective throat swelling and dizziness within 15 minutes of the 0.05-mL dose (CoFAR grade 1; did not meet Brighton criteria). Given a normal physical examination and vital signs, the second dose was administered per protocol. No treatment was administered. Symptoms resolved within 60 minutes. Patient no. 4 reported dizziness and palpitations within 15 minutes of the 0.05-mL injection (CoFAR grade 1; did not meet Brighton criteria). The examination was notable for a heart rate of 113 beats/min compared with a baseline of 100 beats/min. She was normotensive. This self-resolved after 15 minutes, and the patient received the second dose without recurrence. Patient no. 5 reported dizziness 60 minutes after the 0.25-mL dose; however, the vital signs were stable (did not meet CoFAR or Brighton criteria). The patient was monitored for 60 additional minutes, with resolution of her dizziness. Of note, this patient reported preexisting intermittent lightheadedness. Patient no. 6, who had a history of postural orthostatic tachycardia syndrome, developed itching and shortness of breath after the initial dose (CoFAR grade 1; did not meet Brighton criteria). The observing allergist (G.P.) chose to perform a slower graded-dose challenge, so the second dose was administered as 0.125 mL. The patient reported mild abdominal pain and was monitored for an additional 10 minutes. The final 0.125 mL was administered, and she was observed for 75 minutes without any further events.

Of the remaining three patients who were recommended to receive the graded-dose protocol, two opted to not receive a second dose, and one opted for a standard full dose after a risk-benefit discussion. This latter patient's reaction after the first dose of the vaccine included generalized pruritus, flushed sensation, sweating, and minor oral tingling (CoFAR grade 2; did not meet Brighton criteria for anaphylaxis). Immediately after the second dose of the vaccine, she experienced arm and oral tingling, anxiety, flushing sensation, difficulty swallowing, lightheadedness, and nausea (CoFAR grade 3; Brighton Collaboration Criteria 1). The patient received diphenhydramine 50 mg orally, followed by two epinephrine injections 10 minutes apart due to a lack of symptom improvement. Her blood pressure was self-reported to be elevated before the epinephrine administration. A tryptase level was not obtained. This patient's first and second dose reactions were not observed by an allergist. This was the only patient in this cohort to receive epinephrine after getting the second dose of the vaccine.

DISCUSSION

In this retrospective cohort of first-dose, presumed allergic vaccine reaction referrals to the adult Northwestern Medicine Allergy and Immunology Service, 53 of 69 patients (76.8%) who were recommended to proceed with a second dose of the mRNA vaccine were able to receive it either by a single dose or a two-step graded-dose protocol. To our knowledge, this was the first study that described the safety of a one-sixth, five-sixth, two-step graded dose protocol with the Pfizer-BioNTech vaccine. Currently, only one report describes a graded-dose protocol for a COVID-19 mRNA vaccine (Moderna).⁷

The American Academy of Allergy Asthma and Immunology and the American College of Allergy, Asthma and Immunology Joint Task Force Practice Parameters for adverse reactions to vaccines support graded-dose vaccine administration for those who have experienced vaccine reactions.⁸ Specifically, studies that evaluated influenza vaccine administration in patients with an egg allergy by using the two-step graded administration without skin testing were observed to be safe.⁹ During the COVID-19 pandemic, COVID-19 vaccine skin testing is not allowed under the emergency use authorization. Thus, a graded-vaccine dosing strategy was the only immediate option available for risk mitigation and potentially improves the success of receiving a second dose of the mRNA vaccines in patients with a history of presumed allergic reaction to the first dose. This may be an effective approach, especially for individuals who are vaccine hesitant and who are concerned about an allergic reaction and would otherwise forgo the second dose.

Clinicians face challenging dilemmas when evaluating unconfirmed first-dose allergic reactions and risk-stratifying patients who may potentially be at risk for a serious allergic reaction with the second dose. The risks and benefits based on currently available evidence should be discussed with the patient through shared decision-making to improve clinical outcomes because there is no single best-choice approach for guiding second doses.^10,11 The advantages of a graded challenge include supervised observation and reassurance to the patient that a potential reaction can be safely managed. A primary risk of graded-dose administration is the theoretically reduced efficacy of this approach weighed against deferring the second dose and having incomplete immunity.¹¹ There is evolving research, results of which show that fractionated mRNA doses could elicit a robust immune response, which suggests that a two-step graded protocol may not reduce immunogenic efficacy.¹²

A preliminary report that assessed the safety and immunogenicity of half of the dose (50 µg) of the current Moderna mRNA-1273 (100 µg) vaccine showed that binding and neutralizing antibody responses were generally comparable in both groups at all time points (formal statistical testing was not performed).¹³ A clinical trial is ongoing that is investigating the safety and immunogenicity of Pfizer-BioNTech mRNA 20 µg versus 30 µg (NCT04852861).¹⁴ Although further confirmation is needed, including clinical outcomes, this suggests that the amount of mRNA in a second dose of a two-step graded protocol could be sufficient to elicit a robust immune response. Currently, the Canadian Society of Allergy and Clinical Immunology and the European Academy of Allergy and Clinical Immunology support the utilization of a graded protocol if there is concern for an allergic reaction.^15,16 Furthermore, recent recommendations based on an international expert consensus favor vaccination through full or graded dosing, or changing vaccine platforms to another agent over no vaccination.¹¹ Future studies on a larger scale will be important to understand how to best risk stratify patients for single injection versus graded-dose vaccine versus deferral of the vaccine.

The strengths of this study included the use of a standardized history, and all the patients were evaluated by allergy and immunology faculty. Limitations included a small cohort from a single academic institution, so findings may not be generalizable. Severe acute respiratory syndrome coronavirus 2 antibodies were not checked to determine the efficacy of graded-vaccine administration because there is no defined level of immune protection. Logistical considerations, including access to vaccines or institutional restrictions on performing graded-dose administration due to the emergency use authorization, limit this study's generalizability to other clinicians. Also, the Pfizer-BioNTech vaccine was the only available vaccine at our institution at the time, so we were unable to recommend a similar graded-vaccine protocol with the Moderna vaccine.

CONCLUSION

Allergists play a critical role in COVID-19 vaccination efforts, especially related to assessing and mitigating risk and vaccination fears. Analysis of our preliminary data suggests that most of the first-dose vaccine reactions were unlikely immunoglobulin E–mediated reactions because the vast majority of our patients could safely receive their second dose as a single injection. The use of a two-step graded-dose administration could be offered to those patients with a history of presumed mild-to-moderate allergic reactions and who would likely defer the second dose of the vaccine without this option.