Associations of neuralgic amyotrophy with COVID‐19 vaccination: Disproportionality analysis using the World Health Organization pharmacovigilance database

Abstract

Introduction/Aims

There are limited studies on the association of COVID‐19 vaccination with neuralgic amyotrophy (NA). Therefore, we evaluated the association between COVID‐19 vaccination and the occurrence of NA.

Methods

We explored unexpected safety signals for NA related to COVID‐19 vaccination through disproportionality analysis using VigiBase, the World Health Organization's pharmacovigilance database.

Results

On October 15, 2021, 335 cases of NA were identified in the database. The median time to onset of NA after vaccination was around 2 weeks. A significant signal of disproportionality of NA was observed for the ChAdOx1 nCoV‐19 vaccine (AstraZeneca) (information component [IC]₀₂₅ = 0.33, reporting odds ratio [ROR]₀₂₅ = 1.30) and two mRNA‐based COVID‐19 vaccines (BNT162b2 [Pfizer and BioNTech] and mRNA‐1273 [Moderna]) (IC₀₂₅ = 1.74, ROR₀₂₅ = 3.82) compared with the entire database. However, when compared with influenza vaccines, we did not detect any signal of disproportionality of NA for both the ChAdOx1 nCoV‐19 vaccine (IC₀₂₅ = −2.71, ROR₀₂₅ = 0.05) and mRNA‐based COVID‐19 vaccines (IC₀₂₅ = −1.38, ROR₀₂₅ = 0.13).

Discussion

A weak association was observed between NA and COVID‐19 vaccines. However, the risk did not surpass that of influenza vaccines.

Abbreviations

ADR

adverse drug reaction

CI

confidence interval

IC

information component

MedDRA

Medical Dictionary for Drug Regulatory Activities

NA

neuralgic amyotrophy

PT

preferred term

ROR

reporting odds ratio

WHO

World Health Organization

1. INTRODUCTION

Neuralgic amyotrophy (NA) typically starts with severe unilateral shoulder and arm pain, followed by motor weakness and atrophy and less severe sensory deficits due to involvement of the brachial plexus or of its individual branches.1, 2 Recent infection and vaccination are known to be the most frequent risk factors for its development, accounting for 25% and 15% of reported NA cases, respectively. ¹ Its pathophysiology has not been unequivocally determined; however, molecular mimicry and bystander activation of immune‐mediated mechanisms against the brachial plexus in genetically susceptible people may be causes. ² Several patients with NA have been reported after COVID‐19 vaccination, suggesting an association between these events.3, 4

The aim of this study was to assess for any association between NA and COVID‐19 vaccination using VigiBase, the World Health Organization (WHO) global pharmacovigilance database.

2. METHODS

Disproportionality analysis is a method to evaluate the extent to which the incidence of adverse drug reactions (ADRs) increases for specific drugs compared with the subject with no relationship.5, 6 Bayesian disproportionality analyses of the ADRs of BNT162b2 (Pfizer and BioNTech), mRNA‐1273 (Moderna), and ChAdOx1 nCoV‐19 (AstraZeneca) vaccinations were performed using case safety reports in VigiBase, a deduplication database of the WHO, as described in previous work.5, 6, 7, 8, 9, 10 In brief, on October 15, 2021, all cases of NA reported as ADRs from BNT162b2, mRNA‐1273, or ChAdOx1 nCoV‐19 vaccination in VigiBase that used the following preferred terms (PTs) from the Medical Dictionary for Drug Regulatory Activities (MedDRA) were extracted: “neuralgic amyotrophy,” “radiculitis brachial,” and “brachial plexopathy.” ¹¹ Clinical data (age, sex, type of vaccine, time to onset of NA symptoms, reporting regions, seriousness, and final outcomes) were acquired from VigiBase. Time to onset (days) of NA was determined by calculating the interval between vaccination date and occurrence date of the ADR. A serious ADR was defined as one resulting in significant disability, hospitalization, life‐threatening symptoms, or fatality, following the ICH E2B criteria. ¹² Final outcomes were classified as recovered, recovered with sequelae, recovering, not recovered, death, or unknown. ¹² Sequelae were defined as any permanent injuries or complications that resulted from ADRs. ¹² Subgroup analyses were performed according to the type of vaccine (mRNA‐based COVID‐19 vaccines or ChAdOx1 nCoV‐19 vaccine). To reduce selection bias, patients who received BNT162b2 or mRNA‐1273 were analyzed as a combined group, as recommended by VigiBase.

Disproportionalities were analyzed by calculating the information component (IC), which is the logarithmic measure of intensity of the association between a drug and an ADR. A 95% lower end of IC (IC₀₂₅) greater than 0 was considered statistically significant. ¹³ We also determined the reporting odds ratio (ROR) values for sensitivity analyses. ¹⁴ A lower end of the ROR (ROR₀₂₅) 95% confidence interval (CI) of at least 1 was considered significant. ¹⁵ We performed two analyses with two control groups (the entire database, including all drugs and vaccines, and database restricted to influenza vaccines). Statistical analyses were conducted using R version 3.3.3 (R Foundation for Statistical Computing, Vienna, Austria) and SAS version 9.4 (SAS Institute, Cary, North Carolina). Qualitative and quantitative variables are presented as frequency (%) and median (interquartile range), retrospectively. We used the Kruskal‐Wallis test followed by the Mann‐Whitney U test to compare the time to onset and the outcomes of NA between each vaccine type. Bonferroni correction used for multiple comparisons. The calculated P value was multiplied by 3 (the number of comparisons) and displayed in the form of an adjusted P value. An adjusted P ≤ 0.05 was considered significant. Chi‐square tests were used to calculate odds ratio for residual neurological sequelae (recovered with sequelae, recovering, not recovered, and death at last follow‐up) of NA between vaccines. Cases with missing outcomes data were excluded from the analyses. ¹² Our study was approved by the institutional review board of Ewha Woman's University Seoul Hospital (EUMC‐2021‐08‐021).

3. RESULTS

On October 15, 2021, from among 1 731 147 ADR reports related to COVID‐19 vaccines, we identified 335 (0.02%) NA cases (including 219 NA cases, 93 radiculitis brachial cases, and 23 brachial plexopathy cases) reported as ADRs of the BNT162b2 (n = 197), mRNA‐1273 (n = 78), and ChAdOx1 nCoV‐19 (n = 60) vaccines. The clinical characteristics of COVID‐19 vaccine–related NA are presented in Table 1. For all types of COVID‐19 vaccines, the most common age group in which NA occurred was 45 to 64 years. There was no significant difference in the time to onset of NA among the ChAdOx1 nCoV‐19, BNT162b2, and mRNA‐1273 vaccines. The outcomes of each vaccine group are presented in Table 2.

TABLE 1.

Characteristics of COVID‐19 vaccine–related neuralgic amyotrophy cases reported in the pharmacovigilance database of the World Health Organization

	Neuralgic amyotrophy (N = 335)
	ChAdOx1 nCoV‐19 (n = 60)	BNT162b2 (n = 197)	mRNA‐1273 (n = 78)
Age, years
≤11	0 (0.0%)	0 (0.0%)	0 (0.0%)
12‐17	0 (0.0%)	0 (0.0%)	0 (0.0%)
18‐44	11 (18.3%)	63 (32.0%)	21 (26.9%)
45‐64	41 (68.3%)	70 (35.5%)	34 (43.6%)
65‐74	3 (5.0%)	14 (7.1%)	19 (24.4%)
≥75	2 (3.3%)	11 (5.6%)	2 (2.6%)
Unknown	3 (5.0%)	39 (19.8%)	2 (2.6%)
Sex
Male	28 (46.7%)	83 (42.1%)	34 (43.6%)
Female	31 (51.7%)	114 (57.9%)	44 (56.4%)
Unknown	1 (1.7%)	0 (0.0%)	0 (0.0%)
Continent
Africa	0 (0.0%)	0 (0.0%)	0 (0.0%)
Americas	2 (3.3%)	104 (52.8%)	56 (71.8%)
Asia	0 (0.0%)	0 (0.0%)	0 (0.0%)
Europe	58 (96.7%)	92 (46.7%)	22 (28.2%)
Oceania	0.0 (0.0%)	1 (0.5%)	0 (0.0%)
Seriousness
Yes	49 (81.7%)	98 (49.7%)	31 (39.7%)
No	11 (18.3%)	99 (50.3%)	47 (60.3%)
Time to onset, days	15.5 [4.0‐28.8]	10 [4‐15]	10 [3‐22]
Outcome
Recovered	2 (3.3%)	82 (41.6%)	42 (53.8%)
Recovered with sequelae	1 (1.7%)	2 (1.0%)	0 (0.0%)
Recovering	13 (21.7%)	56 (28.4%)	24 (30.8%)
Not recovered	36 (60.0%)	49 (24.9%)	10 (12.8%)
Death	0 (0.0%)	0 (0.0%)	1 (1.3%)
Unknown	8 (13.3%)	8 (4.1%)	1 (1.3%)

Note: Data expressed as number (%) or median [interquartile range].

TABLE 2.

Patients with neuralgic amyotrophy who had sequelae after COVID‐19 vaccination

	ChAdOx1 nCoV‐19,	BNT162b2,	mRNA‐1273,		ChAdOx1 nCoV‐19 vs BNT162b2		ChAdOx1 nCoV‐19 vs mRNA‐1273		BNT162b2 vs mRNA‐1273
	n (%)	n (%)	n (%)	P value	Odds ratio (95% CI)	Adjusted P value	Odds ratio (95% CI)	Adjusted P value	Odds ratio (95% CI)	Adjusted P value
Sequelae (n = 318) a	37 (71.15%)	51 (26.98%)	11 (14.29%)	<.0001	29.75 (7.81‐128.00)	<0.0001	70.64 (15.77‐312.60)	<0.0001	2.38 (1.15‐5.20)	0.13

Abbreviation: CI, confidence interval.

^a Patients with unreported outcome of neuralgic amyotrophy were not included in the analysis. Total analyzed adverse drug reaction reports of neuralgic amyotrophy included 318 patients (ChAdOx1 nCoV‐19, n = 52; BNT162b2, n = 189; mRNA‐1273, n = 77).

When compared with the entire database, a significant signal of disproportionality of NA was noted overall and for each type of COVID‐19 vaccine (overall: IC₀₂₅ = 1.46, ROR₀₂₅ = 3.20; mRNA‐based COVID‐19 vaccines: IC₀₂₅ = 1.74, ROR₀₂₅ = 3.82; ChAdOx1 nCoV‐19 vaccine: IC₀₂₅ = 0.33, ROR₀₂₅ = 1.30). However, when compared with the database with influenza vaccines alone, none of the COVID‐19 vaccines showed any significant signal of disproportionality (Figure 1).

FIGURE 1.

Forest plot with the odds ratio and information component values of neuralgic amyotrophy related to mRNA‐based (BNT162b2, mRNA‐1273) and ChAdOx1 nCoV‐19 vaccines compared with the entire database and influenza vaccines in VigiBase.

4. DISCUSSION

We found that NA showed some association with mRNA‐based COVID‐19 vaccines and the ChAdOx1 nCoV‐19 vaccine when compared with entire database; however, the signal of disproportionality did not surpass that of the influenza vaccines. NA appeared to be more strongly associated with mRNA‐based COVID‐19 vaccines than the ChAdOx1 nCoV‐19 vaccine (yet ChAdOx1 nCoV‐19 vaccination led to more frequent residual deficits) in our study. A stronger association of NA with mRNA‐based COVID‐19 vaccines was speculated in previous case reports.3, 4 The cause of this outcome is unclear. Type I interferon responses elicited by mRNA vaccines may play a significant role in NA development. ¹⁶ Post‐vaccination NA appeared around 2 weeks after first‐dose vaccination in our study. The time interval between vaccination and NA coincides with the well‐known point at which the immune response is maximized after infection or vaccination and suggests a possible association between NA and COVID‐19 vaccination.17, 18

Our study has several limitations. First, VigiBase is a passive reporting database open to physicians and the lay population, and thus may be subject to reporting bias. Second, VigiBase does not offer radiological/laboratory information and does not record the number of vaccinations per patient. Third, ADR in countries where reports are not linked to VigiBase may be omitted. However, VigiBase is composed of the data sent from national drug‐monitoring centers after their primary review. Disproportionality analysis using VigiBase permits rapid assessment of whether clinical ADRs are associated with a drug, based on large‐scale ADR reports from more than 130 countries. The advantage of these methodological approaches was well established in previous studies.6, 19, 20 VigiBase itself evaluates whether drug‐related ADRs are reported more frequently with one drug than with another, but not does confirm causality. Therefore, these results remain to be confirmed through in a future prospective epidemiological study.

The reporting of NA within 1 month of COVID‐19 vaccination suggests that clinicians and epidemiologists should continue to pay attention to the possibility of NA outbreaks associated with the COVID‐19 vaccine. Even if there is a relationship between COVID‐19 vaccination and NA, the risk of NA occurrence after COVID‐19 vacations may be low and similar to that seen with influenza vaccination.

ETHICAL PUBLICATION STATEMENT

We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

AUTHOR CONTRIBUTIONS

Concept and design: Drs Kim, Park, Min, Hong, and Song; acquisition, analysis, or interpretation of data: Drs Kim, Park, Min, Hong, and Song; drafting of the manuscript: Drs Kim, Park, Min, Hong, and Song; critical revision of the manuscript for important intellectual content: Drs Kim, Park, Hong, and Song; statistical analysis: Dr Song; administrative, technical, or material support: Dr Song; supervision: Dr Song. Dr Song has full access to all data in the study and takes responsibility for the integrity of the data and the accuracy of data analysis.

CONFLICT OF INTEREST

The authors declare not potential conflicts of interest.

Kim J‐E, Park J, Min YG, Hong Y‐H, Song T‐J. Associations of neuralgic amyotrophy with COVID‐19 vaccination: Disproportionality analysis using the World Health Organization pharmacovigilance database. Muscle & Nerve. 2022;66(6):766‐770. doi: 10.1002/mus.27734

DATA AVAILABILITY STATEMENT

Anonymized data used in this study are publicly shared by the Uppsala Monitoring Centre or the World Health Organization and can be obtained from VigiBase.