Shoulder injury following COVID-19 vaccine administration: a case series and proposed diagnostic algorithm

ABSTRACT

Background

Shoulder Injury Related to Vaccine Administration (SIRVA) is a preventable adverse event following incorrect vaccine administration, which can result in significant long-term morbidity. There has been a notable surge in reported cases of SIRVA as a rapid national population-based COVID-19 immunization program has been rolled out across Australia.

Methods

Surveillance of Adverse Events Following Vaccination in the Community (SAEFVIC) in Victoria identified 221 suspected cases of SIRVA following the commencement of the COVID-19 vaccination program, reported between February 2021 and February 2022. This review describes the clinical features and outcomes of SIRVA in this population. Additionally, a suggested diagnostic algorithm is proposed, in order to facilitate early recognition and management of SIRVA.

Results

151 cases were confirmed as SIRVA, with 49.0% having received vaccines at state vaccination centers. 75.5% were suspected incorrect administration site, with most patients experiencing shoulder pain and restricted movement within 24 hours of vaccination, lasting on average 3 months.

Conclusion

Improved awareness and education regarding SIRVA is imperative in a pandemic vaccine roll-out. The development of a structured framework for evaluating and managing suspected SIRVA will aid in timely diagnosis and treatment, essential to mitigate potential long-term complications.

KEYWORDS:

• COVID-19 vaccine
• SIRVA
• shoulder injury
• vaccine adverse event
• vaccine complication

1. Introduction

The COVID-19 pandemic has resulted in Australia implementing its largest ever population level vaccine rollout throughout 2021–22, with the first vaccines administered on 22 February 2021. The mass immunization program has resulted in the deployment of a rapidly expanded vaccination workforce comprising a broad range of immunization experience including hospital staff, pharmacists, general practitioners (GP) and general practice nurses administering vaccines across a variety of settings[1].

With such a rapid vaccine rollout and scaling of vaccination workforce, there is an associated risk of vaccine administration error, as well as the potential lack of identification and recognition of these adverse events following immunization (AEFI). Likewise, the use of masks and other personal protective equipment utilized during the pandemic, had a cumbersome effect on communication that could have negatively affected interactions between patients and vaccinators[2]. Both these scenarios may have significant impacts ranging from suboptimal clinical outcomes to development of vaccine hesitancy amongst those affected [3,4].

One such significant AEFI is shoulder injury related to vaccine administration (SIRVA). Whilst transient shoulder pain at the local injection site is common following administration of a vaccine, SIRVA is a rare and under-reported side effect that can result in debilitating disease and even permanent shoulder dysfunction [5,6]. SIRVA can occur when vaccines are delivered into the sub-deltoid bursa or shoulder joint space instead of the deltoid muscle[7]. Most commonly, this can be a consequence of incorrect technique, whereby vaccines are unintentionally delivered too high into the shoulder joint [8,9].

The pathogenesis of SIRVA remains uncertain. It is postulated that incorrect vaccination causes mechanical and/or immune mediated inflammatory reactions that damage surrounding structures including bursa, tendons, and ligaments[10]. Symptoms can include rapid onset of pain at time of injection, decreased range of motion and pain on movement of the arm. Patients with SIRVA are often diagnosed with alternative causes, with the differential diagnosis including deltoid muscle bursitis, tendonitis, rotator cuff tear, frozen shoulder or adhesive capsulitis [4,5,9].

SIRVA can result in both acute and long-term impacts on biological, psychological, and social health. Whilst it is becoming increasingly prevalent, highlighted most recently in population wide vaccine rollouts, it remains under-recognized and under-reported.

This is likely because SIRVA can pose a diagnostic dilemma – as it can be difficult to distinguish from normal injection site pain (often in the shoulder region), that may occur after the first 48 hours post immunization. This emphasizes the need of a standardized criterion that will enable immunizers and practitioners to evaluate suspected SIRVA and proceed to diagnosis and management as early as possible. To date, an established case definition or management guideline for SIRVA has yet to be developed.

Given the current paucity of systematic data on SIRVA related to COVID-19 vaccination, this study aims to describe the clinical features and SIRVA-related outcomes of a single population-level cohort. Secondary outcomes of this study include proposing diagnostic considerations to assist clinicians in recognizing signs and symptoms of SIRVA and an algorithm to facilitate early investigation and management.

2. Methods

Surveillance of Adverse Events Following Vaccination in the Community (SAEFVIC) is the Victoria, Australia, state-wide vaccine safety surveillance system for all adverse events following immunization (AEFI)[11]. It has been operational since 2007 and comprises both active and spontaneous surveillance mechanisms [12,13].

3. Ethics

Follow-up of cases was undertaken as part of public health AEFI management. The SAEFVIC database is a clinical quality registry that forms part of Victoria’s vaccine safety surveillance program.

4. Data Collection

Self and provider reported AEFI after any COVID-19 vaccination to SAEFVIC in the first year of the COVID-19 vaccination program (22 February 2021 to 22 February 2022) were reviewed. Reports containing any of the terms ‘SIRVA, shoulder injury, shoulder pain, frozen shoulder, prolonged shoulder pain, impingement syndrome, adhesive capsulitis, restricted movement and/or bursitis’ were included for initial analysis.

This was correlated with a clinical surveillance phone call at least 6 months after vaccination, as part of standard AEFI follow-up, to further corroborate any information required for suspected diagnosis including clinical symptoms and features.

Data collected included demographic information, vaccinee medical history and clinical features. Post-vaccination data such as imaging or diagnostic tests, management options, revaccination decisions were gathered. The occurrence of long-term (6 months) morbidity and clinical outcomes were also assessed, including whether the patient incurred significant financial costs as part of their treatment.

Based on available information, 151 cases were included for analysis, with inclusion and exclusion criteria shown in Figure 1.

Figure 1. Diagnostic criteria for SIRVA.

5. Data Analysis

Data are presented as counts and proportions. To assess differences in time to symptom onset and time to symptom resolution between sexes, as well as age groups, Kruskal-Wallis H test was applied. Mann-Whitney U-test was used to compare time to symptom resolution between patients with rapid and those with delayed onset of pain. A p-value of <0.05 was considered statistically significant. The statistical programming language R was used throughout this project.

6. Results

Of the 151 cases analyzed, the median age of patients was 49 years old (IQR 42–57) (Figure 2). There was a female predominance (68.2%). The median time for pain onset was double in males (12 hours, IQR 47) compared to females (6 hours, IQR 23). The median time until pain onset was 8 hours (IQR 23.4) for patients 18–49 years old, 6 hours (IQR 3) for patients 50–64 years old and 24 hours (IQR 65) for patients above 64 years of age. No cases were reported in children under the age of 18 years.

Figure 2. Suspected SIRVA cases: Patient Demographics (Age).

6.1. Clinical Features

Table 1 demonstrates that 112 (74.2%) of suspected SIRVA cases self-reported shoulder area pain within 24 hours, with a further 13 (8.6%) patients experiencing pain within 48 hours. The majority of patients reported restricted movement (80.1%) and pain on movement of the shoulder (97.3%). 75.5% of patients also self-reported that they were suspicious the vaccine was administered incorrectly into the wrong anatomical site, most often describing that the injection was suspected to be given too high on the arm.

Table 1. Clinical features of SIRVA cases.

Onset of pain			Medical follow-up*			Imaging investigations			Diagnosis on imaging			Length of time to symptom resolution		Medication**
	N	%		N	%		N	%		N	%	Age group	Median no. of days		N	%
Immediate/rapid	50	33.1%	GP	123	81.5%	X-ray	24	15.9%	Bursitis	56	68.3%	18–49 years	35 days	OTC	122	80.8%
Within 24hrs	112	74.2%	Allied health	75	49.7%	Ultrasound	91	60.3%	Adhesive capsulitis	19	23.2%	50–64 years	90 days	Prescription medication	37	24.5%
Within 48hrs	125	82.8%	Specialist	16	13.9%	MRI	19	12.6%				64+ years	45 days	Cortisone injection	37	24.5%
						None	47	31.1%				No resolution at time of follow-up	80 (53.0%)

(MRI: multi-resonant Imaging, OTC: Over the counter medication)

** Figures include patients who may have had multiple different types of medical follow-up or taken multiple different medications.

Most patients had imaging investigations (66.2%), with ultrasound the most common modality. Other imaging modalities included X-ray and magnetic resonance imaging (MRI). Of those who received ultrasound imaging, 82.4% had abnormalities detected, with bursitis the most common diagnostic finding. Other diagnoses included adhesive capsulitis and supraspinatus tendinosis.

Most cases sought medical attention for their symptoms, with 123 (81.5%) patients having presented to their GP post vaccination. Just over half of the patients in our study received treatment with allied health practitioners including physiotherapists (Table 1). A significant proportion of patients reported the use of adjunctive medications to help manage their symptoms, including simple analgesia such as paracetamol and/or ibuprofen as well as stronger prescription-only analgesia. Nearly a quarter of patients (24.5%) also received direct joint corticosteroid injections.

The median duration until symptom resolution was 51.5 days. This was longer in males (median = 74.5 days, IQR 33.5 to 140.8) and those aged 50–64 years (median = 90 days). Patients who reported rapid onset of pain were found to have a significantly longer recovery time than those who did not report their onset of pain as rapid (81 vs 44 days, p < 0.01).

Our study demonstrated that SIRVA has the potential to negatively impact patient’s sleep, work, exercise, driving and ability to perform activities of daily living (ADL) (Figure 3). Even though 18.5% of patients reported incurring significant loss of income defined as monetary loss of >$2,000 AUD (approx. $1270 USD), which was above the threshold set by the Australian Government COVID-19 vaccine claims scheme [14,15], only eight patients (5.3%) patients actually participated in the claims scheme. Over 74% of patients went on to receive further vaccinations, including scheduled immunizations, influenza and COVID-19 vaccinations or boosters.

Figure 3. Negative impact on day-to-day functioning.

The most common locations for COVID-19 vaccine administration in reported cases of suspected SIRVA were government vaccine hubs (n = 74, 49.0%), followed by general practices (n = 42, 27.8%). Eleven (7.5%) of cases reported a pharmacy as the place of vaccine administration.

The type of immunization provider across all cases was similarly distributed, with 47.7% of cases receiving vaccines administered by hub vaccinators, 15.9% by GP practice nurses, 11.3% by GPs and 6.6% by pharmacists. Approximately 20% of patients were not aware of the type of immunizer administering their vaccine. Suspected SIRVA was most commonly self-reported (62.9%), with 13.2% reported by GP’s and 5.3% by GP practice nurses.

7. Discussion

7.1. Background

SIRVA was first described by Atanasoff et al as a secondary phenomenon following various vaccines more than 10 years ago[5]. It has been characterized by protracted shoulder pain and restricted movement that begins within the first 48 hours post vaccination [4,5]. SIRVA has since been highlighted via isolated case reports across various jurisdictions, albeit with a paucity of a standardized definition and management algorithm.

Global surges in COVID-19 vaccine administration likely attributed to recent rises in reported cases of SIRVA. For example, in Australia alone, more than 94% of the population >16 years were fully vaccinated with a primary course in just over 12 months[1]. This increase became evident from April 2021 with increasing case reports of SIRVA after COVID-19 vaccination, followed by Bass et al describing a large cohort from the US-based Vaccine Adverse Event Reporting system (VAERS) [4,16].

7.2. Demographics

Our cohort was analogous to VAERS in demographic distribution, with a similar median age (51.8 years VAERS vs 49 years SAEFVIC) and sex distribution (female prevalence 67.1% VAERS vs 68.2% SAEFVIC). Likewise, pain and limited range of motion were the most commonly reported symptoms across both cohorts[4]. There were no pediatric cases reported and the overall incidence in this age group remains unknown. This may be credited to the COVID-19 vaccine only becoming available to Australian children <12 years old from 10 January 2022, and the use of the anterolateral thigh as an alternate injection site in infants [17,18].

Our study highlighted a significant difference in the time of symptom onset between genders, with males reporting a median time double that of females (6 hours compared to 12 hours respectively). Furthermore, patients aged over 64 years of age had a median time of symptom onset up 3–4 times longer than younger patients. There is currently insufficient data to comment on the causality of these differences in reported timing of pain onset.

The age distribution of our results and notable female predominance of patients, despite the vaccine program being population wide, raises the question of potential autoimmune or anatomical differences that may predispose certain subgroups to developing SIRVA[19]. Other risk factors for SIRVA have included low body mass index and a small deltoid muscle[7]. This highlights an important area of future research and consideration to mitigate potential long-term morbidity.

The top locations of vaccine administration reported by patients with SIRVA (namely state vaccination centers and GP practices) correlated with the most common vaccine administration sites for COVID-19 vaccines in the general population. Despite the majority of vaccines being administered at state vaccination centers, most suspected SIRVA cases were reported by patients themselves or by general practitioners after patients presented for evaluation.

This is likely a reflection of the unique methodology of AEFI reporting during the COVID-19 vaccine rollout. Unlike other vaccinations, the COVID-19 rollout in Victoria had active as well as passive surveillance systems in place, with prompts to vaccinees to report any AEFI to SAEFVIC. Nonetheless, the discrepancies in provider reporting despite many cases seeking medical attention indicate that there is underreporting of SIRVA cases. Further emphasis on improving recognition and reporting for both consumers and practitioners is paramount to be able to evaluate the true size of this AEFI.

7.3. Identification and modifiable factors

Improved recognition and diagnosis of SIRVA as an AEFI, as well as addressing modifiable factors is paramount in the current climate of increasing vaccine administration. We have previously reported on our experience with SIRVA pre-COVID-19 vaccination across a 13-year period [20]. The findings have helped us to develop proposed SIRVA diagnostic criteria (Figure 1), which have been utilized in this study to standardize evaluation of reported cases. These criteria, along with other characteristics that add to the clinical suspicion of SIRVA such as abnormalities on medical imaging (including deltoid muscle bursitis, tendonitis, rotator cuff tear, or adhesive capsulitis) or poor response to over-the-counter analgesia, may help to increase awareness and facilitate earlier detection and diagnosis.

Improved provider awareness of SIRVA may aid in addressing poor injection technique, which is the largest modifiable factor that may lead to SIRVA. Mackenzie et al highlighted poor health practitioner awareness of SIRVA in a recent cross-sectional study. They also identified that only 52% of authorized vaccinators accurately identified correct anatomical landmarks integral to correct immunization technique[21]. A significant number of patients in our study reported suspicion that their vaccine was administered incorrectly. This may have been an incorrect position (such as too high on the arm), or potentially an incorrect direction of the needle causing the vaccine to be injected into the bursa rather than the deltoid muscle[8].

As such, awareness and increasing attention to surface landmarking on each patient’s differing anatomy is critical. At risk subgroups as described above may benefit from the use of different needle types, considered injection techniques and risk counseling[22]. Key recommendations for correct administration of injected vaccines have been summarized and are available for public reference via the Melbourne Vaccine Education Center (MVEC)[18].

7.4. Investigations & management

There is currently no gold standard approach to the investigation and management of SIRVA. In addition to the aforementioned diagnostic criteria, we propose a structured framework for evaluating and managing suspected SIRVA to aid in early recognition (Figure 4).

Figure 4. Proposed SIRVA Investigation & Management Algorithm.

The use of imaging tests to aid in diagnosis of SIRVA is important. Whilst x-ray is a commonly used first-line investigation, ultrasound (where available), may be more effective in highlighting connective tissue and joint capsule injury, which may then precipitate specific treatment options[4].

Treatment options center around two main goals: symptomatic relief with pharmacological modalities, coupled with non-pharmacological therapies designed to improve and restore motion and function. Pharmacological options include simple analgesia such as non-steroidal anti-inflammatories, along with intra-articular corticosteroid injections, both of which have been shown to improve recovery if initiated early [3,8]. Oral corticosteroids have been trialed for some patients, although there is a paucity of evidence that this reduces symptoms or decreases recovery time. [8]

A large proportion of patients sought allied health treatment, namely physiotherapy, to improve range of motion and prevent complications such as frozen shoulder. Early treatments such as these, that form part of a management regime have been shown to improve recovery[23]. Whilst a limited number of patients in our cohort had surgery, other groups have found that surgical repair occurred in up to 32.6% of cases[9].

Despite investigations and treatment, many patients describe ongoing symptoms more than 6 months post vaccination (Table 1). Longer recovery time was more prevalent in patients who initially described rapid onset of pain. The significant morbidity extended primarily to patients’ abilities to perform activities of daily living such as dressing, cleaning, driving, washing and sleeping which require anti-gravity use of the shoulder joint. The poor participation in the Australian government COVID-19 vaccine claims scheme may have been secondary to lack of awareness, as well as difficulty meeting eligibility criteria such as need for hospital admission – the majority of SIRVA cases are managed in an ambulatory setting [1,15].

7.5. Future actions

This study highlights some of the potential vocational ramifications as well as the emotional and mental health impacts of SIRVA. These can be difficult to quantify, and further studies will help elucidate the full implications. It is to be noted that a quarter of patients, although eligible, had not gone on to receive any further vaccinations at the time of the 6-month review. An experience of an AEFI like this, whilst not strictly directly caused by the vaccine, may also contribute to future vaccine hesitancy and decreased confidence in healthcare providers[3].

SIRVA can be avoided through improved regulation, guidance and education of immunization providers. Mandatory training of vaccinators highlighting the importance of preventing SIRVA can help minimize incorrect vaccine administration. Furthermore, educational resources can help immunization providers consolidate and evaluate their skills[18]. With the advent of new rises in vaccine hesitancy and the surge in vaccinations associated with global pandemics, minimizing preventable AEFI’s such as SIRVA is becoming paramount[24].

8. Limitations

Given the nature of this study, which relies on active and passive surveillance mechanisms, there is likely under-reporting of SIRVA cases. Further awareness of SIRVA as outlined above may help to boost reporting rates. The surveillance systems also limited information provided by users in some instances, resulting in several cases being excluded from analysis due to insufficient information. Furthermore, given the retrospective nature of data collection as part of safety surveillance, no data were collected on needle length and therefore no analysis can be made on whether this contributes to the mechanism of SIRVA. Comparison between various treatment modalities and associated outcomes in this study was unable to be performed.

9. Conclusion

SIRVA is an under-recognized AEFI that has become more frequently apparent in the pandemic COVID-19 vaccine roll-out. Incorrect injection administration is the largest modifiable factor leading to SIRVA. Improving injection technique, along with identifying at-risk patients and general awareness of SIRVA may reduce the incidence of this important AEFI.

By proposing a structured framework for evaluating and managing suspected SIRVA that can be used across jurisdictions, the aim is to aid early recognition. Timely diagnosis and treatment of SIRVA will help mitigate long-term patient morbidity, vaccine hesitancy and healthcare system burden. Minimizing mistrust in healthcare systems and reducing the incidence of AEFI’s will play a vital role in any future pandemics. Further studies are required to understand the outcomes of SIRVA and better direct its treatment and work to establish a standardized case definition and best practice management guideline should be undertaken.

Article highlights

• There has been an increase in cases of SIRVA following mass COVID-19 immunization programs in Australia.

• A standardized diagnostic modality will aid in early recognition of SIRVA

• SIRVA has the potential to cause significant morbidity

• Ongoing efforts are required to formulate treatment guidelines for SIRVA

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

The SAEFVIC nursing team for their assistance and support in clinical surveillance of patients.

This work has been presented as an oral presentation at the CDIC 2022 and IPA 2023 conferences.

Additional information

Funding

This paper was not funded.