Shoulder injury related to vaccine administration following SARS-CoV-2 inoculation: Case series and review of literature

Abstract

Introduction

Shoulder Injuries Related to Vaccine Administration (SIRVA), describes those cases of shoulder severe post-inoculation complications, including pain and prolonged disability. Most of the reported cases have been secondary to influenza vaccination. This study retrospectively describes a series of 18 patients following SARS-CoV-2 inoculation and compares the findings with those previously reported for other vaccines.

Materials and methods

Inclusion criteria was onset of symptoms within 48 h after injection, symptoms duration of at least seven days, and restricted range of motion in absence of symptoms prior to vaccination. Average age was 59.4 years old (38–76), and 72.2% were women.

Results

In many cases (58%) the initial diagnosis was not clear, which lead to incorrect treatment. The most common pathological finding was subacromial-subdeltoid bursitis (66.6%). All patients who received depot corticosteroids followed by a gentle rehabilitation program showed strong clinical improvement but did not completely resolve the symptoms at 7.2 months average final follow-up. Surgical intervention was necessary in one of the patients due to the persistence of symptoms despite conservative treatment.

Conclusions

Shoulder injury related to vaccine administration is rare, but when present, its torpid evolution makes it difficult to treat. We have found in our case series a similar pattern to that already described for other vaccines. A high index of suspicion helps to pick up the condition promptly and early treatment can bring satisfactory outcome.

Highlights of the study

• •Shoulder injury related to vaccine administration has a torpid evolution.
• •Local pain and loss of active range of motion was present in all cases.
• •A correct approach should include early diagnosis and ruling out infection or neurological injury.
• •Initial treatment with depot corticosteroids and a gentle rehabilitation program are the first treatment choice.

1. Introduction

The outbreak of the Covid SARS 2 pandemic has forced a massive, worldwide vaccination campaign. The chosen inoculation site is the deltoid muscle region¹ because it is an area of easy exposure, it generates an adequate antigen-antibody reaction,² and does not create a walking impairment. There are reports of 60%–80% minor and transient adverse effects after vaccine application in the deltoid muscle region.³ However, there is a small percentage of cases of severe and persistent local symptoms, including sharp pain, loss of active and passive mobility, and functional impairment. These and other complications were first reported by Atanasoff et al.⁴ in a series of 13 cases with pain and disability following Influenza and tetanus-containing vaccination in the deltoid region. This situation was named Shoulder Injuries Related to Vaccine Administration (SIRVA).⁴ Isolated cases of SIRVA have been reported after SARS-CoV-2 vaccination5, 6, 7, 8, 9, 10 but, according to our literature search, there is not a profound analysis of a case series published yet. The possible underreporting of local effects has been mentioned.¹¹

The aim of this study is to describe the findings in a series of cases with local complications following SARS-CoV-2 vaccination, compare them with those previously reported for other vaccines, and share our therapeutic experience.

2. Materials and Methods

We included patients referred from seven institutions from July 2021 to November 2021. Cases should meet the following criteria: onset of symptoms within 48 h after injection, symptoms duration of at least seven days, restricted range of motion, and absence of shoulder symptoms prior to vaccination. Exclusion criteria were cases in which all inclusion criteria were not met, ongoing shoulder pain due to active shoulder pathology prior to vaccination, rapid response to conservative treatment, and a history of psychiatric disorders.

Our clinical evaluation was performed by an orthopedic surgeon specialized in shoulder surgery. Data was extracted from patients' clinical records and stored in a spreadsheet that was available to the investigators. Each patient was assigned an encrypted alphanumeric code. We evaluated age, gender, affected side, commercial brand of COVID-19 vaccine applied, dose at which symptoms occurred, symptoms (pain, dysfunction, swelling), onset and duration of symptoms, indicated treatment, response to treatment and follow-up time. Pain was subjectively graded as mild, moderate, or severe by the patients. Functional evaluation included active and passive range of motion (ROM), and Simple Shoulder Test (SST) score. Bone and soft-tissue evaluation were performed by radiographs and ultrasound or magnetic resonance imaging (MRI) in all cases. Notification of SIRVA cases to competent authorities was mandatory. Ethics approval was waived by an Institutional Review Board (Registration Code 6197).

3. Results

Eighteen cases met the inclusion criteria. Demographics and clinical findings are shown in.

Table 1. All included individuals were adults since minors had not yet been eligible for vaccination at the time of inclusion. Average age was 59.4 years old (38–76), 72.2% were women. Vaccination was performed on the left shoulder in ten cases, and in eight on the right. Nine patients were vaccinated with Sputnik V™ vaccine (Gamaleya Research Institute), seven with Covishield™ (AstraZeneca, University of Oxford), one with BioNtech™ (Fosun Pharma, Pfizer), and one with Sinopharm BBIBP™ (Beijing institute of Biological Products, Sinopharm).

Table 1.

Clinical findings.

	Clinical Findings
Age	59.4 years (range 38–76)
Gender	Female: 13 (72.2%)			Male: 5 (27.8%)
Shoulder side	Left: 10 (55.6%)			Right: 8 (44.4%)
Dose prior to symptoms	First dose: 14 cases (77.8%)			Second dose: 4 cases (22.2%)
Vaccine	Sputnik V (9)	Covishield (7)		Sinopharm (1)	BioNTech (1)
Onset of pain	Immediate (<2 h): 8 cases (44.4%)			>2 h: 10 cases (55.6%)
Signs and symptoms	Pain 15 (83.3%)		Limited active ROM: 16 (88.8%)		Limited passive ROM: 6 (33.3%)
Subjective degree of pain at initial evaluation	Mild: 0		Moderate: 11 (61.1%)		Severe: 7 (38.8%)
SST score average	Initial: 3.4			At final follow-up: 8.5
Diagnostic tests	MRI: 16 Ultrasound: 1 Both:1
Treatment	Depot corticosteroids: 11			Physioterapy: 15

First evaluation was done by our team between 2 and 20 weeks after inoculation (average: ten weeks). Symptoms began after first dose in fourteen cases, and after the second one in four. In the latter cases, none have had complications after first dose. Two patients who presented symptoms after first application refused to receive the second dose.

Local pain and loss of active ROM was present in all cases (Fig. 1). Moderate pain was reported in eleven cases (61.1%), and severe pain in seven (38.8%). Onset of pain was reported as occurring immediately (within 2 h) after application in eight patients (44.4%), within the first 24 h but not immediately in seven (38.8%), and between 24 and 48 h in three (16.6%). Even though all of the patients suffered some degree of limitation of their range of motion it was moderate in most of the cases. Just one patient presented with pseudo-paralysis. Persistent swelling and tenderness were found in two cases, a deep hematoma and axillary nerve neuropraxia was found in one case. A 70-year-old female patient developed a neuralgic amyotrophy (Parsonage Turner syndrome) 2-days after the first dose of CoV-19 vaccine, which was given intramuscularly in the right deltoid, suffering severe pain, mild wasting of the deltoid with hypotonia and hypoesthesia. Power of deltoid and supraspinatus was Medical Research Council Grade 0/V. Nerve conduction study (NCS) showed reduced compound muscle action potential amplitudes of suprascapular nerve and axillary nerve.

Fig. 1.

Marked loss of mobility in a patient suffering SIRVA.

Four patients complained of traumatic inoculations and painful injections which produced severe symptoms shortly after. Of these, thanks to photographic material it was possible to verify a “too high” injection site of the vaccine (Fig. 2).

Fig. 2.

a) Patient pointing at a high injection site. b) Oblique - coronal MRI image of the same patient.

shows subacromial bursitis and an area of bone edema in the proximal epiphysis.

Average SST score at first evaluation was 3.4 (0–7). The final evaluation of the patients was conducted between 5 and 10 months with an average of 7.2 months. At last follow-up, the average SST score was 8.5 (1–12). Thirteen patients (72.2%) persisted with pain and some degree of loss of ROM at final follow-up. Surgical intervention was necessary in one of the patients due to the persistence of symptoms despite conservative treatment. Arthroscopic subacromial bursectomy was performed due to the presence of chronic painful bursitis which was confirmed by pathological anatomy (Fig. 3).

Fig. 3.

Signs of adhesive capsulitis on MRI sagittal plane.

Imaging findings are shown in Table 2. Radiographic images did not provide any useful information, but they allowed us to rule out calcific tendinitis and osteoarthritis. Fifteen patients underwent MRI, one ultrasonography, and one, both. Only one case presented with normal MRI. The most common pathological finding was subacromial-subdeltoid bursitis, which occurred in twelve cases (66.6%), four of whom presented other associated injuries. Four partial and two full-thickness rotator-cuff tears were found (33.3%), but due to the high prevalence of these injuries in this group's average age, we were unable to define whether they were caused by vaccination. One case (5.5%) presented with biceps tenosynovitis and another (5.5%) with bone edema. A very particular finding among SIRVA patients is the presence of fluid collection within the fibers of the deltoid muscle (Fig. 4). This was found in three cases (16.6%).

Table 2.

Imaging findings. (In several patients there was more than one finding.)

Imaging findings
Subacromial-Subdeltoid bursitis	12 cases (66.6%)
Deltoid fluid collections	3 cases (16.6%)
Partial Rotator cuff tear	4 cases (22.2%)
Full-thickness rotator cuff	2 cases (11.1%)
Biceps tenosynovitis	1 case (5.5%)
Bone edema	1 case (5.5%)
Adhesive capsulitis	1 case (5.5%)
Neuralgic amyotrophy	1 case (5.5%)

Fig. 4.

Deltoid effusion on MRI coronal plane.

Clinicians or orthopedic surgeons of various specialties conducted initial treatment, prior to consulting a shoulder surgeon from our participating institutions.

Fifty eight percent of the cases were initially misdiagnosed, which lead to an incorrect treatment. One of the patients referred was initially diagnosed with Parsonage Turner Syndrome and was prescribed Pregabalin for three months without clinical response. Studies (brachial plexus MRI, electromyogram, and nerve conduction) ruled out neurological disorders, finally the correct diagnosis was a subacromial bursitis. Another case with a clinical suspicion of infection was treated with antibiotics for 2 weeks without any previous assessment. Once in our care, we ruled out infection and diagnose a severe subacromial bursitis.

The average time undergoing treatment until a shoulder surgeon took over varied from 1 to 20 weeks (average 10 weeks). None of the patients had been previously treated with depot corticosteroids.

Once under our treatment, eleven patients received intramuscular depot corticosteroids; three of whom received a second dose later. The remaining cases did not received corticosteroids for several reasons: it was not indicated by the treating surgeon in four cases, one patient was under suspicion of an ongoing shoulder infection, which would later be ruled out, and another one refused to receive the medication.

A rehabilitation program including passive and assisted active-motion exercises was indicated in fifteen patients. Five cases had been treated with rehabilitation with poor results due to increased pain related to the type of physiotherapy they received. Once they were switched to the appropriate plan and under the effect of corticosteroids, they had a good evolution.

One patient refused to follow medical treatment but resolved symptoms spontaneously after five months.

Average follow-up time was 7.2 months (range, 5–10 months). All patients who received depot corticosteroids showed strong clinical improvement but did not completely resolve their symptoms. All patients showed slow but progressive recovery rates at final follow-up. The combination of depot corticosteroids and a gentle exercise program determined better results as shown by the comparison of SST before and after treatment.

4. Discussion

Bodor and Montalvo¹² were the first to describe vaccination-related shoulder dysfunction in 2007. In 2010 the term SIRVA was described by Atanasoff et al.⁴

SIRVA is a clearly defined and known condition,^4,13, 14, 15, 16, 17, 18, 19 but its incidence is currently unknown for the SARS-CoV-2 vaccines. There is, however, information from previous vaccines. In 2018, there were 10,230 reported application-related complications of varying severity of the influenza vaccine, representing only 0.006% of the target population. In this group, only 27 patients (0.000017%) described self-limited shoulder pain or synovitis.²⁰ In our series, it is impossible to know what proportion of the total vaccinated patients suffered from SIRVA, since we are not a referral or reporting center for this pathology. We estimate that the incidence is low but that there is also a high rate of underreporting, as described by Menni et al.¹¹ Cagle et al.,²¹ identified 56 reported accounts of SIRVA worldwide in 2021, whatever the disease for which the vaccine was applied for.

Our findings demonstrate a few key points. In the first place, due to the low prevalence and non-specific clinical aspects of this condition, many patients who suffered from SIRVA were initially misdiagnosed and took an average of 10 weeks to consult a shoulder surgery specialist. Secondly, pain and loss of ROM, and consequently low SST scores were found in all patients. Lastly, a rehabilitation program and the systemic application of depot corticosteroids showed positive results, although a complete recovery was not a common finding.

Comparing our study to Atanasoff et al.'s⁴ published in 2010 on various vaccines, several similarities and differences can be found. Average age was higher in our series (59.4 vs. 50 years), but it most likely was due to Covid vaccination campaigns starting at older populations. Female sex was predominant in both series (70.5% and 85%), which may be explained by women's smaller deltoid muscle mass,⁴ being the subacromial space and axillary nerve at risk from chemical and traumatic injury.

Pain was present in all cases of the two series. In our group, immediate onset of pain occurred in lower proportions (44.4% vs. 55.6%). Severe pain was reported in 33.3% of our cases. Loss of ROM was the second most frequent finding on both case series (70.5% vs. 88.8%).

In two cases in our series, patients were psychologically affected by this complication, and untrusty of the vaccine. They refused to take the second dose even though they understood its benefits. We believe it is important to provide psychological assistance as they go through this prolonged complication.

In Atanasoff's study⁴ a query of the Vaccine Injury Compensation Program database was conducted to identify claims for injuries related to vaccine. There was a varied origin of the cases that were presented to make administrative claims. Our series just included a group of patients assisted by shoulder and elbow specialists who have been in contact with each other to attempt to resolve this condition, without the presence of any administrative claims.

When analyzing MRI scans, subdeltoid-subacromial bursitis was the most frequent finding (66.6%), followed by deltoid effusion (16.6%). Diverse types of rotator-cuff tears were found in six patients (33.3%), but due to the high prevalence of these injuries in general population at this age group, it is quite difficult to define whether they were related to vaccine administration.

In many cases, shoulder injury is generated by an inadequate application technique.3, 4, 5, 6^,22, 23, 24^,27 Safe inoculation area, appropriate needle length based on patient characteristics, and other technical aspects have been described in detail.^14,15,25,26 The technique can fail due to the vaccine being applied very high in the arm, which causes compromise of the subacromial-subdeltoid bursa, or very low, with possible neurological compromise, or very deep, with aggression on the periosteum or the joint, or very shallow, depositing the vaccine in the subcutaneous cellular tissue that is not sufficiently vascularized to facilitate an adequate immune response and is associated with an increase in local adverse reactions.²⁸

Beyond this, the injury caused by vaccination is usually greater than what would be expected from a simple trauma with a needle.^4,7 When a vaccine is injected into the muscle belly, it creates an antigen-antibody reaction at the muscular level that causes transient and moderate discomfort. When the vaccine is applied into a synovial space such as the shoulder joint, preexisting antibodies from previous infections or vaccines can lead to a prolonged inflammatory reaction.^4,23,27 Although COVID-19 is a new disease and the population is receiving vaccines for the first time, it is unlikely that an adult has not been exposed to adenovirus in the past.

Finally, aluminum compounds in the adjuvant of some vaccines may also have contributed to severe inflammatory reactions.¹³

There is no consensus in literature on how to manage SIRVA. We consider that the first step is to rule out infection,^6,28 neurological lesions,29, 30, 31, 32 and preexisting conditions to arrive at a correct diagnosis. The early use of intramuscular depot corticosteroids has proven to be effective for most of the possible clinical presentations.^33,34 Some authors propose its local subacromial application, but we prefer to avoid further aggression to an already painful and swollen area. On the other hand, local and systemic corticosteroid application have not shown clinical differences in shoulder pathology.^35,36 A gentle exercise program should be incorporated as soon as pain is controlled. The first objective is to regain passive ROM, following by active ROM, and finally strength in a progressive fashion. Its early implementation contributes to pain control and helps prevent the development of adhesive capsulitis. In our series, we treated with surgery a case of severe subacromial bursitis and adhesive capsulitis related to poor initial management of rehabilitation. Adhesive capsulitis can be secondary to pain and poor management of rehabilitation,^24,37,38 but it can also present as an immune response. Thompson et al.²² reported a case of bilateral adhesive capsulitis following influenza vaccination, suggesting an immune-mediated reaction as its probable cause.

In SIRVA cases with severe subacromial bursitis and/or adhesive capsulitis with poor response to conservative treatment, surgical intervention has been indicated.^4,23 Remarkably, in Atanasoff's series,⁴ four cases required one or more surgical interventions. Just one of our patients underwent surgical treatment thus far.

At the time of writing this study we found no case series reports of this size regarding SIRVA injuries related to COVID vaccination. The weaknesses of this study are the fact that it is a retrospective and multicenter evaluation. We have tried to compensate for this with strict inclusion criteria that meet the SIRVA definition and with the help of complementary images in all cases. Another weakness is not having specific percentages of the incidence of the condition in the total vaccinated population.

5. Conclusions

Shoulder injury related to vaccine administration is rare, but when present, its torpid evolution makes it difficult to treat. We have found in our case series a similar pattern to that already described for other vaccines. This condition should be suspected in all individuals, without a history of shoulder pain or dysfunction, who experience sudden pain and reduced ROM within 48 h, following vaccine application, which persist for over seven days. A correct approach should include early diagnosis, ruling out any prior infection or neurological injury, evaluation of specific soft tissue injuries by imaging studies, initial treatment with depot corticosteroids, gentle rehabilitation program, and report to health authorities. It is important to inform patients of its prolonged evolution and provide psychological support when needed.

Funding/sponsorship

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Informed consent

The patients have authorized the use of their images. The Institutional Ethical Committee Approval of our hospital has not considered it necessary to sign an informed consent because it is a retrospective study.

Institutional Ethical Committee Approval: Authorization code: 6197. Proof is attached as a separate file.

Authors contribution

DM: Conceptualization- Data curation- Supervision- Writing – Original draft DG: Conceptualization-Investigation-Writing – Original draft NA: Formal analysis-Methodology-Validation-Visualization Writing – original draft Writing – review & editing FA: Resources-Validation- Writing – review & editing JMPC: Investigation-Resources-Visualization FD: Resources-Visualization- Writing – review & editing PHM: Review & editing GP: Review & editing.

Declaration of competing interest

None.