How Long Does SIRVA Last?
One of the most important questions after a suspected Shoulder Injury Related to Vaccine Administration (SIRVA) is:
How long should the symptoms last?
There is no single answer.
Some patients recover within weeks. Others require several months of treatment. A smaller subset develops persistent pain, stiffness, or functional limitation that can continue for a year or longer.
The literature can appear contradictory because different studies examine very different populations.
A prospective or registry cohort that follows all reported cases may show relatively favorable recovery.
A medicolegal claims database will disproportionately contain patients whose symptoms were sufficiently persistent or severe to pursue compensation. Case reports are even more selective because unusual, prolonged, or treatment-resistant cases are more likely to be published. Understanding these differences is essential when discussing prognosis.
Ordinary Injection Soreness Is Not SIRVA
Pain at the injection site for a short period after vaccination is common. SIRVA refers to a different clinical pattern: shoulder pain and dysfunction that is more severe and persistent than expected from routine post-injection soreness.
The original Atanasoff series that introduced the term SIRVA involved 13 patients whose shoulder pain and reduced range of motion persisted for more than six months.
That historical fact is important, but it can also create a misleading impression. The original patients were selected specifically because they had severe, prolonged symptoms. They should not be interpreted as establishing that all SIRVA lasts at least six months.
One of the Best Long-Term Studies Found a Median Recovery of 8 Weeks
Marshall and colleagues provide particularly useful prognosis data because they followed a larger clinical series longitudinally. Their Australian SAEFVIC cohort included 102 SIRVA cases. Patients were followed for at least six months, and 85 were successfully contacted for long-term follow-up. Among those 85 patients:
86% reported complete resolution of symptoms, while
14% continued to have pain and/or restricted range of motion.
Even more informative was the time to recovery:
Median time to resolution: 8 weeks. Interquartile range: 3–16 weeks.
This is probably one of the most useful numbers available when someone asks, broadly, “How long does SIRVA last?”
But it should not be converted into a rule that SIRVA should resolve by eight weeks. A median means half resolved sooner and half later.
Some Patients Have Persistent Symptoms
The same Marshall study demonstrates that a meaningful minority had a much more prolonged course. Among patients with imaging abnormalities, 18% continued to report symptoms at follow-up. By comparison, every patient whose imaging showed no abnormality reported complete symptom resolution at follow-up.
The patients with persistent symptoms could experience substantial consequences. The study reported ongoing limitations in activities of daily living, and some patients had difficulty working or caring for dependents. Among those with continuing symptoms, two had modified their careers because they could no longer perform required manual tasks.
Thus, although the overall prognosis in this cohort was favorable, persistent clinically meaningful SIRVA clearly occurred.
Hesse and colleagues examined 476 conceded SIRVA claims from the National Vaccine Injury Compensation Program. Only 24.3% were documented as having resolved by the last medical visit available in the records. Patients with residual symptoms continued to report pain, restricted motion, impingement, weakness, and muscle atrophy.
At first glance:
Marshall: 86% complete resolution
versus
Hesse: 24.3% resolved at last documented visit
looks like an enormous contradiction. It probably is not.
These Populations Should Not Be Compared Directly
The Hesse cohort consisted of conceded compensation claims. That creates substantial selection bias toward more persistent and clinically consequential cases.
The Marshall cohort was a vaccine-safety registry population followed specifically to determine longer-term outcomes.
There is another important difference. Hesse reported whether symptoms had resolved by the last available medical-record visit. That does not necessarily mean that everyone with symptoms at the last recorded visit remained symptomatic indefinitely.
A patient might improve after the medical record ended. Our natural-history evidence dataset therefore deliberately classifies Hesse as secondary natural-history evidence from a selected medicolegal population and specifically cautions against pooling its 24.3% resolution figure with ordinary clinical cohorts.
This is exactly why prognosis cannot be estimated simply by averaging percentages from different SIRVA studies.
“Still Symptomatic at Last Follow-Up” Does Not Mean Permanent
This distinction is particularly important.
Suppose a case report follows a patient for six months and states:
“Symptoms improved but mild pain persisted.”
What can we conclude? We can say:
The patient had residual symptoms at six months.
We cannot say:
The patient had permanent symptoms.
Likewise, if a report ends after three months, absence of further published follow-up does not establish either recovery or continued disability.
For this reason, our SIRVA natural-history dataset uses deliberately conservative outcome definitions. Complete recovery is counted only when the authors explicitly report resolution, asymptomatic status, or return to baseline. “Improved,” “near complete,” or full range of motion with residual pain is not coded as complete recovery.
That distinction matters when interpreting the literature.
Why Can SIRVA Last Longer in Some Patients?
SIRVA is not a single shoulder pathology. Reported diagnoses include:
bursitis,
rotator cuff tendinopathy,
adhesive capsulitis,
synovitis,
inflammatory joint pathology,
tendon injury,
osseous abnormalities,
and mixed presentations.
These conditions have different natural histories. A relatively isolated acute bursitis may resolve much faster than established adhesive capsulitis.
Similarly, a patient with persistent tendon pathology or an evolving osseous abnormality may have a different course from someone with transient inflammatory bursitis.
This heterogeneity is one reason it is difficult to provide one universal prognosis for “SIRVA.”
Adhesive Capsulitis Can Prolong Recovery
Adhesive capsulitis deserves particular attention. Once progressive capsular stiffness develops, the course may resemble frozen shoulder more broadly, with pain followed by substantial loss of motion and potentially prolonged recovery.
This means that two patients who both initially satisfy a SIRVA definition may subsequently follow very different trajectories:
SIRVA → acute bursitis → improvement over weeks
versus
SIRVA → progressive capsulitis → prolonged pain and stiffness over months.
The eventual diagnosis may therefore be more prognostically useful than the SIRVA label itself.
Chronic Imaging Abnormalities Have Been Documented
Persistent symptoms are also supported by chronic imaging studies. Donners and colleagues evaluated patients with chronic SIRVA at a median of approximately 35 weeks after vaccination and identified ongoing abnormalities including greater tuberosity erosions, infraspinatus tendinitis, capsulitis, synovitis, and bone marrow edema.
Serial imaging in one patient demonstrated progression of a greater tuberosity erosion over more than two years. These observations demonstrate that objective abnormalities can persist well beyond the initial injection in selected patients.
They should not, however, be interpreted as showing that chronic imaging abnormalities occur in most SIRVA patients. The Donners study was a small, selected chronic SIRVA cohort.
Does Persistent Pain Mean Permanent Injury?
No. Duration and permanence are different concepts.
A patient may have symptoms for:
3 months,
6 months, or even
12 months
and subsequently improve.
The Marshall cohort is particularly useful in demonstrating that substantial recovery can occur over time. Even though every patient had a condition serious enough to be reported as suspected SIRVA, 86% of those successfully followed ultimately reported complete resolution.
Therefore:
Chronic does not necessarily mean permanent.
Permanent prognosis should generally not be inferred merely from the fact that symptoms have persisted beyond an arbitrary number of months.
Treatment Also Complicates the Meaning of “Natural History”
Another challenge is that very few published SIRVA patients simply receive no treatment and are observed.
Most undergo some combination of:
NSAIDs,
physical therapy,
corticosteroid injection,
hydrodilatation,
image-guided procedures,
or surgery.
In the Hesse cohort, 80% underwent physical or occupational therapy, 60.1% received at least one shoulder corticosteroid injection, and 32.6% underwent surgery.
The Marshall cohort similarly included analgesic use, allied health treatment, specialist evaluation, and corticosteroid injections.
Therefore, most published outcome data describe the clinical course under treatment, not the untreated natural history of SIRVA.
Case Reports Can Make SIRVA Look More Chronic Than It Is
Case reports are valuable for understanding unusual presentations and treatment-resistant disease.
But they are poorly suited to answering:
“What percentage of all SIRVA patients recover?”
A patient who develops mild bursitis and completely recovers after several weeks is unlikely to become a published case report.
A patient with three years of persistent pain, unusual MRI findings, multiple failed treatments, and an uncommon procedure is much more likely to be published. This is classic publication bias.
Our natural-history dataset therefore separates individual case reports from larger longitudinal cohorts rather than treating every published patient as representing the same underlying population.
Pharmacovigilance Data Should Also Be Interpreted Carefully
VAERS and other pharmacovigilance databases sometimes report whether a patient was “recovered” at the time the report was submitted.
Those figures should not be confused with longitudinal prognosis. Our reconciled natural-history dataset, for example, records the Hibbs VAERS study as having a median reported recovery duration of approximately 70 days among resolved reports, while noting that most reports were not recorded as resolved at the time of submission. But the dataset explicitly classifies these as reporting-status outcomes, not longitudinal clinical follow-up.
A report submitted while a patient is still symptomatic tells us the patient had not recovered yet.
It does not tell us that the patient never recovered.
Can We Predict Who Will Have Persistent Symptoms?
The evidence is still limited. In the Marshall cohort, median time to resolution did not differ significantly based on:
sex,
provider type,
presence versus absence of imaging abnormalities,
or rapid versus delayed symptom onset.
Interestingly, patients with imaging abnormalities were more likely to remain symptomatic at follow-up, but the study was not designed to establish a validated prognostic model.
At present, there is no reliable SIRVA calculator that can predict exactly how long an individual patient will remain symptomatic.
Prognosis Should Be Diagnosis-Specific
A more useful clinical question than:
“How long does SIRVA last?”
may be:
“What pathology does this patient have, and what is the expected course of that pathology?”
For example, prognosis may differ substantially among:
isolated bursitis
adhesive capsulitis
rotator cuff pathology
persistent synovitis
direct tendon or osseous injury, and
mixed inflammatory/structural disease.
This is why accurate diagnosis matters not only for treatment but also for prognosis.
How Should Future Medical Care Be Considered?
Persistent symptoms do not automatically mean that indefinite future treatment is medically necessary.
Likewise, improvement does not mean that no additional treatment will be required.
Future-care analysis should consider:
current symptoms and function,
objective findings,
specific diagnosis,
treatment already attempted,
response to those treatments,
whether improvement is continuing,
whether additional treatment has a reasonable clinical indication,
and whether a plateau has actually been reached.
The published SIRVA literature does not support a universal point at which all patients should be considered permanently recovered or permanently impaired.
A Practical Way to Explain the Timeline
Based on the current evidence, a useful way to describe prognosis is:
Weeks: Many patients improve substantially, and in the Marshall cohort the median complete resolution time was approximately 8 weeks.
Several months: Persistent symptoms are well documented, particularly with adhesive capsulitis, tendon pathology, or more substantial inflammatory disease.
Six months and beyond: Chronic SIRVA clearly occurs, but published estimates vary greatly depending on how patients were selected.
Years: Persistent symptoms and objective abnormalities have been reported in selected cases, but these reports should not be interpreted as the typical SIRVA course.
This is not a validated staging system. It is a practical summary of the range represented in the literature.
The Bottom Line
So, how long does SIRVA last? The strongest available longitudinal evidence suggests that many patients recover within several weeks to several months.
In the Marshall clinical cohort:
Median time to complete resolution was 8 weeks, with an interquartile range of 3–16 weeks, and 86% of patients available for long-term follow-up reported complete recovery.
But that is not the experience of every patient.
Persistent symptoms lasting six months, a year, or longer are well documented, particularly in selected chronic cohorts, medicolegal populations, and case reports. The original SIRVA series itself consisted of patients symptomatic for more than six months.
The most defensible interpretation is therefore:
SIRVA does not have a fixed duration. Many patients recover within weeks to months, while a smaller subset develops prolonged symptoms. Prognosis should be based on the specific shoulder pathology, objective findings, treatment response, and trajectory over time rather than symptom duration alone.
And importantly:
Persistent symptoms at a particular follow-up point do not establish permanent injury, just as improvement does not necessarily establish complete recovery.
References
Atanasoff S, Ryan T, Lightfoot R, Johann-Liang R. Shoulder injury related to vaccine administration (SIRVA). Vaccine. 2010;28(51):8049-8052.
Hesse EM, Atanasoff S, Hibbs BF, et al. Shoulder Injury Related to Vaccine Administration (SIRVA): petitioner claims to the National Vaccine Injury Compensation Program, 2010-2016. Vaccine. 2020;38(5):1076-1083.
Hibbs BF, Ng CS, Museru O, et al. Reports of atypical shoulder pain and dysfunction following inactivated influenza vaccine, Vaccine Adverse Event Reporting System (VAERS), 2010-2017. Vaccine. 2020;38(5):1137-1143.
Marshall T, Addison M, Crawford NW, Buttery JP, Cheng DR. Aiming too high: Shoulder injury related to vaccine administration (SIRVA): a case series. Vaccine. 2022;40(52):7505-7509.
Donners R, Gehweiler JE, Kovacs BK, et al. Chronic stage magnetic resonance imaging findings in patients with shoulder injury related to vaccine administration (SIRVA). Skeletal Radiol. 2023;52:1695-1701.
Comments