Injection infection? Another surprising way to freeze a shoulder

One possible cause of frozen shoulder is infection. The causality isn’t proven, but it does rest on some evidence (previously discussed), and it’s a good example of an unexpected and maybe underestimated pathological mechanism for a common painful problem, a slow-burn infection of the joint with Propionibacterium acnes. How does it get in there? P. acnes is endemic on our skin and in our pores, harder to wipe out than bed bugs — so even a properly disinfected incision site isn’t always P. acnes free, and it can get dragged into the joint by surgical tools.

If surgery can do it, then there’s probably a much more common way for P. acnes to catch a ride into the joint capsule…

Shoulder injury related to injections

The number one candidate for a non-surgical mechanism of infection is injection. While injections cause a lot fewer complications than surgery, anything that breaks the skin is still “invasive” and involves some infection risk. The most common injections are vaccinations, which is why this phenomenon is often referred to as “shoulder injury related to vaccine administration” (SIRVA).

But this has nothing to do with vaccination itself — this is just about needles, not what’s in them, and the last thing the world needs now is more demonization of vaccinations. So let’s not call it SIRVA! That’s a bullshit term that seems to go out of its way to emphasize the wrong thing, at the wrong time in history. It should be “shoulder injury related to injection” — SIRI! 😜

Most injections are in the shoulder, because most injections are intramuscular, and the shoulder is a convenient spot with a bunch of muscle. But sometimes shoulder injections miss: the needle goes too deep, into the joint capsule instead of the muscle tissue. Oops. Within hours, or days at most, it becomes clear that the shoulder is much more sore than it should be. The symptoms may drag on and on: “Patients will often visit their physicians months later because they are not able to carry out daily tasks that were possible before the vaccination” (Bancsi). “The prolonged pain and stiffness…is distinct—in other words, much worse—than typical soreness from shots” (Zhang). The conventional wisdom is that this is an ordinary minor injury, which is often mistaken for conditions like bursitis, rotator cuff tears, and adhesive capsulitis. Bancsi et al.:

During physical examination and on ultrasound scan, SIRVA will not appear to be any different from routine shoulder injuries. The only difference is that the shoulder symptoms will have started within days of a vaccination [or any other injection].

But is that the “only difference”? It’s not really normal for a minor physical trauma to drag on for weeks and months! It seems like something more might be going on here.

Unfortunately, it’s plausible that injections can trigger genuine adhesive capsulitis by dragging P. acnes right into the joint capsule, just like a surgery probably can, and then excessive and prolonged inflammation ensues — much more than minor physical trauma would ever cause. And that creates a genuine case of frozen shoulder.

Propionibacterium acnes, a bacterium that does not belong in shoulders.

Citation needed, but where there’s smoke…

Injection injuries are certainly a real thing, and frozen shoulder is a real thing, but … do they overlap? Is there any evidence that injection injury specifically can cause frozen shoulder, via the mechanism of P. acnes infection? Not direct evidence, no. But we do have a lot of suggestive case reports. There’s a 2015 paper reporting three injection injuries leading to a confirmed diagnosis of frozen shoulder (Saleh). After 2020 there was, unsurprisingly, a surge of many more case reports related to Covid (and other) vaccinations. Here’s a few:

• A 2023 Malaysian paper reported seven cases, concluding with the opinion that the phenomenon is “under-reported” and frozen shoulder is “a potential complication arising from improper COVID-19 vaccine administration” (Foong et al).
• An American report of “the first known case of adhesive capsulitis following improper tetanus-diphtheria (Td) vaccine administration” (Weinberg et al).
• A Chinese report of sixteen cases of shoulder injury: 63% frozen shoulder, but I suspect that should be higher (Chu).

And so on. In fact, there were so many post-pandemic reports that we even got an updated systematic review in 2022 from MacMahon et al, concluding that chronic shoulder pain after injections is probably “exceedingly rare” — but still roughly ten times more common than “one in a million.” That review doesn’t even mention infection as a mechanism, and is instead focussed on the possibility of “immune-mediated inflammatory response to vaccine antigens” — for which there is “no confirmatory experimental evidence.” Just as there is none for infection, although I think infection is a more likely culprit.

Case reports are formal anecdotes, far from the best possible evidence … but this is also about as good example of where-there’s-smoke-there’s-fire as you can get. Infection as a complication of both surgery and injection are highly plausible, and there’s good evidence of the first. And it’s also plausible that such infection could lead to frozen shoulder, and there’s also both direct and indirect evidence for that (again, see previous post).

From those solid premises, it’s just a reasonable hop to the hypothesis that injection injuries are the explanation for some rare-but-real cases of post-injection frozen shoulder. Far from proven… but downright likely in my opinion.

Injection safety

The next time I get an injection — probably a vaccination, which I obviously will do, because the benefits dramatically outweigh the risks — I’m going to drown my shoulder in rubbing alcohol. I want as few surviving P. Acnes on my skin as possible when that needle goes through my skin.

“I say we take off & nuke the entire site from orbit. It’s the only way to be sure.”

Ellen Ripley, Aliens

What about getting shots in the butt instead? Surgeon Dr. Michael Skyhar: “This is all avoidable by simply asking that the vaccine be placed in a different location, such as the upper/outer buttocks.”

Unfortunately, that may not be an upgrade! Gluteal injections actually risk an even nastier injury to the sciatic nerve. Mishra et al:

“Sciatic nerve injury from an intramuscular injection in the upper outer quadrant of the buttock is an avoidable but persistent global problem, affecting patients in both wealthy and poorer healthcare systems. The consequences of this injury are potentially devastating. Safer alternative sites for intramuscular injection exist.”

Safer sites like the shoulder!

If done correctly, the risk could be reduced to near zero… but that can be said of deltoid injections too. Is it worth avoiding the tiny risk of a shoulder injury by pivoting to even greater risk of hip harm? Especially when so many people dispensing injections are unfamiliar with how to do gluteal injections safely? It seems wise to just stick with the standard.

This is an excerpt from my frozen shoulder tutorial, updated in a few ways, but mainly with the post-pandemic case reports … and a brand-new audio version.