Adhesive Capsulitis

(08/03/2026) Week 217

Frozen Shoulder - Review

Challoumas D, Biddle M, McLean M, Millar NL. Comparison of treatments for frozen shoulder: a systematic review and meta-analysis. JAMA Netw Open. 2020;3(12):e2029581. doi:10.1001/jamanetworkopen.2020.2958

Adhesive Capsulitis TL:DR:

The incidence is 1 - 5% in the general population with women > men and usually presenting between ages 40-60. DM and hypothyroidism are associated conditions.

Classically, the history of frozen shoulder has three phases: pain (2-9 months), stiffness (4-12 months) and recovery (5-24 months) with all three phases progressing on average over 1-2 years. Full recovery is expected, but some patients will continue to have functional limitations beyond that time frame. It is also extremely rare to get adhesive capsulitis again in the same shoulder.

Clinically, the pain is described as dull and non-localized with limited range of motion.

If the clinical presentation is suspicious for adhesive capsulitis, MRI should be considered to evaluate for other causes of symptoms.

What’s the underlying cause? Unknown.

Adhesive Capsulitis Treatment:

Intra-articular (IA) steroid injection is associated with better short term outcomes (pain and range of motion), but the effect was not seen in long term outcomes. Injections also appeared to beat PT, but, again, only for short term outcomes. Home exercises had some benefits in short term pain reduction, but the effect was modest.

All of these benefits were small when compared to no treatment.

In fact, none of the studied treatments improved long term outcomes.

The conclusion from the JAMA review article was that IA steroid injection was an effective way to improve short term outcomes (mostly pain) when paired with PT and/or home exercises.

I would agree that is a reasonable take, but given the lack of sham/saline comparators in those studies, I suspect there is unmeasured placebo effect.

My personal takeaway is that when I have a patient who I suspect has adhesive capsulitis (frozen shoulder), it is important to evaluate for other causes of symptoms. Not doing so could lead to erroneous watchful waiting.

I had been planning on stopping here, but something kept bothering me.

Why is the evidence base so weak?

1. Is it because we do not understand the disease process to begin with?

2. Is it because we do not design adequate trials, whether that be due to lack of funding, rareness of the disease or misaligned incentives in our procedure based medical economy?

The messy reality is probably born out of both reasons, and it is my opinion we do not teach this uncertainty adequately.

I can think of many other disease processes where our lack of understanding of the disease itself threatens the evidence base. HFpEF, Diabetes, Lupus and to some extent even common diseases like hypertension and hyperlipidemia. As my mentor recently asked me, What is “Hypertension”?

I would posit that the confidence of your answer is inversely proportional to your ability to see the forest for the trees.