Glenohumeral Osteoarthritis

Who: Usually over 50. Also the 30-year-old whose shoulder dislocated at 22.
What: Deep aching pain with progressive stiffness and crepitus. Worse overhead and at end range.
How it behaves: Mechanical and load-related, with morning stiffness that eases as they get moving.

The illness script in one line: a stiff, crepitant, aching shoulder in an older adult — where both active and passive range are lost together, and the plain film will confirm it.

The numbers that matter — and how few of them there are:

(First two: [1], narrative review, no confidence intervals. Third: [2], n = 154, mean 15.2 years.)

Why those figures are so soft: six competing radiographic classification systems make the true incidence — and therefore the prevalence — unobtainable [3]. And globally, the shoulder isn't even modelled separately: the GBD study reports 595 million people with OA in 2020 with the knee as the most common site, and the shoulder sits inside an unmodelled "other sites" category [4].

Compare that with the knee OA page, where Australia has a national standard and a Cochrane review of 139 trials. Same disease, same joint surface, a fraction of the evidence.

⚠️ No Australian guideline exists for this presentation. Anchored to the APTA Physical Therapist Management of Glenohumeral Joint Osteoarthritis CPG [5], the only physiotherapy-specific guideline in existence, with the AAOS CPG (2020, evidence to June 2019) as the medical and surgical anchor. Australia contributes registry data, not guidance — the AOANJRR.

Pattern > Label. The decision isn't "is it arthritis". It's is this the capsule, the joint surface, or the cuff — and §3 is that decision.

Five stages, in your course's order. Insidious, so BEHAVIOUR before HISTORY.

MAIN PROBLEM

The reach they've lost. Seatbelt, back pocket, top shelf, hair, doing up a bra. Same functional list as frozen shoulder — which is exactly why the subjective alone won't separate them.

BODY CHART

• Deep and poorly localised, over the joint line and deltoid. They cup it rather than point.
• May refer to the upper arm. Rarely below the elbow — clear the cervical spine.

BEHAVIOUR

• Aggs: overhead, end range, reaching behind, carrying, lying on it.
• Eases: rest, and — the useful one — movement, after a warm-up period.
• 24-hr pattern: stiff first thing, loosens with movement, sore again by the end of a loaded day.
• Severity: VAS 0–10 at best / at worst / right now.
• Crepitus and grinding the patient volunteers unprompted. Ask, because it's one of the three EULAR-style signs and it's the one that separates this from a capsule.
• Irritability: vigour, severity, time to settle.

HISTORY

• Gradual, over months to years.

• The three history features associated with glenohumeral arthritis ([6], systematic review of 21 studies):

  Feature
  Prior shoulder dislocation
  Age over 75
  A diagnosis of knee osteoarthritis

  The review reports no sensitivities, likelihood ratios or odds ratios in its abstract — so these are associations to ask about, not numbers to quote.

• Ask about dislocation properly, because this one has figures. In 154 patients under 40 with anterior instability, 22.7% had symptomatic GHJ OA at a mean of 15.2 years [2]. Risk factors within that cohort:

  	Odds ratio
  Hyperlaxity	10.1 (95% CI 1.4–72.4)
  Manual labourer	6.1 (1.02–36.1)
  Smoking	4.3 (1.1–16.5)

  Wide intervals, small numbers — but a 25-year-old with a dislocated shoulder is someone whose future you can actually mention.

• Previous imaging, and what they were told. See §5.

SPECIAL / SCREENING QUESTIONS

General health, weight loss, fever, malignancy history, inflammatory features, medications, neurological symptoms. See §4.

A note on how thin this is: the evidence base for this pattern holds almost no subjective research — most of it is surgical. The best published description of the presenting pattern is one sentence in a family-medicine review: over 50, it "usually presents as gradual pain and loss of motion" [7]. That is the state of the literature.

The discrimination, and the honest state of it

This page's central problem: a capsular pattern (ER > ABD > F) is commonly taught for GHJ OA, and there is no evidence for it.

The only study that directly tested whether a consistent capsular pattern exists in stiff shoulders concluded, in its own words: "No single capsular pattern emerged" [8]. The Cyriax capsular pattern is a teaching convention. No study has measured the presenting range profile in GHJ OA at all.

What that same study does give you is the one measured anchor in this whole comparison — and it belongs to the other condition:

So teach magnitude, not pattern. External rotation that has collapsed to near-nothing with a firm end-feel and no crepitus is a capsule. External rotation reduced but present, with a hard end-feel and crepitus, is a joint surface. Active loss with preserved passive is a cuff.

And say the second part out loud: that contrast is reasoning, not a validated rule. One of those three descriptions has a number behind it. The other two do not.

Jones' hypothesis categories that change your plan: sources of symptoms (capsule vs joint surface vs cuff), pathophysiology (this one is genuinely structural, unlike most of what you'll see), and prognosis.

Same stiff shoulder, three stories. Crepitus and a hard end-feel is the joint. No crepitus and a firm end-feel is the capsule. Active-only loss is the cuff.