Shoulder instability: the age-at-first-dislocation rule.

The rule, plainly.

A young person dislocates a shoulder. They come into the emergency department, the shoulder is reduced, they go home in a sling. The orthopaedic registrar tells them to follow up with the GP. Most of the time, that is where the story ends. It should not be.

If the patient is under 25, dislocates the shoulder in contact sport, and has no surgical input over the next six months, the chance of dislocating that same shoulder again is over 80%.¹ This is not a soft probability. It is the most reliable epidemiological signal in the whole of orthopaedic shoulder practice.

The decision about whether to operate on a first-time shoulder dislocator is therefore not really a clinical decision. It is an actuarial one. The number is the decision.

Why age does the work.

Three reasons:

1. The labrum tears, and it doesn't heal. The first dislocation in an anatomically normal young shoulder almost always tears the anterior labrum — the Bankart lesion. Once detached, the labrum does not knit back to bone on its own. The front of the socket has lost its bumper, and the shoulder is more likely to come out again.
2. The bone starts to disappear. Each subsequent dislocation grinds a little more bone off the front of the glenoid socket. By the time a patient has dislocated three or four times, the imaging often shows a measurable defect at the anterior glenoid rim — and the surgical decision becomes considerably harder.
3. The activity profile. A 19-year-old surfer who dislocates her shoulder is not going to stop surfing. A 22-year-old rugby player is not going to stop tackling. The mechanism that caused the first dislocation will recur. Older patients can sometimes modify activity around the shoulder; younger patients usually cannot, and should not have to.

What "stabilisation" actually means.

The umbrella term — shoulder stabilisation — covers two fundamentally different operations. The choice between them is the single most consequential surgical decision in shoulder instability surgery.

Arthroscopic Bankart repair.

The keyhole repair of the torn anterior labrum. Small suture anchors are placed into the front of the glenoid, and the labrum is sutured back to its native bony footprint. This restores the anterior labral bumper. Modern arthroscopic Bankart repair has good long-term results in shoulders with minimal bone loss — reported re-dislocation rates of 5–15% in well-selected patients.²

It is the right operation in patients with:

• Recurrent anterior shoulder instability
• Minimal glenoid bone loss (typically less than 15% on CT)
• An "off-track" engaging Hill-Sachs lesion that can be addressed with an arthroscopic remplissage

It is not the right operation when the bone is gone. Read about arthroscopic Bankart repair →

Latarjet procedure.

An open bone-block transfer. A small finger of bone from the front of the scapula (the coracoid process) is taken with its conjoint tendon attached and transferred to the deficient front edge of the glenoid socket. The operation provides three protective mechanisms simultaneously:

• Bone — the coracoid block restores the anterior glenoid.
• Dynamic sling — the conjoint tendon tensions across the front of the shoulder during abduction and external rotation, preventing the head from translating forward.
• Capsular repair — the coracoacromial ligament is used to repair the anterior capsule over the new bone block.

Long-term re-dislocation rates after Latarjet are reported at 2–5%.³ It is the operation for shoulders with significant glenoid bone loss or where an arthroscopic Bankart has previously failed. Read about shoulder stabilisation and Latarjet →

The bone-loss threshold — and why the line is moving.

Burkhart and De Beer's 2000 paper drew the original line at approximately 25% anterior glenoid bone loss — above that, arthroscopic Bankart was unreliable.⁴ Subsequent work refined the threshold downward to around 15–20%.

The 2014 on-track/off-track concept by Di Giacomo⁵ changed the conversation again — what matters is not just the size of the glenoid defect, but the relationship between the glenoid track and the Hill-Sachs lesion on the back of the humeral head. An "off-track" Hill-Sachs that engages the front of the glenoid in functional positions of abduction and external rotation will cause recurrent instability even after a technically excellent labral repair.

In 2026, the practical threshold for considering bone augmentation is moving lower again — closer to 13–15% bone loss when combined with an off-track Hill-Sachs lesion. The reason: the published failure rate of arthroscopic Bankart in shoulders that crossed the older threshold has been higher than the early literature suggested.

Shoulder Instability — Part Two: the surgical-management half of the series, paired with the surgical sections of this essay.

For a deeper dive on where remplissage fits in the modern decision algorithm — the operation that sits between Bankart and Latarjet for the off-track-but-subcritical-bone-loss patient — read the Remplissage essay →

The new option: arthroscopic distal tibia allograft.

The Bankart-versus-Latarjet dichotomy is no longer the only choice. A third option — arthroscopic distal tibia allograft for anterior glenoid bone loss — is becoming a serious alternative for selected patients. The technique uses a piece of cadaver distal tibial bone that has both bone stock and a cartilaginous surface that matches the native glenoid better than a coracoid transfer. Critically, the entire operation can be done arthroscopically.

For young patients with significant bone loss who want to avoid the open Latarjet, this is now a serious option. Dr Coory observed the technique in detail at the Smith+Nephew Inspire Shoulder course in Singapore in May 2026, and is adding it to the practice in selected cases. Notes from Singapore and Sydney →

When NOT to operate.

Three groups are usually better not operated on:

• The single first-time dislocator over 40 — the recurrence rate is low and the issue is more often a co-existing rotator cuff tear, which is a different problem.
• The patient with generalised joint hyperlaxity and multidirectional instability — the right answer is usually physiotherapy and neuromuscular control, not surgery.
• The patient who has dislocated but is unwilling to modify activity during a six-month rehabilitation period. The operation does not work if the rehabilitation does not.

What this means for your shoulder.

If you (or a young athlete in your family) has dislocated a shoulder, three questions are worth asking:

1. Has the imaging quantified the bone loss with a 3D CT — not just an MRI?
2. What is the on-track/off-track status of the Hill-Sachs lesion?
3. Given the activity profile, what is the realistic re-dislocation rate without surgery, with arthroscopic Bankart, and with Latarjet (or allograft)?

The answer to those three questions is what the operation decision turns on. Part Two of the series will cover the surgical decision tree in more detail.

References.

1. Owens BD, Dawson L, Burks R, Cameron KL. Incidence of shoulder dislocation in the United States military. J Bone Joint Surg Am. 2009;91(4):791–796.
2. Boileau P, et al. Risk factors for recurrence of shoulder instability after arthroscopic Bankart repair. J Bone Joint Surg Am. 2006;88(8):1755–1763.
3. Hovelius L, et al. Long-term results of the Bristow-Latarjet procedure. J Shoulder Elbow Surg. 2012;21(5):647–660.
4. Burkhart SS, De Beer JF. Traumatic glenohumeral bone defects and their relationship to failure of arthroscopic Bankart repairs. Arthroscopy. 2000;16(7):677–694.
5. Di Giacomo G, et al. Evolving concept of bipolar bone loss and the Hill-Sachs lesion: from "engaging/non-engaging" to "on-track/off-track". Arthroscopy. 2014;30(1):90–98.