Hill-Sachs Lesions

A Hill-Sachs lesion is a compression fracture of the posterolateral humeral head, made when the head is driven against the anterior glenoid rim during an anterior dislocation. It is one half of the "bipolar bone loss" of anterior shoulder instability; the anterior glenoid is the other half.

How common. A Hill-Sachs lesion is present in 40 to 90 percent of first-time anterior dislocations and in up to 90 to 100 percent of shoulders with recurrent instability.

Why it matters. A large or "off-track" lesion engages the anterior glenoid rim in the at-risk position and levers the head out of the socket. A lesion that is missed carries a high risk of recurrence.

Where it forms. The defect sits on the posterolateral aspect of the humeral head, superior to the greater tuberosity. The dislocation is an anterior one with the arm attempting to internally rotate, typically a traumatic abduction and external rotation injury, and the lesion is the dent left when the soft cancellous bone of the head impacts the hard cortical bone of the anterior glenoid rim.

View larger

Bipolar bone loss. The Hill-Sachs lesion interacts with any anterior glenoid bone loss (the bony Bankart). It is the combined loss that reduces the arc of stability.

Engagement. In abduction and external rotation, the at-risk position, the perceived width of the glenoid track narrows. If the Hill-Sachs lesion is wide enough to bridge the narrowed track, the anterior glenoid rim falls into the defect, and that levering action forces the humeral head out of the socket.

History. A history of recurrent anterior dislocations, and a feeling of the shoulder "locking" or getting stuck in abduction and external rotation.

Examination. The findings to elicit are these:

• Apprehension test: positive in abduction and external rotation
• Relocation test: apprehension relieved by a posteriorly directed force
• Crepitus: may be felt in the mid-range of rotation if the lesion is large
• Engaging sign: symptoms (a clunk or apprehension) reproduced at lower degrees of abduction may suggest an engaging lesion

Radiographs. The standard AP often misses the posterolateral defect, so request the views that profile it:

• AP in internal rotation
• Axillary view
• Stryker notch view: hand on the head, beam tilted 10 degrees cephalad; the best plain view for a Hill-Sachs lesion
• West Point view: the best plain view for glenoid bone loss

View larger

CT. The gold standard for quantifying bone loss; the key CT view is the axioscapular. The protocol is a 3D reconstruction with the humeral head subtracted, which gives the en-face glenoid view on which the track is measured.

MRI. Its job is the soft tissue: the Bankart lesion and rotator cuff tears. It can estimate bone loss, but CT is superior; on MRI the axial cuts are where the depth of the lesion is seen.

View larger

A posterolateral humeral head defect or a "locking/apprehensive" shoulder is not always a true engaging Hill-Sachs lesion. Distinguish it from the following:

The two questions. The decision is made in order: first the glenoid, then the track. Both bipolar lesions are always addressed.

Step 1: how much glenoid is missing. Glenoid bone loss greater than 20 to 25 percent goes to a Latarjet, regardless of the Hill-Sachs lesion. Less than that, and the track decides. The 20 to 25 percent figure is older teaching; contemporary work puts the point at which recurrence rises as low as 13.5 percent in high-demand athletes, which lowers the threshold for augmentation in that group.

Step 2: is the Hill-Sachs on-track. An on-track lesion is treated with an arthroscopic Bankart repair alone. An off-track lesion is treated with an arthroscopic Bankart repair plus a remplissage; a Bankart repair alone for an off-track lesion has a high recurrence rate.

Algorithm

View larger

The options. Chosen by patient demand and bone loss:

• Non-operative: only for low-demand patients and first-time dislocators with small on-track lesions; rarely successful in young active patients
• Arthroscopic Bankart repair: soft-tissue repair only; failure rate high if the lesion is off-track
• Remplissage: infraspinatus capsulotenodesis; "fills" the defect with tendon, making it extra-articular
• Latarjet: coracoid transfer; increases the width of the glenoid track (lengthens the track)
• Humeral head allograft: for massive defects (greater than 40%) where remplissage is insufficient
• Resurfacing arthroplasty: for older patients with degenerative changes

A minority of lesions are too large or too deep for soft-tissue filling, and there the humeral side itself must be reconstructed rather than filled or made extra-articular. The ladder rises with defect size and articular involvement:

View larger

The reverse Hill-Sachs (McLaughlin) lesion is the posterior mirror image of the classic defect, and it is examined just as often. It is an impaction fracture of the anteromedial humeral head, made when a posterior dislocation drives the head onto the posterior glenoid rim. The classic causes are the "three E's": Epilepsy (seizure), Electrocution and Extreme trauma.

Why it is missed. The arm is locked in internal rotation and adduction with no external rotation, and the AP film can look near-normal. Look for the "light-bulb" sign of fixed internal rotation, the rim sign and the trough-line sign, and always obtain an axillary view or a CT.

View larger

Management by defect size. The reverse lesion is stratified by the proportion of the articular surface involved:

• Under roughly 25 percent: closed reduction and immobilisation (some surgeons brace in slight external rotation); a small, stable, congruent lesion can be managed non-operatively
• Roughly 25 to 50 percent: the McLaughlin procedure, transfer of the subscapularis tendon into the defect so that it fills the engaging notch and blocks re-dislocation; the modified McLaughlin (Neer/Hawkins) transfers the lesser tuberosity with its attached subscapularis into the defect, for bone-to-bone healing
• Over roughly 50 percent, or chronic/locked: disimpaction and bone grafting, segmental osteochondral allograft, partial resurfacing, or arthroplasty in the older or arthritic patient; chronic locked posterior dislocations usually require open reduction and reconstruction

View larger

View larger

View larger

View larger

After remplissage. The complications specific to the tenodesis:

• Stiffness: excessive loss of external rotation if the sutures are tied too tight (a tenodesis effect)
• Pain: posterior cuff pain or infraspinatus spasm (cramping) is common early after surgery
• Recurrent instability: when the glenoid bone loss was underestimated and the track calculation was wrong
• Infraspinatus strength: a minor deficit in external rotation strength, usually clinically insignificant

After Latarjet. The complications are neurological, hardware-related and graft-related:

• Musculocutaneous nerve: the most common nerve injury, about 5 percent, a traction injury during retraction of the conjoined tendon
• Axillary nerve: at risk during the subscapularis split or tenotomy
• Screw breakage or back-out: causes irritation
• Proud screws: "kissing lesions" on the humeral head, leading to rapid arthritis
• Graft problems: non-union (a fibrous union is usually stable) and osteolysis, resorption of the graft
• Recurrence: lower than after a Bankart repair, less than 5 percent, but revision is difficult (Eden-Hybinette)

Any instability surgery. Infection is less than 1 percent for arthroscopy and slightly higher for an open Latarjet. Chondrolysis was historically associated with pain pumps and thermal capsulorrhaphy (radiofrequency shrinkage), both now largely abandoned. Stiffness follows overtightening of the anterior capsule (accessory anteroinferior instability repair).

The protocol. After a Bankart repair with remplissage the standard protocol is 6 weeks in a sling, protecting both the capsulolabral repair and the tenodesis, before motion and then strength are rebuilt in stages:

External rotation. The infraspinatus tenodesis is under tension in internal rotation, yet it is excessive external rotation that is guarded against, because that is what pulls out the anchors. Most surgeons limit external rotation to neutral for 6 weeks.

Active elevation. Avoided for 6 weeks, to protect the Bankart repair from the shear of the humeral head translating.

Bankart repair alone. For an on-track lesion the success rate is 85 to 90 percent. For an off-track lesion the recurrence rate is high, greater than 20 to 30 percent.

Adding the remplissage. For an off-track lesion, Bankart repair plus remplissage gives success rates that approach those of the Latarjet, 90 to 95 percent, with lower morbidity.

Latarjet. The gold standard for collision athletes and significant bone loss, with greater than 95 percent stability.

Global Epidemiology

• Anterior dislocation accounts for over 95 percent of glenohumeral dislocations; a Hill-Sachs lesion is present in roughly 40 to 90 percent of first-time anterior dislocations and approaches 90 to 100 percent in recurrent instability.
• Peak incidence is bimodal: young males (15 to 30 years, sport/trauma) and older adults (falls). Young age, male sex, high-demand/contact sport and bipolar bone loss are the dominant risk factors for recurrence.
• Bipolar (combined glenoid plus humeral) bone loss is the single most important predictor of failure after soft-tissue stabilisation across populations.

Society Guidance, Side by Side

Registry & High-Level Evidence Notes

• There is no dedicated implant registry for instability (no implant), so evidence is driven by RCTs and cohorts rather than arthroplasty-style registries.
• Convergent message across high-level data (MacDonald RCT; Haroun meta-analysis; Schwihla long-term cohort): address the off-track lesion, and remplissage matches Latarjet stability with fewer complications when glenoid loss is subcritical.

High- vs Limited-Resource Practice Variation

• Well-resourced settings: routine preoperative 3D CT with en-face glenoid reconstruction for track quantification; arthroscopic Bankart plus remplissage is the default for off-track subcritical lesions.
• Limited-resource settings: MRI or plain CT (or even instability radiographs and intraoperative dynamic assessment) may substitute for 3D CT; open Latarjet remains a reliable, instrument-light, single-stage option where arthroscopic remplissage expertise or equipment is unavailable.

• The 83 percent figure is a population mean, not a hard threshold. The original cadaveric track was 84 plus or minus 14 percent of glenoid width (Yamamoto/Itoi 2007); the working 0.83 multiplier is a convenient approximation. Borderline (near-track) lesions sit in a grey zone where intraoperative dynamic assessment still matters.
• Does off-track status independently predict failure, or is glenoid loss the real driver? Long-term data (Schwihla 2023) show off-track lesions fail isolated Bankart repair at high rates, yet some cohorts (Park 2019) found glenoid defect size, not off-track status per se, was the more reliable predictor once selective remplissage was used. The honest position: bipolar interaction matters, and glenoid loss carries heavy weight.
• Remplissage vs Latarjet for subcritical loss. Meta-analytic data (Haroun 2020) show equivalent stability with fewer complications for remplissage, but Latarjet retains advocates for collision athletes and revision settings. There is no consensus "right answer" — the decision is patient- and surgeon-specific.
• The "subcritical" glenoid threshold is moving. Older teaching used 20 to 25 percent; contemporary work suggests instability and recurrence rise from as little as 13.5 percent in high-demand athletes, lowering the threshold for augmentation.
• Clinical engagement vs radiographic off-track are correlated but not identical. A lesion can be radiographically off-track yet not reproduce clinical locking, and vice versa; relying on one alone risks over- or under-treatment.