Anterior shoulder instability — from the essential labral lesion through glenoid bone loss quantification, glenoid track engagement, arthroscopic anchor repair, and bony augmentation with Latarjet / remplissage
When the humeral head dislocates anteriorly, the anteroinferior labrum and its attached inferior glenohumeral ligament (IGHL) complex are peeled off the glenoid rim like a stripped gasket. Bankart described this avulsion in 1938 as the "essential lesion" of recurrent anterior instability — the injury that, if left unrepaired, converts a single traumatic dislocation into a lifelong pattern of recurrent instability because the labral "bumper" and capsuloligamentous checkrein against anterior translation are lost.
The glenoid labrum is a fibrocartilaginous ring that deepens the shallow glenoid socket by ~50%, increasing concavity-compression stability and providing the attachment site for the glenohumeral ligaments. The anteroinferior quadrant is the biomechanically critical zone:
• Inferior glenohumeral ligament (IGHL) complex: anterior band, axillary pouch, posterior band — acts as a hammock that reciprocally tightens with arm rotation. The anterior band is the primary checkrein against anterior translation with the arm in the 90°/90° abduction-external rotation position — precisely the position of a contact-sport tackle or a throwing cocking phase. • Labrum: anchors the IGHL to bone circumferentially and increases the effective glenoid concavity. • Glenohumeral joint stability here relies more on this soft-tissue "bumper" than on bony containment, because the glenoid is a shallow, roughly pear-shaped socket articulating with a much larger humeral head (a ball on a golf tee).
During a forceful anterior dislocation, the humeral head levers against the anteroinferior rim, avulsing the labrum-IGHL complex off the bone, often with the periosteum stripped medially (a "sleeve" avulsion) so the tissue heals, if untreated, medialized and non-anatomic — a healed-but-incompetent Bankart lesion.
Not every anteroinferior labral injury is identical; variants carry different biomechanical consequences and treatment implications:
• Classic (soft-tissue) Bankart lesion: labrum-IGHL avulsed from the glenoid articular margin with an intact bony rim underneath. Amenable to arthroscopic suture anchor re-fixation. • Bony Bankart lesion: a fragment of the anteroinferior glenoid rim avulses along with the labrum. If the fragment is large enough and can be anatomically reduced, it effectively restores glenoid bone stock; if comminuted or resorbed, it becomes a source of glenoid bone loss. • Perthes lesion: labrum stripped but periosteum remains partially intact, so the labrum is displaced medially but not fully detached — easily missed on MRI without ABER (abduction-external rotation) positioning. • ALPSA lesion (Anterior Labroligamentous Periosteal Sleeve Avulsion): the labrum-IGHL sleeve retracts medially and heals in a medialized, non-anatomic position along the glenoid neck — a chronic variant that will not heal correctly without surgical mobilization and repair. • GLAD lesion (Glenolabral Articular Disruption): labral tear with an associated articular cartilage flap injury, usually without instability.
As the humeral head dislocates anteriorly and levers over the anteroinferior glenoid rim, the relatively soft posterosuperior articular surface of the humeral head impacts against the dense cortical bone of the anterior glenoid rim, causing a compression (impaction) fracture — the Hill-Sachs lesion, first described radiographically in 1940. This creates a wedge-shaped, often vertically oriented divot in the posterosuperior humeral head, typically located at or above the equator of the humeral head with the arm at the side.
The Hill-Sachs lesion is not merely a static defect — its clinical significance is entirely dynamic: whether it engages the glenoid rim depends on its size, location, and orientation relative to the glenoid track (Stage 3). A small, superiorly located, obliquely oriented Hill-Sachs lesion may never engage; a large, more equatorial, vertically oriented lesion combined with glenoid bone loss engages readily and drives recurrent instability even after an isolated labral repair.
The paired bipolar bone loss of the Bankart and Hill-Sachs lesions is the biomechanical crux of anterior instability surgery: neither lesion exists in isolation from a decision-making standpoint. Glenoid bone loss decreases the glenoid arc available to contain the humeral head, while Hill-Sachs size determines how much of the humeral articular surface is missing — together they determine whether a "bumper" repair alone will hold or whether bone augmentation is required.
Recurrent instability transforms the glenoid from a round-to-pear-shaped socket into a progressively truncated, "inverted pear" shape as repetitive anterior subluxation events erode the anteroinferior rim. Accurately quantifying this bone loss on CT — using an en-face 3D reconstruction and a best-fit circle overlay — is the single most important preoperative decision point in anterior instability surgery, because it determines whether isolated arthroscopic soft-tissue repair will succeed or whether bony augmentation is required from the outset.
The normal glenoid, viewed en-face (looking directly down the glenoid articular surface), is pear-shaped: narrower superiorly, wider inferiorly, and roughly circular in its inferior two-thirds. As recurrent anteroinferior labral and bony injury accumulates, the anteroinferior rim progressively erodes, flattening or concavely truncating what should be the widest, roundest part of the socket. Burkhart and De Beer coined the term "inverted pear" for this eroded morphology, and demonstrated in a landmark 2000 series that its presence on arthroscopic inspection predicted a dramatically higher failure rate after isolated arthroscopic Bankart repair (~67%) compared with glenoids retaining a normal pear shape (~4%).
This single observation reframed anterior instability surgery: the glenoid's bony geometry, not just the soft-tissue labral injury, is often the dominant driver of recurrence, and surgical planning must start with an objective measurement of how much bone is actually missing.
Two complementary techniques are used on 3D CT reconstructions (humeral head digitally subtracted to expose the glenoid face en-face):
• Best-fit circle (linear) method: a circle is fitted to the intact inferior glenoid articular surface (which is reliably circular even after anterior rim erosion). The distance from the circle's anterior edge to the actual eroded anterior rim (the bone loss width, "d") is measured and expressed as a percentage of the circle's diameter ("D"): %bone loss = d/D × 100. • Pico (surface area) method: rather than a linear measurement, the area of the missing bone loss segment is compared to the area of the best-fit circle, giving a direct percentage of articular surface area lost — considered by many to be more accurate because bone loss is inherently a 2D area deficit, not simply a linear one.
Both methods require a true en-face view (perpendicular to the glenoid face) — an oblique CT cut can dramatically over- or under-estimate bone loss, which is why dedicated 3D reconstruction protocols with humeral head subtraction have become the imaging standard of care preoperatively.
Multiple biomechanical and clinical outcome studies converge on a critical threshold, commonly cited in the range of ~13.5–20% glenoid bone loss, above which isolated arthroscopic soft-tissue Bankart repair carries an unacceptably high recurrence rate:
• <13.5% bone loss: isolated arthroscopic Bankart repair generally has good outcomes and is the standard of care. • 13.5–20% (the "subcritical" zone): outcomes are less predictable; isolated repair may still fail, particularly in young, contact-sport, or hyperlaxity patients — many surgeons individualize the decision using the glenoid track concept (Stage 3) and patient risk factors. • >20–25%: isolated soft-tissue repair failure rates rise sharply (recurrence commonly cited from ~35% up to 70%); bone block augmentation (Latarjet) is generally recommended.
This threshold is not a rigid cutoff but the center of a decision zone that also incorporates the reciprocal Hill-Sachs lesion size (bipolar bone loss), patient age, activity/contact-sport participation, and number of prior dislocations — inputs that converge in the glenoid track model explored next.
The Instability Severity Index Score (ISIS) integrates age, sport type/level, shoulder laxity, Hill-Sachs presence on AP radiograph, and glenoid contour loss into a 10-point score; scores ≥6 predict a high failure rate with isolated arthroscopic repair and favor primary bony augmentation.
Yamamoto and colleagues introduced the glenoid track concept in 2007 to unify Bankart and Hill-Sachs bone loss into a single predictive model: rather than evaluating each lesion in isolation, the model asks whether, throughout functional abduction and external rotation, the Hill-Sachs defect will remain within the contact "track" the glenoid sweeps across the humeral head — or whether it will fall off that track and engage the anterior glenoid rim, precipitating dislocation.
The glenoid track (GT) is the contact-arc width on the humeral head that the glenoid articular surface sweeps across during arm elevation and rotation — anatomically, it is measured as:
GT = 0.83 × D − d
where D is the glenoid diameter (best-fit circle) and d is the anteroinferior glenoid bone loss width. The constant 0.83 reflects that, even in an intact shoulder, the rotator cuff footprint and the glenoid's own coverage limit the track to about 83% of the glenoid diameter rather than 100%. Any glenoid bone loss (d) subtracts directly from this available track width — the mechanism by which glenoid bone loss narrows the safety margin against Hill-Sachs engagement even before the humeral lesion is measured.
The Hill-Sachs interval (HSI) — the medial-to-lateral width of the Hill-Sachs defect plus a margin to the rotator cuff insertion — is then compared against GT. If HSI < GT, the lesion is "on-track": as the arm rotates, the defect stays medial to the glenoid contact zone and never reaches the anterior rim. If HSI ≥ GT ("off-track"), the defect is wide enough that during abduction-external rotation it rotates past the glenoid's contact arc and catches directly on the anterior glenoid rim.
"Engagement" is the biomechanical event underlying recurrent dislocation in the throwing or overhead-reaching position (abduction + external rotation, the position of apprehension): the Hill-Sachs defect, rather than the smooth articular surface of the humeral head, contacts the glenoid rim. Because the defect is a step-off rather than a smooth curve, it catches on the rim edge instead of gliding across it, and the humeral head levers anteriorly out of the socket with relatively little additional force.
On-track lesions do not exhibit this mechanism — the humeral articular surface remains congruent with the glenoid throughout the arc of motion, so an isolated arthroscopic Bankart labral repair, by restoring the anteroinferior bumper and capsular tension, is generally sufficient. Off-track lesions will continue to engage and redislocate even after a technically perfect labral repair, because the labral repair does nothing to address the humeral-sided bony deficiency that is mechanically catching on the rim — these cases require an additional procedure that either fills the humeral defect (remplissage) or augments the glenoid (Latarjet) to move the effective rim posteriorly/laterally out of the engagement path.
This simulator's two sliders map directly onto the glenoid track equation: increasing glenoid bone loss narrows the available track width (GT falls), while increasing Hill-Sachs depth widens the effective Hill-Sachs interval (HSI rises, because deeper, larger lesions require a greater safety margin). When HSI exceeds GT, the on-screen humeral head defect crosses out of the blue track band during its rotational sweep and the status flips to "Off-Track", with the model's predicted recurrence risk rising sharply — visually reproducing why bipolar bone loss (glenoid loss + Hill-Sachs size) must be assessed together, not independently, when planning surgery.
Because both lesions interact multiplicatively rather than additively, a moderate glenoid bone loss (~15%) combined with even a moderate Hill-Sachs lesion can be off-track, while the same Hill-Sachs lesion with a pristine glenoid may be safely on-track — this is why isolated measurement of either lesion, without the combined track calculation, can misclassify surgical risk.
For on-track lesions with subcritical glenoid bone loss, arthroscopic Bankart repair remains the workhorse procedure: the avulsed anteroinferior labrum and IGHL complex is mobilized off the glenoid neck, the bony rim is freshened to bleeding bone, and suture anchors re-fix the capsulolabral tissue back onto the articular margin — restoring both the labral "bumper" that deepens the socket and the ligamentous checkrein against anterior translation.
Arthroscopic Bankart repair follows a consistent sequence:
1. Diagnostic arthroscopy confirms the labral avulsion, assesses the glenoid track/bone loss intraoperatively, and rules out concomitant pathology (SLAP tear, rotator cuff, biceps). 2. Labral mobilization: an elevator releases the medially scarred labrum-IGHL sleeve off the glenoid neck (critical in chronic ALPSA lesions) so the tissue can be restored to its anatomic position on the articular margin rather than repaired in situ where it healed medialized. 3. Glenoid rim preparation: a shaver or rasp decorticates the anteroinferior glenoid neck to bleeding cancellous bone, creating a vascularized bed for tendon-to-bone healing. 4. Anchor placement: 3–4 suture anchors (biocomposite or all-suture) are placed sequentially along the glenoid articular margin from roughly the 5:30 to the 3:00 clock position (right shoulder) — spanning the full extent of the avulsed IGHL anterior band and axillary pouch attachment. Anchors are placed directly on the articular surface edge (not on the glenoid neck) to restore an anatomic, not medialized, labral height. 5. Suture passage and tensioning: sutures are shuttled through the labral-capsular tissue and tied arthroscopically, advancing the tissue superiorly and laterally onto the prepared rim to re-tension the IGHL and reconstruct the labral bumper. 6. Capsular plication: in patients with an element of capsular redundancy or multidirectional laxity, additional plication sutures independently reduce capsular volume.
Biomechanical and clinical studies show that both anchor number and precise rim-edge placement influence the durability of the repair:
• Fewer than 3 anchors is associated with higher recurrence, particularly when the avulsion extends inferiorly past the 6 o'clock position — an inferior anchor is required to address the axillary pouch component of the IGHL, and omitting it leaves a residual sulcus of instability. • Anchors placed too medial on the glenoid neck (rather than directly on the articular margin) restore capsular tension but fail to restore the labral height that deepens the glenoid socket — biomechanical studies show medialized repairs restore substantially less concavity-compression stability than anatomic rim repairs, even when suture tension is identical. • Labral height restoration — the projection of the repaired labral tissue above the glenoid articular surface — directly correlates with restored glenoid concavity and resistance to translation in cadaveric testing; anatomic anchor-edge repairs typically restore labral height to within 85–95% of the native, uninjured state, while medialized repairs may restore well under half of that. • Modern "double-pulley" and horizontal mattress suture configurations further improve labral height restoration compared with simple loop stitches by compressing a broader area of tissue against the bone bed.
When patient selection is appropriate — on-track lesion, subcritical glenoid bone loss, and absence of major risk factors (young age, contact/collision sport, hyperlaxity, high ISIS score) — arthroscopic Bankart repair achieves recurrence rates generally under 10%, with high patient satisfaction and return to sport.
However, outcomes deteriorate substantially outside this window: in young (<20 years), contact-sport athletes with even modest bone loss, recurrence after isolated arthroscopic repair has been reported from roughly 20% up to 70% in various series — this wide range reflects heterogeneity in bone loss severity, activity level, and surgical technique across studies, but consistently demonstrates that soft-tissue repair alone cannot compensate for a meaningful bony deficiency or an off-track Hill-Sachs lesion. This outcome gap is precisely what drives the bone-loss-based and track-based algorithms discussed in Stages 2–3, and what leads surgeons to add or substitute bony procedures in high-risk patients (Stage 5).
A repair that is technically flawless in suture tensioning can still fail clinically if the underlying bone loss was not accounted for — the glenoid track and critical-bone-loss frameworks exist precisely because labral repair addresses the soft-tissue lesion but cannot correct a bony deficiency on either side of the joint.
When glenoid bone loss exceeds the critical threshold, the lesion is off-track, or a prior isolated Bankart repair has already failed, bone-restoring or defect-filling procedures are required. The Latarjet procedure transfers the coracoid process with its attached conjoint tendon to the anteroinferior glenoid, while arthroscopic remplissage fills the Hill-Sachs defect itself — together these procedures convert a mechanically unstable bipolar bone loss pattern into a stable one.
Described by Michel Latarjet in 1954, the procedure transfers the coracoid process (with the conjoint tendon of biceps short head and coracobrachialis still attached) through a split in the subscapularis and fixes it flush to the anteroinferior glenoid neck with screws. Its stabilizing effect is often described as "triple-blocking":
1. Bone block effect: the transferred coracoid extends the glenoid articular arc anteriorly, directly restoring the bony containment that was lost — effectively reversing the inverted-pear deficit and, importantly, also extending the effective glenoid track width, which can convert an off-track Hill-Sachs lesion back to on-track without touching the humerus at all. 2. Sling (dynamic) effect: the conjoint tendon, running through and below the subscapularis, acts as a dynamic hammock that tightens across the anteroinferior joint line precisely when the arm is abducted and externally rotated — the exact position of instability risk — reinforcing the static bone block with a dynamic muscular restraint. 3. Capsuloligamentous effect: the remaining anteroinferior capsule/labral tissue and the coracoacromial ligament stump (left attached to the transferred coracoid) are typically repaired over the bone block, adding a soft-tissue restraint layer as well.
Because it directly restores bone stock and adds a dynamic sling, the Latarjet procedure is durable even in the presence of large glenoid bone loss where isolated soft-tissue repair would be expected to fail, and is generally considered the procedure of choice for glenoid bone loss above the critical threshold, for revision cases after a failed prior Bankart repair, and for many contact/collision athletes regardless of measured bone loss.
Remplissage (French for "filling") addresses the Hill-Sachs lesion directly rather than augmenting the glenoid: the posterior capsule and infraspinatus tendon are tenodesed into the Hill-Sachs defect using suture anchors, converting the intra-articular bony defect into an extra-articular, capsule-tendon-filled space. Mechanically, this accomplishes two things:
• It physically obliterates the defect's edge as a catching hazard, so the humeral head can no longer engage the glenoid rim through that defect. • It functions as a capsulodesis, limiting terminal external rotation slightly — a trade-off that is generally well tolerated in most patients but is weighed carefully in overhead athletes and throwers who depend on maximal external rotation.
Remplissage is most commonly added as a combined procedure alongside arthroscopic Bankart repair — "Bankart-plus-remplissage" — in patients with an off-track Hill-Sachs lesion but preserved or only subcritical glenoid bone stock, avoiding the greater surgical morbidity of a bone block procedure while still addressing the humeral-sided engagement mechanism. It is not a substitute for glenoid bone augmentation when the glenoid bone loss itself is the primary problem.
The dramatic difference in reported recurrence rates across the instability literature is best explained not by any single technique being inherently superior, but by whether the chosen procedure was matched to the underlying bipolar bone loss pattern:
• Isolated arthroscopic Bankart repair performed in a high-risk population — young age, contact/collision sport participation, and/or unrecognized significant glenoid bone loss or an off-track Hill-Sachs lesion — has produced recurrence rates reported anywhere from roughly 20% to 70% across published series, because the procedure does not address the bony mechanical deficiency driving the instability. • The identical arthroscopic soft-tissue technique, applied to a correctly selected on-track, subcritical-bone-loss population, produces recurrence rates under 10%. • Latarjet and remplissage, applied specifically to the high-risk/off-track/critical-bone-loss population they were designed for, also bring recurrence back down to under 10% in most contemporary series — despite being used in the objectively higher-risk patients.
The clinical lesson threaded through all five stages of this simulator is therefore one of matching, not hierarchy: accurately characterizing the Bankart lesion, quantifying glenoid bone loss, calculating on-track/off-track status, and only then selecting soft-tissue repair versus bone block versus remplissage (or a combination) is what separates a <10% recurrence outcome from a 20–70% one using otherwise equally valid techniques.
The Latarjet procedure and arthroscopic remplissage are not competitors to Bankart repair but complements to it — in current practice the majority of Latarjet and remplissage procedures are still performed together with anteroinferior labral/capsular repair, addressing the bony and soft-tissue components of instability simultaneously rather than treating bone loss and labral injury as separate problems.