Arthroscopic repair of supraspinatus/infraspinatus tears — tear classification, suture-anchor construct biomechanics, augmentation, and tendon-to-bone healing
Before any arthroscopic instrument enters the joint, the operative plan is dictated by four independent classification systems assessed on MRI and confirmed at arthroscopy: tear thickness (partial vs. full), tear size (linear dimension in the coronal and sagittal planes), tear pattern (crescent, U-shaped, or L-shaped), and the Goutallier grade of fatty infiltration of the muscle belly — the single strongest predictor of reparability and postoperative function.
Partial-thickness rotator cuff tears involve incomplete disruption of the tendon and are further subdivided by location:
• Bursal-side tears: begin on the outer (bursal) surface of the tendon, often associated with subacromial impingement and bursal inflammation; tend to propagate with continued mechanical abrasion against the acromion • Articular-side tears: begin on the deep (articular) surface, the most common partial-thickness pattern (roughly 2–3× more frequent than bursal-side); associated with internal impingement and posterosuperior glenoid contact in overhead athletes • Intratendinous (interstitial) tears: horizontal cleavage within the substance of the tendon with intact bursal and articular surfaces — easily missed on standard MRI and at arthroscopy unless probed
Partial tears are graded by depth using the Ellman classification: Grade 1 (<3mm, <25% thickness), Grade 2 (3–6mm, 25–50%), Grade 3 (>6mm, >50% thickness). Grade 3 partial tears (>50% thickness) are generally treated as functional full-thickness tears and converted to complete tears intraoperatively before repair, since debridement alone leaves a mechanically incompetent residual tendon.
Full-thickness tears communicate the subacromial bursa with the glenohumeral joint and are the tears addressed by the bulk of anchor-based repair techniques discussed in this simulator.
Tear size is measured as the maximum tendon defect dimension (typically anteroposterior) and drives the classic DeOrio and Cofield categories:
• Small: <1 cm — usually supraspinatus only, minimal retraction, straightforward primary repair • Medium: 1–3 cm — full supraspinatus, may involve anterior infraspinatus fibers • Large: 3–5 cm — two-tendon involvement common, moderate retraction, tension at repair increases substantially • Massive: >5 cm OR full-thickness involvement of ≥2 tendons (supraspinatus + infraspinatus, or supraspinatus + subscapularis) — the category most associated with irreparability, poor tissue quality, and need for augmentation
Tear pattern describes the geometry of the defect and dictates mobilization strategy:
• Crescent-shaped: the most common pattern; the tendon edge retracts minimally medially, mobilizes directly laterally to the footprint with low tension — straightforward direct repair • U-shaped: the tear apex retracts far medially while the anterior and posterior tendon margins remain relatively fixed near the greater tuberosity — direct lateral mobilization of the apex creates excessive tension; margin convergence (side-to-side sutures closing the U before anchor fixation) is required • L-shaped (or reverse-L): an asymmetric tear combining a longitudinal split with a transverse component, requiring rotational mobilization of one tendon leaf to restore normal anatomy before fixation
Goutallier classification (graded on oblique-sagittal MRI or CT through the scapular Y-view, at the level of the scapular spine):
• Grade 0: normal muscle, no fatty streaks • Grade 1: some fatty streaks within the muscle belly • Grade 2: fatty infiltration present but less fat than muscle • Grade 3: fat and muscle roughly equal in volume • Grade 4: more fat than muscle
Fatty infiltration is largely irreversible once established — muscle atrophy and fat replacement do not regress even after anatomically successful repair. Grade ≥3 fatty infiltration is a strong independent predictor of retear and poor functional outcome, and is a key criterion pushing the surgeon toward augmentation or alternative reconstructive strategies (Stage 4) rather than isolated primary repair.
Patte classification stages tendon retraction relative to the glenoid on coronal MRI:
• Stage 1: tendon stump retracted to near its normal footprint insertion • Stage 2: stump retracted to the level of the humeral head • Stage 3: stump retracted medially to the glenoid rim
Greater retraction correlates with longer tear chronicity, higher fatty infiltration grade, and reduced tendon excursion/mobility — all of which compound to reduce the probability of a durable, tension-free repair.
Three families of arthroscopic anchor constructs are used to re-approximate torn tendon to the greater tuberosity footprint. They differ primarily in how many rows of anchors are placed and whether the sutures compress the tendon directly against bone across the entire footprint width or merely tack the tendon edge to a single line of fixation — a difference with measurable consequences for contact area, contact pressure, and ultimate load to failure.
Single-row repair places a single line of suture anchors (typically 2–4) directly at the lateral edge of the footprint, along the articular margin of the greater tuberosity. Each anchor carries sutures passed through the tendon edge as simple or mattress stitches, pulling the tendon down to a linear strip of bone.
Advantages: technically simpler, faster to perform, lower implant cost, fewer anchors, and less risk of over-tensioning the tendon medially. For small, mobile, crescent-shaped tears, single-row repair achieves excellent clinical outcomes with lower cost and operative time.
Limitation: the repaired tendon contacts only the lateral strip of the anatomic footprint rather than the full medial-to-lateral insertional area, restoring roughly 60–70% of native footprint contact area. Reduced contact area translates to a smaller surface for tendon-to-bone healing and a lower initial construct stiffness, contributing to higher retear rates in medium-to-large tears where the biomechanical margin for error is smaller.
Double-row repair adds a second, medial row of anchors placed near the articular cartilage margin (chondral edge), in addition to the standard lateral row near the tuberosity edge. Medial-row sutures are passed through the tendon first and tied, then lateral-row anchors (often knotless) capture the same sutures or independent limbs to re-tension the construct laterally.
The medial row re-creates the medial extent of the anatomic footprint, while the lateral row restores the lateral extent — together restoring 80–90% of native footprint contact area. This translates biomechanically into higher initial fixation strength, greater construct stiffness, and improved resistance to gap formation under cyclic loading compared with single-row repair, at the cost of more anchors, longer operative time, and higher implant cost.
Suture-bridge repair — also termed transosseous-equivalent (TOE) repair — is a variant of double-row fixation in which the medial-row sutures, after being tied, are passed laterally over the top of the tendon and secured under tension by knotless lateral-row anchors. This creates horizontal mattress "bridges" of suture that compress the entire width of the tendon against the footprint, mimicking the compressive effect of a traditional open transosseous tunnel repair without requiring bone tunnels.
This compressive bridging maximizes tendon-bone contact area (>90% of native footprint) and contact pressure, and produces the highest reported load-to-failure and construct stiffness among the three techniques in biomechanical cadaveric testing. Suture-bridge fixation is considered the current biomechanical gold standard for full-thickness tears amenable to primary repair, particularly medium-to-large tears where footprint restoration meaningfully affects healing potential — though single- and double-row repairs remain entirely appropriate for well-selected small tears.
A repair construct is only as strong as its weakest link in the chain from bone to anchor to suture to tendon. Understanding anchor pullout mechanics, knot versus knotless fixation, and the strategies used to reduce repair-site tension — margin convergence and medialization — explains why two repairs of the same tear, performed with the same implants, can have very different failure rates.
Suture anchors fail by one of two mechanisms: suture breakage/slippage at the eyelet, or anchor pullout from bone. Pullout strength depends on anchor design (screw-in vs. push-in, threaded vs. barbed), anchor diameter, insertion angle (ideally 45° deadman angle to maximize resistance to pullout), and — critically — local bone mineral density at the greater tuberosity.
Reported pullout strengths for modern suture anchors in cadaveric humeral bone range roughly 80–200 N, but this value falls sharply in osteoporotic or elderly bone, and in revision surgery where prior anchor tunnels have already compromised the local bone stock. Because in vivo repair-site forces during early active motion and simple activities of daily living can approach or exceed this range, anchor pullout — not suture breakage — is frequently the limiting factor for construct failure in poor-quality bone, favoring techniques that distribute load across more anchors (double-row, suture-bridge) rather than concentrating all forces on 2–3 single-row anchors.
Traditional knotted anchors rely on the surgeon tying a secure arthroscopic knot (typically a sliding-locking knot backed by three or more reversing half-hitches) to maintain suture tension after tendon-to-bone approximation. Knot security depends on suture material (braided sutures hold knots better than monofilament), number of throws, and knot configuration — a poorly tied or under-tensioned knot is a recognized mode of early construct failure independent of anchor or bone quality.
Knotless anchors eliminate this variable by mechanically locking the suture within the anchor itself (interference screw, cam, or barbed mechanism) under surgeon-controlled tension at the time of insertion, and are especially favored for lateral-row fixation in suture-bridge constructs where multiple suture limbs must be tensioned simultaneously. Comparative biomechanical and clinical studies show knotless and well-tied knotted constructs achieve comparable failure loads and retear rates — the decisive factor is meticulous surgical technique and adequate initial tension, not the presence or absence of a knot per se.
The tension-free repair principle holds that the single strongest modifiable predictor of healing is minimizing the force required to hold the tendon edge at the footprint — high repair-site tension increases gap formation under cyclic load, reduces vascular perfusion at the tendon-bone interface, and is one of the most consistent independent risk factors for retear across the literature.
Margin convergence: for U-shaped tears, the medially retracted apex is under far greater tension than the anterior/posterior margins if pulled directly laterally to the footprint. Side-to-side (margin convergence) sutures are placed first, approximating the free tear margins to each other and converting the U-shape toward a crescent — redistributing load through the tendon substance itself (which tolerates tension well) rather than concentrating it at the anchor-tendon interface. Only after margin convergence is the residual, now much smaller, defect fixed to bone.
Medialization: for larger or chronically retracted tears where the tendon cannot reach the native lateral footprint edge without excessive tension, the surgeon may intentionally fix the tendon to a more medial point on the footprint (partial footprint restoration) rather than force it fully laterally. This sacrifices some footprint contact area and biomechanical restoration in exchange for a genuinely tension-free construct — an explicit trade-off, since a partially medialized but tension-free repair reliably outperforms a fully lateralized but overtensioned one.
When a tear is massive, chronically retracted, or accompanied by advanced (Goutallier grade ≥3) fatty infiltration, primary suture-anchor repair alone often cannot achieve durable healing. A spectrum of augmentation and salvage options exists, ranging from biologic and mechanical reinforcement of a repairable tear to complete reconstructive or arthroplasty solutions for tears deemed irreparable.
For massive or high-risk tears where the native tendon is repaired but tissue quality or tension raises concern for failure, a patch graft can be applied over (onlay) or interposed within the repair to mechanically reinforce it and provide a scaffold for cellular ingrowth. Two broad categories are used:
• Dermal allograft (acellular human or porcine dermal matrix): a decellularized collagen scaffold that is biologically incorporated over months, providing mechanical support during the vulnerable early healing window while gradually being remodeled by host tissue • Synthetic patches: non-degradable or slowly degradable polymer meshes (e.g., polyester, polypropylene-based) providing purely mechanical reinforcement without expectation of true biologic incorporation
Clinical series and meta-analyses of patch-augmented repair in massive tears report meaningful reductions in retear rate compared with repair alone — commonly cited on the order of a 20–30 percentage-point reduction — though heterogeneity in patch material, tear severity, and surgical technique across studies limits precise generalization.
Superior capsule reconstruction addresses massive, irreparable posterosuperior rotator cuff tears (supraspinatus and infraspinatus) in patients who are not yet candidates for reverse total shoulder arthroplasty — typically younger, higher-demand patients without advanced glenohumeral arthritis. A graft (fascia lata autograft or dermal allograft) is fixed medially to the superior glenoid and laterally to the greater tuberosity footprint, substituting for the absent superior cuff tendon.
The reconstructed superior capsule restores the critical function of superior humeral head stabilization — resisting proximal migration of the humeral head against the deltoid's upward pull — thereby re-establishing a stable fulcrum for the remaining functional rotator cuff and deltoid to generate elevation, even though the graft itself is not a contractile tendon. Graft thickness (≥8 mm autograft associated with better structural outcomes than thin allograft) and secure medial/lateral fixation are key technical determinants of graft healing and clinical success.
Biologic augmentation: platelet-rich plasma (PRP) and bone marrow aspirate concentrate (BMAC) have been investigated as adjuncts applied at the repair site to deliver growth factors (PDGF, TGF-β, VEGF) and mesenchymal progenitor cells intended to enhance the biology of tendon-bone healing. Evidence remains mixed — some randomized trials and meta-analyses show modest reductions in retear rate, particularly for high-risk repairs, while others show no significant benefit; PRP/BMAC is best considered an adjunct of uncertain but plausible incremental benefit rather than a reliable stand-alone solution.
Partial repair vs. debridement-only: when a tear is too large or retracted to fully cover the footprint, a partial repair restoring the "force couple" balance between anterior (subscapularis) and posterior (infraspinatus/teres minor) cuff forces can improve function even without complete footprint restoration. Debridement alone (removing frayed tissue and treating associated bursitis or biceps pathology without attempting tendon fixation) is reserved for irreparable tears in low-demand patients, providing pain relief without restoring cuff mechanics.
Reverse total shoulder arthroplasty (rTSA): for older, lower-demand patients with massive irreparable tears and pseudoparalysis (inability to actively elevate the arm above 90° despite passive motion being preserved), rTSA reliably restores active elevation by medializing and lowering the glenohumeral center of rotation, allowing the deltoid to substitute for the absent rotator cuff. rTSA is generally favored over SCR or repair salvage once glenohumeral arthritis or established pseudoparalysis is present, given its more predictable functional outcome in this population.
Even a technically perfect repair with maximal footprint coverage and high initial fixation strength must survive a biological healing process that never fully recreates native anatomy. Understanding why the healed interface is mechanically inferior to the native enthesis — and which patient and surgical factors predict retear — is essential to interpreting outcomes and designing a rehabilitation protocol that protects healing tissue without causing stiffness.
The native rotator cuff insertion is a specialized fibrocartilaginous enthesis with four histologic zones — tendon, uncalcified fibrocartilage, calcified fibrocartilage, and bone — that gradually transitions mechanical stiffness from compliant tendon to rigid bone, distributing stress and minimizing the stress-concentration effects of an abrupt tendon-bone junction.
After surgical repair, healing does not regenerate this graded fibrocartilaginous transition. Instead, the tendon-bone interface heals by formation of disorganized fibrovascular scar tissue, with collagen fibers oriented less consistently than native Sharpey-like fibers and no reconstitution of the tidemark between calcified and uncalcified fibrocartilage. This scar-mediated healing is mechanically inferior to the native enthesis — lower failure load, lower stiffness, and a persistent zone of relative weakness — even in structurally intact, clinically successful repairs at long-term follow-up. This biological ceiling is a central reason why footprint coverage and repair-site tension (Stages 2–3) matter so much: maximizing initial mechanical fixation partially compensates for the fact that biological healing will never fully restore native tissue quality.
Multivariate analyses across large cohort studies consistently identify a recurring set of independent retear risk factors:
• Tear size: the single strongest structural predictor — larger tears retear more often regardless of technique • Chronicity: long-standing tears have more retracted, atrophic, fatty-infiltrated tissue with reduced healing capacity compared with acute tears repaired early • Fatty infiltration grade (Goutallier ≥3): reflects both reduced tissue quality and, often, tear chronicity; strongly associated with retear and inferior functional recovery even when structural healing occurs • Patient age: older age (commonly cited threshold >65–70 years) correlates with lower intrinsic tendon healing capacity and higher retear rates independent of tear size • Diabetes mellitus: impairs collagen cross-linking and microvascular healing at the tendon-bone interface • Smoking: nicotine-mediated vasoconstriction reduces perfusion to the healing repair site and is a modifiable risk factor patients should be counseled to address perioperatively • Repair-site tension: as established in Stage 3, higher tension at fixation is directly associated with gap formation and retear
Reported retear rates vary substantially with tear size and technique: small tears retear in roughly 10–20% of cases across most series, while massive tears show a wide reported range of roughly 40–90% depending on technique, tissue quality, and patient risk-factor burden — the wide range itself reflecting how strongly these compounding risk factors, more than any single technical choice, determine outcome in the massive-tear population.
Rehabilitation after arthroscopic rotator cuff repair follows a graduated, biology-informed timeline designed to protect the healing tendon-bone interface during its most vulnerable early weeks while avoiding the opposite complication of postoperative stiffness (adhesive capsulitis):
• Weeks 0–6 (immobilization/protection phase): the arm is maintained in a sling or abduction brace; only passive range-of-motion exercises are performed by a therapist or via pendulum/pulley exercises, avoiding any active contraction of the repaired tendon while the fibrovascular healing response is at its weakest and most vulnerable to gap formation under load • Weeks 6–12 (passive-to-active-assisted transition): sling is discontinued; passive ROM is progressed toward full range, and active-assisted motion begins as the repair gains early mechanical integrity; the repair remains too weak for resisted strengthening • Weeks 12–16+ (active ROM and early strengthening): active range of motion without resistance is emphasized first, restoring neuromuscular control before load is added • Months 4–6 (progressive strengthening): resistance training of the rotator cuff and periscapular musculature begins and progresses gradually, with return to full unrestricted activity (including overhead sport) typically not permitted before 6 months
This timeline is deliberately conservative relative to many other orthopedic soft-tissue repairs because tendon-to-bone healing is slower and mechanically weaker than tendon-to-tendon or bone-to-bone healing; larger tears, patch-augmented repairs, and patients with additional retear risk factors are frequently protected with more conservative (slower) protocols, while very small, low-tension repairs may progress somewhat faster — always balancing the risk of retear from premature loading against the risk of permanent stiffness from excessive immobilization.
The central biomechanical thesis of modern rotator cuff surgery: initial fixation strength (anchor construct, footprint coverage, tension-free technique) and biological healing capacity (tissue quality, vascularity, patient factors) are two independent, additive determinants of success. A biomechanically excellent suture-bridge repair cannot overcome Goutallier grade 4 fatty infiltration in a heavy smoker — and conversely, ideal tissue biology cannot fully compensate for a technically overtensioned, poorly covered repair construct.