What Happens to the Rotator Cuff in Reverse Shoulder Replacement?

In a reverse shoulder replacement, the rotator cuff is deliberately sidelined as the shoulder’s primary engine and replaced, biomechanically, by the deltoid muscle. The surgery flips the ball-and-socket anatomy so that a metal ball sits where the socket used to be, and a plastic cup goes on the upper arm bone. This geometric swap changes which muscles do the heavy lifting and reshapes the mechanical role of every remaining rotator cuff tendon. What happens to those tendons, whether they are repaired, ignored, or augmented with transfers, turns out to be one of the most debated questions in shoulder surgery today.

Why the Rotator Cuff Is No Longer in Charge

A normal shoulder relies on the rotator cuff to compress the humeral head into the socket while the deltoid raises the arm. When the cuff is torn beyond repair, that compression fails and the deltoid cannot lift the arm effectively. Reverse shoulder arthroplasty solves this by medializing and lowering the center of rotation, which lengthens the deltoid’s lever arm and lets it power the shoulder without needing the rotator cuff to hold things in place.¹ In effect, the prosthesis itself acts as the stabilizer that the cuff used to provide, creating a fixed fulcrum that the deltoid can work against. This is why reverse replacement is sometimes described as a deltoid-dependent procedure.²

But “sidelined” is not the same as “irrelevant.” The rotator cuff has four tendons, and while the design eliminates the need for the cuff to compress the joint, each tendon still contributes to rotation and fine motor control. The surgical team has to make decisions about every cuff tendon it encounters, and those decisions affect what the shoulder can and cannot do afterward.

The Subscapularis Debate

The subscapularis is the large tendon at the front of the rotator cuff. During surgery, it is almost always detached to gain access to the joint. Whether or not to reattach it afterward has generated years of conflicting research. On one side, the logic seems straightforward: repairing a tendon should be better than leaving it loose. On the other, the prosthesis is designed to work without the cuff, and a repair adds surgical time and imposes restrictions during early recovery.

One meta-analysis found that repairing the subscapularis was associated with much lower dislocation rates compared to leaving it unrepaired. But the same study revealed that implant design mattered enormously: when the center of rotation was lateralized rather than medialized, dislocation rates dropped just as dramatically even without a subscapularis repair.³ This suggests that the subscapularis repair was partly compensating for a design limitation in older-style implants rather than being universally necessary.

Two later systematic reviews reached even more muted conclusions. One pooled seven studies and found that dislocation occurred in about 0.7% of patients with subscapularis repair and about 1.5% without it, a difference that was not statistically significant.⁴ Another meta-analysis echoed this, finding no significant difference in dislocation risk, and concluded that subscapularis repair produces no clinically meaningful benefits, particularly with lateralized implant designs.⁵

Where subscapularis repair does seem to help is internal rotation, the motion you use to tuck in a shirt or reach behind your back. Patients whose subscapularis was repaired showed modestly better internal rotation after surgery, while those without repair tended to have slightly better external rotation.⁶ This trade-off makes sense anatomically: a functioning subscapularis pulls the arm inward, and without it, external rotators face less opposition.

Does the Subscapularis Actually Heal?

Even when a surgeon reattaches the subscapularis, the tendon does not always stay attached. Ultrasound studies paint a mixed picture. One study using a standard single-row repair technique found that only about 53% of repaired subscapularis tendons were still intact on ultrasound afterward. Patients with intact repairs had significantly better internal rotation, though overall clinical scores between the healed and failed groups were similar.⁷ A separate study using a more robust double-row repair technique achieved a higher healing rate of about 83%.⁸

The fact that many subscapularis repairs fail without obvious clinical consequences reinforces the broader point: in a reverse replacement, the prosthesis carries the stabilizing load that the subscapularis would normally share. A healed repair is a bonus for internal rotation, but a failed repair is not the catastrophe it would be in a conventional shoulder replacement.

The Posterior Cuff and External Rotation

If the subscapularis debate centers on stability and internal rotation, the posterior rotator cuff conversation revolves around external rotation: the motion of turning your arm outward, reaching to the side, or bringing a cup to your mouth. The infraspinatus and teres minor are the two posterior cuff muscles responsible for this movement, and their condition before surgery is one of the strongest predictors of how well external rotation recovers afterward.

A systematic review found that fatty infiltration of the teres minor and infraspinatus, where muscle tissue is gradually replaced by fat, negatively affects both patient-reported outcomes and external rotation recovery after reverse replacement.⁹ Patients walking into surgery with severe posterior cuff degeneration tend to walk out with less ability to rotate the arm outward, regardless of how well the deltoid-driven elevation works.

However, the story changes with implant design. A study specifically looking at lateralized glenoid implants found that neither infraspinatus nor teres minor fatty infiltration negatively influenced range of motion or patient-reported outcomes after about 15 months of follow-up.¹⁰ A more recent study of lateralized humeral implants reached a similar conclusion: patients with severe infraspinatus fatty infiltration achieved the same range of motion and clinical scores as those with mild infiltration, though their absolute strength in forward elevation and external rotation was lower.¹¹

This pattern, where lateralized designs seem to buffer the impact of posterior cuff degeneration, likely comes down to tension. When the humerus sits further out from the body, the remaining posterior cuff fibers and the posterior deltoid are stretched to a more functional length, allowing them to contribute more to rotation even when the muscles are partially degenerated.

How Implant Design Changes the Cuff’s Role

The tension relationship between the implant and the remaining rotator cuff is one of the most important and least intuitive aspects of reverse replacement. The original Grammont-style design placed the center of rotation right at the bone surface, which was brilliant for deltoid efficiency but left the remaining cuff tendons slack and mechanically disadvantaged. Newer lateralized designs push the humerus outward, which tightens the cuff tendons and gives them more mechanical leverage.

But this tightening is a double-edged sword. A biomechanical study found that repairing the rotator cuff in combination with a lateralized glenosphere actually creates an antagonistic effect, increasing the forces on both the deltoid and the joint itself.¹² In plain terms, when you tension the cuff more aggressively through implant design and then add a tight surgical repair on top of that, the muscles end up fighting each other rather than cooperating. This is one reason some surgeons opt not to repair the subscapularis with lateralized implants: the prosthesis is already doing the tensioning job that a repair would provide, and stacking both together may do more harm than good.

The clinical implication is that there is no universal “best” approach to the rotator cuff in reverse replacement. A medialized design with a torn subscapularis may benefit from repair to reduce dislocation risk. A lateralized design may do just as well without it. A patient with a functioning teres minor may get good external rotation with no additional intervention, while one with a completely degenerated posterior cuff may need a tendon transfer. The implant and the patient’s remaining cuff tissue form a system, and what you do with one dictates what you need from the other.

Tendon Transfers When the Posterior Cuff Is Gone

Some patients arrive at surgery with essentially no functioning posterior rotator cuff at all. The infraspinatus is gone and the teres minor is atrophied or absent. In these cases, the reverse replacement restores the ability to lift the arm overhead, courtesy of the deltoid, but external rotation remains severely limited. This combination of restored elevation and absent rotation has a name: the “hornblower’s sign,” where a patient can raise the arm but cannot rotate it outward to bring a hand to the face without hiking the elbow above the shoulder.

For these patients, surgeons sometimes combine the reverse replacement with a tendon transfer. The most studied approach involves rerouting the latissimus dorsi and teres major tendons from their normal position on the front of the arm to the back, turning them into substitute external rotators. In a prospective series of 11 patients with complete loss of both elevation and external rotation, this combined procedure improved active elevation from an average of 70 degrees to 148 degrees and external rotation from negative 18 degrees to positive 18 degrees.¹³ A separate series of 15 patients confirmed that the combined procedure restored both active elevation and external rotation in cases of massive cuff tears that included the teres minor.¹⁴

Tendon transfers are not routine. They add complexity, require longer rehabilitation, and depend on the transferred muscles being healthy enough to function in their new role. But for the subset of patients with a truly devastated posterior cuff, they represent the only current option for restoring meaningful external rotation.

Compensatory Muscles and New Activation Patterns

When the rotator cuff is bypassed or absent, other muscles step in. An electromyography study measured which muscles actually fire during external rotation after reverse replacement and found a surprising pattern. During external rotation beyond the neutral position, the teres major initiated the movement, followed by the teres minor and posterior deltoid, with all three contributing roughly equally.¹⁵ The teres major is not traditionally thought of as an external rotator; its recruitment in this role reflects the neuromuscular adaptation that occurs when the shoulder’s geometry is fundamentally altered.

There is also evidence that the teres minor can hypertrophy, or enlarge, as a compensatory response when the infraspinatus fails. In patients with advanced cuff tears heading toward surgery, teres minor hypertrophy has been found to correlate with infraspinatus tear and atrophy, suggesting the body attempts to compensate before and possibly after surgical intervention.¹⁶

How the Scapula Compensates

The shoulder is not just a ball and socket. It is a chain of joints, and the joint between the shoulder blade and the rib cage, the scapulothoracic articulation, plays a much larger role after a reverse replacement than it does in a normal shoulder. Studies consistently show that after reverse replacement, patients rely more on scapular rotation and less on glenohumeral motion to raise the arm.¹⁷

One motion analysis study found that the ratio of humeral-to-scapular movement in reverse replacement patients stayed relatively flat throughout abduction, around 2.4 to 2.8, in contrast to healthy shoulders where the scapula contributes less at the beginning of the movement and progressively more as the arm rises higher.¹⁸ In practical terms, the shoulder blade starts rotating from the very beginning of arm elevation rather than kicking in partway through, as it does normally. Another study was more striking still: four out of ten patients who underwent reverse replacement performed a weighted lifting task using almost exclusively scapulothoracic motion, with very little contribution from the ball-and-socket joint itself.¹⁹

This scapular compensation is both a feature and a vulnerability. It allows patients to raise the arm even when the glenohumeral joint’s range is restricted, but it also places greater demands on the muscles that control the shoulder blade, particularly the trapezius and serratus anterior. This has implications for both rehabilitation and complications.

Complications Linked to Cuff and Deltoid Mechanics

One of the unique complications of reverse replacement is scapular notching, where the humeral component mechanically impinges on the scapular neck during arm movements, gradually eroding bone. This is more common with medialized designs and has been linked to implant positioning factors like the angle between the glenosphere and the humerus and the amount of inferior overhang of the baseplate.²⁰ While notching does not always cause symptoms, severe cases can loosen the implant over time.

More acutely, acromial and scapular spine fractures are a recognized risk. These stress fractures occur because the deltoid, now working harder as the shoulder’s primary mover, pulls more forcefully on its attachment points. Contributing factors include poor bone quality (common in the elderly population that receives most reverse replacements), excessive deltoid tension from overtightening the construct, and technical factors like screw placement. Prevention strategies focus on optimizing bone health and avoiding excessive deltoid tension during implant placement.²¹

What Rehabilitation Looks Like Without a Functional Cuff

Rehabilitation after reverse replacement is fundamentally different from rehab after a conventional replacement precisely because the rotator cuff is no longer the star. The focus shifts almost entirely to the deltoid and the scapular stabilizers. A rehabilitation framework for cuff-deficient shoulders emphasizes deltoid loading before and after surgery, with exercises performed in the scapular plane and at higher abduction angles, where the deltoid and periscapular muscles activate most efficiently.²²

The early postoperative period typically involves immobilization in a sling, but the specifics depend partly on what was done to the subscapularis. If it was repaired, internal rotation is restricted for several weeks to protect the repair. If it was not, mobilization can proceed more aggressively. Either way, active-assisted range of motion usually begins within the first few weeks, progressing to active motion and then strengthening over three to six months. Scapular control exercises are increasingly recognized as critical given the heightened reliance on scapulothoracic motion described above.

Proprioception After the Cuff Is Bypassed

The rotator cuff does more than move the arm. Its tendons are packed with nerve endings that sense joint position, contributing to proprioception, your awareness of where your arm is in space without looking at it. Removing or bypassing the cuff raises the question of whether proprioception suffers.

The answer appears to depend on why the replacement was done. One study of patients who received a reverse replacement for cuff tear arthropathy found that their operated shoulder actually had better joint position sense than their non-operated side.²³ This counterintuitive result likely reflects the fact that the non-operated shoulder was also diseased, and the surgery restored enough mechanical stability for the remaining sensory receptors to function effectively. In contrast, a study of patients who received a reverse replacement for complex upper-arm fractures found significantly worse proprioception in the operated shoulder compared to the healthy opposite side.²⁴ The difference makes sense: fracture patients lost healthy tissue suddenly rather than gradually adapting over years, and the trauma itself may damage sensory nerve endings that chronic cuff disease leaves partially intact.

For patients, the practical takeaway is that proprioceptive training, exercises that challenge the brain to track joint position through different planes, belongs in the rehabilitation program. The degree of proprioceptive recovery varies, but it is not a lost cause simply because the cuff is no longer mechanically active.