What Is a Reverse Total Shoulder Replacement

A reverse total shoulder replacement swaps the ball and socket of the shoulder joint. A metal ball goes on the shoulder blade, held in place with a baseplate and screws, and a stem and socket go into the arm bone. This lets the deltoid muscle lift the arm when the rotator cuff no longer can.

• What is it? The formal name is reverse total shoulder arthroplasty. “Arthroplasty” means joint replacement, and “reverse” means the ball and socket trade places compared with a normal shoulder.
• Who is it for? It was built for shoulders with both arthritis and a torn, unrepairable rotator cuff. It is now also used for fractures, failed prior replacements, and some cases of arthritis with an intact cuff.
• How well does it work, and how long does it last? It reliably relieves pain and restores the ability to raise the arm overhead. About nine in ten implants are still in place ten years later.

How is a reverse shoulder different from a normal shoulder?

A normal shoulder is a round ball on top of the arm bone (the humerus) sitting against a shallow socket on the shoulder blade (the glenoid). Rather than being a true ball and socket, the glenohumeral joint is more like a golf ball on a tee. The joint depends on the rotator cuff, a group of four tendons that hold the ball centered while the larger deltoid muscle lifts the arm.

A standard (or anatomic) shoulder replacement keeps that arrangement. It replaces the worn ball and socket with implants of the same shape. It still needs a working rotator cuff to function.

A reverse replacement turns the design around. A metal half-sphere called the glenosphere is fixed to the shoulder blade. A stem is placed down the arm bone, and a plastic cup on top of it fits over the ball.

Why does it work without a rotator cuff?

When the rotator cuff is torn beyond repair, the ball doesn’t stay centered on the tee. The deltoid pulls, but the arm does not lift. Some patients cannot raise the arm at all even though the deltoid is healthy. This is called pseudoparalysis.

The reverse design fixes the problem mechanically. The center of rotation of the shoulder moves inward, closer to the body, and slightly lower. That does two things:

• The ball-and-socket joint is inherently stable. The socket is deeper than the normal “golf tee” glenoid and fits over the ball, so the joint stays centered without a rotator cuff holding it there.
• The deltoid gets better leverage. Moving the center of rotation lengthens the deltoid’s lever arm, the distance between the pivot and where the muscle pulls. Lowering the arm puts the muscle under more tension. The same muscle now lifts the arm effectively.

A crowbar is a good comparison. Move the pivot point, and the same effort moves a heavier load.

The design depends on a working deltoid. A shoulder with a badly damaged deltoid or the nerve that powers it is a poor candidate.

Who needs a reverse shoulder replacement?

The most common reasons are:

• Cuff tear arthropathy. This is arthritis that develops after a long-standing, massive rotator cuff tear. It was the original reason the operation was designed.
• A massive rotator cuff tear that cannot be repaired. This applies especially when the patient has lost the ability to raise the arm, even without advanced arthritis.
• Severe fractures of the top of the arm bone in older adults. The data shows patients with severe proximal humerus fractures do very well with a low complication rate and faster pain relief when treated with a reverse shoulder replacement.
• A failed prior shoulder replacement or fracture repair. Revision surgery is more complex and carries higher risk than a first-time replacement.
• Arthritis with an intact rotator cuff. Some surgeons use a reverse here, particularly in older patients or shoulders with significant bone loss in the socket. In a matched comparison of 134 patients, pain relief and function scores two to three years after surgery were similar with reverse and anatomic replacements.

The use of the reverse shoulder replacement has grown quickly. By 2020, reverse replacements made up nearly 70% of first-time shoulder replacements in the United States.

What will the shoulder be able to do?

The reverse is best at two things: relieving pain and restoring the ability to raise the arm overhead. Most patients regain elevation above shoulder height.

Rotation is less predictable.

• Reaching behind the back is often limited. Tucking in a shirt, reaching a back pocket, or fastening a bra can remain difficult.
• Turning the arm outward is also often limited, especially when the small rotator muscle at the back of the shoulder (the teres minor) is torn or has turned to fat.

Most people return to recreational activity. A review of 19 studies covering more than 3,000 patients found that about 85% returned to sport, and roughly 70% returned at the same level or higher. Lower-demand activities such as swimming, fitness, and golf are the most common.

How have reverse shoulder designs changed?

The original Grammont design placed the pivot point very far inward. It worked, but it had two drawbacks. The plastic cup could rub against the edge of the shoulder blade when the arm was at the side, and turning the arm outward was limited.

Newer designs shift the pivot point outward. These are called lateralized designs. Studies comparing the two approaches report similar pain relief and overall function scores. Lateralized designs preserve more rotation and show much less of the bone wear seen on X-rays. No single design has been shown to be best for every patient.

What are the risks?

Most patients do well, and serious problems are uncommon. The complications specific to a reverse replacement are:

Complication	What it is	How common	What it means for you
Dislocation	The ball comes out of the socket	Fewer than 2 in 100 first-time replacements; about 6 in 100 revisions	Most happen in the first two months. It usually needs a return to the operating room.
Acromial stress fracture	A crack in the bony roof of the shoulder (the acromion), from the extra pull of the deltoid	About 3 in 100 across 90 studies; higher in some series	New pain on top of the shoulder in the months after surgery. Function is often worse afterward, even when the bone heals.
Scapular notching	Wear on the edge of the shoulder blade where the cup rubs	Common on X-rays with the original inward design; much less common with lateralized designs	Usually painless and seen only on X-rays. It is a reason designs have evolved.
Infection	Bacteria settling on the implant	Uncommon	Covered in detail in Can a Shoulder Replacement Get Infected?

Dislocation is more common after revision surgery and in men. Acromial stress fractures are more common in patients with inflammatory arthritis, such as rheumatoid arthritis.

How long does a reverse shoulder replacement last?

Reverse replacements have been used widely for about two decades in the US, so long-term data is still limited. A review of seven studies following patients for at least ten years found that about 88% of implants were still in place without revision surgery at ten years.

Two groups don’t do as well. Implants placed during revision surgery fail more often than first-time replacements. Patients younger than 60 also have lower ten-year survival than older patients, likely because they place higher demands on the shoulder for more years.

Function tends to hold up over time, though some decline appears in longer follow-up. Part of that decline reflects aging itself.

What is recovery like?

Most patients go home the same day; some spend one night in the hospital. The arm usually rests in a sling for comfort and protection in the early weeks. Surgeons differ on how soon motion begins.

A randomized trial compared starting therapy immediately with keeping the arm immobilized for six weeks. Both groups gained similar motion and function by one year, and early motion did not cause harm. Your surgeon will choose a timeline based on your shoulder, the bone quality, and the reason for surgery.

Most daily activities return within a few months. Continued gains in strength and comfort are common through the first year.

Dr. Patel has his patients wear a sling for 2 weeks (his full reverse total shoulder therapy protocol is available to download). After that, you begin immediate range of motion and return to normal daily activities (getting dressed, carrying a coffee mug). No golf or higher impact activities until 3 months postoperatively.

---

Midhat Patel, MD is a fellowship-trained shoulder and elbow surgeon at Banner Health in Phoenix and Scottsdale, Arizona, and a Clinical Assistant Professor of Orthopaedic Surgery at the University of Arizona College of Medicine – Phoenix.

Sources

1. Design principles, deltoid mechanics, rotational limits, notching, original indications. Boileau P, et al. Grammont reverse prosthesis: design, rationale, and biomechanics. J Shoulder Elbow Surg. 2005. Narrative review. Medialized, distalized center of rotation increases deltoid recruitment and tension; restores active elevation above 90°. External rotation often limited, particularly with absent or fatty teres minor; internal rotation rarely restored. Inadequate deltoid tension associated with instability; notching from cup impingement on scapular neck.
2. Lateralized versus medialized designs: rotation and notching. Berton A, et al. Medialized versus lateralized center of rotation in reverse total shoulder arthroplasty: a systematic review and meta-analysis. J Clin Med. 2021. 24 studies in qualitative synthesis, 19 in meta-analysis. External rotation gain with arm at side 20.4° lateralized vs 8.3° medialized; scapular notching 6.6% vs 47.7%.
3. Lateralized versus medialized designs: outcome scores similar. Helmkamp JK, et al. The clinical and radiographic impact of center of rotation lateralization in reverse shoulder arthroplasty: a systematic review. J Shoulder Elbow Surg. 2018. 18 studies. No clear difference in outcome scores; lateralized COR associated with greater ROM and less notching; no other clear complication differences.
4. Lateralized versus Grammont design: prospective comparison. Freislederer F, et al. Lateralized vs classic Grammont style reverse shoulder arthroplasty for cuff deficiency Hamada stage 1–3: does the design make a difference? J Shoulder Elbow Surg. 2021. Prospective comparative cohort, n = 67 (44 lateralized, 23 Grammont), 24-month follow-up. Constant and SPADI not different (P ≥ .654). External rotation higher with lateralized design (P = .012); notching 2.9 times higher with Grammont (P = .001). Reaching L3: 70% vs 48% (P = .26).
5. Implant survival at ten years. Biner M, Klotz S, Andronic O, Karczewski DC, Zingg L, Wieser K, Kriechling P. Long-term outcomes following reverse total shoulder arthroplasty: a systematic review with a minimum follow-up of 10 years. JBJS Open Access. 2025;10(2):e25.00025. Systematic review, 7 studies. Weighted mean revision-free survivorship 88% at 10 years (5 studies; range 73%–94%). Overall complication rate 36%. Weighted mean loss to follow-up 63.3% (range 47%–80%); mean follow-up 11.8 years; mean age 70.5. doi:10.2106/JBJS.OA.25.00025
6. Survival by indication and age. Survivorship of reverse shoulder arthroplasty according to indication, age and gender. J Clin Med. 2022. Retrospective cohort. 10-year revision-free survival 91.0% primary vs 80.9% revision (P < .001). By age: <60 years 75.7%; 60–69, 88.8%; 70–79, 91.3%; ≥80, 94.3% (P < .001). PMCID PMC9145012.
7. Reverse versus hemiarthroplasty for fracture, multicenter trial. Jonsson EÖ, Ekholm C, Salomonsson B, Demir Y, Olerud P; Collaborators in the SAPF Study Group. Reverse total shoulder arthroplasty provides better shoulder function than hemiarthroplasty for displaced 3- and 4-part proximal humeral fractures in patients aged 70 years or older: a multicenter randomized controlled trial. J Shoulder Elbow Surg. 2021;30:994–1006. RCT, 99 randomized (48 rTSA, 51 HA), 84 analyzed; minimum 2-year follow-up; mean age 79.5. Constant 58.7 vs 47.7, mean difference 11.1 (95% CI 3.0–18.9; P = .007). Flexion 125° vs 90°; abduction 112° vs 83° (both P < .001). No difference in WOOS, pain, or EQ-5D.
8. Reverse versus hemiarthroplasty for fracture, single-center trial. Sebastiá-Forcada E, Cebrián-Gómez R, Lizaur-Utrilla A, Gil-Guillén V. Reverse shoulder arthroplasty versus hemiarthroplasty for acute proximal humeral fractures: a blinded, randomized, controlled, prospective study. J Shoulder Elbow Surg. 2014;23(10):1419–1426. RCT, n = 62 (>70 years), mean follow-up 28.5 months. RSA vs HA: UCLA 29.1 vs 21.1; Constant 56.1 vs 40.0; forward elevation 120.3° vs 79.8°; abduction 112.9° vs 78.7° (P = .001). doi:10.1016/j.jse.2014.06.035
9. Reverse versus anatomic replacement for arthritis with intact cuff. Kirsch JM, Puzzitiello RN, Swanson D, Le K, Hart PA, Churchill R, Elhassan B, Warner JJP, Jawa A. Outcomes after anatomic and reverse shoulder arthroplasty for the treatment of glenohumeral osteoarthritis: a propensity score-matched analysis. J Bone Joint Surg Am. 2022;104(15):1362–1369. Retrospective matched cohort, n = 134 (67 per group), mean follow-up 30 ± 10.7 months. No significant differences in final VAS pain, ASES, or SANE. doi:10.2106/JBJS.21.00982
10. Growth in use. Mayfield CK, Korber SS, Hwang NM, et al. Volume, indications, and number of surgeons performing reverse total shoulder arthroplasty continue to expand: a nationwide cohort analysis from 2016–2020. JSES Int. 2023;7(5):827–834. Nationwide database analysis. Primary rTSA rose from 55% (2016) to nearly 70% (2020) of primary shoulder arthroplasty; rTSA volume 14,781 (2016) to peak 23,644 (2019). doi:10.1016/j.jseint.2023.05.002
11. Dislocation rate, timing, and risk factors. ASES Complications of RSA Research Group; Lohre R, Swanson DP, Mahendraraj KA, et al. Predictors of dislocations after reverse shoulder arthroplasty: a study by the ASES complications of RSA multicenter research group. J Shoulder Elbow Surg. 2024;33(1):73–81. Multicenter retrospective cohort, n = 6,621; mean follow-up 19.4 months. Dislocation 2.1% overall; 1.6% primary (99/6,019); 6.5% revision (P < .001). Median 7.0 weeks after surgery (IQR 3.0–36.0). doi:10.1016/j.jse.2023.05.028
12. Acromial and scapular spine stress fracture rate. King JJ, Dalton SS, Gulotta LV, Wright TW, Schoch BS. How common are acromial and scapular spine fractures after reverse shoulder arthroplasty? A systematic review. 2019. Systematic review, 90 studies. Incidence 2.8% (253/9,048). Primary vs revision 2.8% vs 2.1% (P = .4). Highest after RSA for inflammatory arthritis (10.9%) and massive rotator cuff tears (3.8%). PMID 31154841.
13. Early versus delayed motion after surgery. Hagen MS, Allahabadi S, Zhang AL, Feeley BT, Grace T, Ma CB. A randomized single-blinded trial of early rehabilitation versus immobilization after reverse total shoulder arthroplasty. J Shoulder Elbow Surg. 2020;29(3):442–450. RCT, 107 shoulders enrolled, 86 (80.3%) completed 1-year follow-up (44 delayed, 42 immediate). Similar improvements in ROM and outcomes between protocols; delayed group immobilized 6 weeks. doi:10.1016/j.jse.2019.10.005
14. Return to sport. Singh A, Cho V, Abdel Khalik H, Dagher D, Sheth U, Khan M. Return to sport after reverse shoulder arthroplasty: a systematic review. Sports Health. 2026;18(2):457–465. 19 studies, n = 3,092; mean age 72.5. Return to sport at any level 85.1% (95% CI 84.6–85.6); same or higher level 69.5%. doi:10.1177/19417381251343083