What is involved robotic-assisted shoulder arthroplasty? How do we find out if it's worth it?

A recently published article Robotic Assisted Reverse Total Shoulder Arthroplasty: Narrative Review and Surgical Technique of Humeral and Glenoid Preparation summarized the evolution of advanced execution modalities and describes the surgical technique and workflow of the Zimmer Robotic Surgical Assistant (ROSA) Shoulder System for a robotic assisted reverse total shoulder arthroplasty (RSA). 

Two of the authors are early adopters of this technology: Zimmer Biomet Announces Successful Completion of World’s First Robotic-Assisted Shoulder Replacement Surgery with ROSA Shoulder System and UBMD Orthopaedics & Sports Medicine Becomes 3rd Site in the World to Perform a Robotic Shoulder Arthroplasty. As the authors describe:

The Robotic Surgical Assistant system consists of two units: one corresponds to the robotic arm and the other to the optical unit, which includes an infrared camera mounted on a separate arm.

This setup enables communication with the optical markers implanted in the patient’s humerus and coracoid. Operating in a semi-active mode, the robot assists in both humeral and glenoid preparation through three modes: automatic, collaborative, and static. In automatic mode, the robot positions the arm in the surgical area. Once in place, it switches to collaborative mode, allowing the surgeon to move the arm within the limits established in the surgical plan. After the final position is determined, the robot transitions to static mode to perform precise cuts or reaming.

Humeral head resection is performed through robotic positioning of an extramedullary cutting guide, whose final position is adjusted in collaborative mode. The robotic arm controls the cut guide along the plane of the planned resection and the guide is then secured to the humerus with pins. The robot is then placed in static mode and the humeral cut is performed through this guide at the planned version, inclination, and height. The cut surface of the humerus is then validated to confirm appropriate resection of the humeral head. Following validation the remainder of the humeral preparation including reaming and broaching is performed manually.

Glenoid preparation is performed with a conventional reamer attached to the robotic arm that is powered by a standard cordless reamer. This allows glenoid reaming, controlled by the robot with live tracking in collaborative mode, to the desired version, inclination, and depth. Bone preparation for aTSA involves two robotic controlled steps. The first reamer prepares the face of the glenoid to the correct version, inclination and depth. The second prepares the central hole for the hybrid glenoid central post. The remaining glenoid preparation for the implant performed manually. In RSA the glenoid is prepared in a single step ream for the glenoid face and central boss with the robot controlling the version, inclination and depth in collaborative mode.

The steps include 

Humeral clamp placement on the proximal humerus

Reproduced from Marigi et al., JSES Reviews, Reports & Techniques, 2026, under CC BY-NC-ND 4.0

Humeral registration across various landmarks utilizing the probe

Reproduced from Marigi et al., JSES Reviews, Reports & Techniques, 2026, under CC BY-NC-ND 4.0

Robotic insertion of the humeral cut guide

Reproduced from Marigi et al., JSES Reviews, Reports & Techniques, 2026, under CC BY-NC-ND 4.0

The authors provided videos of this technique

Reproduced from Marigi et al., JSES Reviews, Reports & Techniques, 2026, under CC BY-NC-ND 4.0

Reproduced from Marigi et al., JSES Reviews, Reports & Techniques, 2026, under CC BY-NC-ND 4.0

 Coracoid array tracker secured on the coracoid process using two pins.

Reproduced from Marigi et al., JSES Reviews, Reports & Techniques, 2026, under CC BY-NC-ND 4.0

Glenoid registration across various landmarks using the probe.

Reproduced from Marigi et al., JSES Reviews, Reports & Techniques, 2026, under CC BY-NC-ND 4.0

Robotic assisted glenoid reaming using an implant specific reamer mounted on the robotic arm.

Reproduced from Marigi et al., JSES Reviews, Reports & Techniques, 2026, under CC BY-NC-ND 4.0

 A, Reamed glenoid prepared for baseplate implantation. B, Baseplate inserted onto the prepared glenoid.

Reproduced from Marigi et al., JSES Reviews, Reports & Techniques, 2026, under CC BY-NC-ND 4.0

A trial reduction is performed to assess and optimize soft tissue tension and range of motion. While robotic implantation accurately executes pre-operative planning based on the CT scan anatomy, which can assist in soft tissue balancing, intraoperative adjustments may be necessary to achieve optimal soft tissue balance.

Postoperative radiograph of a RSA with humeral and glenoid based robotic preparation.

Reproduced from Marigi et al., JSES Reviews, Reports & Techniques, 2026, under CC BY-NC-ND 4.0

The ROSA system is reported to cost between $1,000,000 and $1,500,000.  As the authors point out, "The disadvantages associated with the implementation of robotic-assisted SA are not limited solely to the costs involved. This surgery will entail a steep learning curve for surgeons already accustomed to other techniques, and it could also introduce a cognitive bias in trainee surgeons. Furthermore, in centers that do not yet have the robot, it will be necessary to modify operating rooms to provide sufficient space for the installation of the robotic unit."

Robotics is a technology to transfer a preoperative plan to the patient. Thus, this technique guide is not expected to present data on the efficacy of a robotic approach in optimizing component positioning (What Reverse Total Shoulder Geometry Will Give My Patient the Best Function and Lowest Complication Risk?)

This technique guide is not expected to present data on the value of robotics to the patient in terms of comfort, function, and reduced revision rate.

In his recent article, Robot-assisted shoulder arthroplasty  Sanchez-Sotelo opined: "The main theoretical benefits of robot-assisted shoulder arthroplasty include accuracy and precision, data acquisition, and with certain robots, the promise to avoid soft-tissue injury with haptic boundaries, prepare a bone through minimally invasive or cuff-preserving exposures, and the potential for motion assessment and soft-tissue balance. The disadvantages include cost, a certain learning curve, complications related to array insertion, potential for cognitive bias, need for a larger operating room space, and the potential for malfunction. Although adoption is likely to happen in many centers, cost and space constrains may favor alternative technologies, such as mixed reality navigation, especially in ambulatory surgery centers."

When openevidence.com was asked "does robotic-assistance improve patient reported outcomes for anatomic or reverse shoulder arthroplasty?", it concluded "There is currently no clinical evidence that robotic arthroplasty improves patient-reported outcomes for anatomic or reverse shoulder arthroplasty".

Let's think on this a bit.

Great-horned Owl

Seattle

2021

Follow on twitter/X: https://x.com/RickMatsen

Follow on facebook: https://www.facebook.com/shoulder.arthritis

Follow on LinkedIn: https://www.linkedin.com/in/rickmatsenmd

[1-2]

What Reverse Total Shoulder Geometry Will Give My Patient the Best Function and Lowest Complication Risk?

While we know that many patient (osteoporosis, steroid use, inflammatory arthropathy, female sex) and shoulder factors (cuff status, prior surgery, coracoacromial arch integrity, humeral and glenoid deformity) drive the outcome of reverse shoulder arthroplasty, I was curious to see what the literature has to say about the surgeon-controlled variables related to the geometry of the reconstruction. I've tried to focus on the position, rather than the design, of the components. Here's what I think I learned - as always - your comments are welcome.

Avoiding Complications

Minimizing risk of inferior impingement, scapular notching, and baseplate loosening

Place the baseplate flush with the inferior glenoid rim

  Select and place humeral component to achieve a 135° liner-shaft angle

Target 4–10 mm of lateralization of the glenosphere center of rotation (CO): defined as the distance from the glenosphere center of rotation (COR) to the glenoid bone surface (includes the thickness of the baseplate, bone graft and/or augment). Know your glenospheres and don't rely on the numbers on the box: for example, in one implant system the 3mm baseplate plus a "32-4" glenosphere lateralizes the center of rotation by 9mm.

Place baseplate in 0 - 10 degrees of inferior tilt: central screw parallel to floor of supraspinous fossa so that the humeral force on the glenosphere is perpendicular to the screw fixation.

Minimizing risk of neurologic injury and pain.

Avoid excess humeral distalization: acromiohumeral interval (AHI) <30mm, humeral lengthening (pre to postoperative change in AHI) <20mm.

Minimizing risk of acromial and scapular spine fracture

The distance from the glenosphere center (COR) to the most lateral point on the acromial undersurface (DA) should exceed the distance from the glenosphere center (COR) to the lateral tip of the greater tuberosity (DGT). DA ≥ DGT. 

Minimize humeral-sided contribution to global lateralization. Humeral sided lateralization directly increases the distance from the glenosphere center (COR) to the lateral tip of the greater tuberosity (DGT) while leaving the distance from the glenosphere center (COR) to the most lateral point on the acromial undersurface (DA) unchanged, worsening the DA - DGT difference. By contrast glenoid-sided lateralization (increasing lateralization of the glenosphere center of rotation (CO), changes both the DA and DGT simultaneously, preserving more control over the DA - DGT difference.

 

Minimizing instability risk

Strive for humeral retroversion 0°–20° and glenoid retroversion 0°–20° (Recall that soft tissue tension, humero-scapular impingement, liner geometry, and other factors play major roles in rTSA stability).

Optimizing Function

Deltoid efficiency

Position the glenosphere center of rotation inferiorly and posteriorly to maximize the deltoid's mechanical advantage during abduction and flexion

Motion

4–10 mm glenoid-sided lateralization improves internal rotation by displacing the humeral cup away from the scapular neck, reducing the impingement that limits motion.

Glenoid retroversion 0°–20° optimizes external rotation

A 135° liner-shaft angle increases adduction and rotational range

Avoid excess humeral distalization: Keeping acromiohumeral interval (AHI) <30mm, humeral lengthening (change in AHI) <20mm facilitates flexion and external rotation.

"Glenosphere Lateralization" - Resolving the Nomenclature Confusion

The geometry of reverse shoulder arthroplasty reconstruction is frequently discussed in terms of a single number ("glenoid lateralization” , "metallic offset", "lateralization shoulder angle (LSA)), yet none of these numbers captures the important surgeon-controlled variables and their use results in contradictory results in the rTSA literature. 

Three distinct measurements describe different aspects of glenosphere construct geometry. 

Global Lateralization (GL): distance from the glenoid bone surface to the lateral tip of the greater tuberosity (= GT + humeral component contribution). Includes the baseplate and augments. This is the distance that the tuberosity is lateralized from the native glenoid bone. Increases in GL tighten the shoulder - increasing stability, but may also increase the risk of contact between the tuberosity and the acromion when the arm is elevated.

Effective Glenosphere Thickness (GT): distance from the glenoid bone surface to the lateral edge of the glenosphere (= CO + glenosphere radius). Includes the baseplate and augments. This is the glenosphere contribution to global lateralization

.

Center of Rotation Offset (CO): distance from the glenoid bone surface to the glenosphere center of rotation (COR). Includes the baseplate and augments. The COR is the pivot point around which the tuberosity rotates. The position of the COR defines the moment arm for deltoid action. 

All three are measured from the same landmark (glenoid bone surface) but to different endpoints. Each can vary independently of the others through implant selection and surgical technique. Reporting all three measures on postoperative radiographs would resolve most of the apparent conflicts.

A clear-eyed view

Peregrine Falcon

Union Bay Natural Area

2019

Follow on twitter/X: https://x.com/RickMatsen

Follow on facebook: https://www.facebook.com/shoulder.arthritis

Follow on LinkedIn: https://www.linkedin.com/in/rickmatsenmd

References

1. Arenas-Miquelez A, Murphy RJ, Rosa A, Caironi D, Zumstein MA. Impact of Humeral and Glenoid Component Variations on Range of Motion in Reverse Geometry Total Shoulder Arthroplasty: A Standardized Computer Model Study. J Shoulder Elbow Surg. 2021.

2. Dean EW, Dean NE, Wright TW, et al. Clinical Outcomes Related to Glenosphere Overhang in Reverse Shoulder Arthroplasty Using a Lateralized Humeral Design. J Shoulder Elbow Surg. 2022. [Note: “overhang” is measured as glenosphere-to-baseplate offset on 2D Grashey radiograph, not glenosphere-to-native-bone.]

3. Pak T, Kilic AI, Ardebol J, et al. Glenoid-Sided Lateralization Decreases Scapular Notching With a 135° Humeral Component Arthrex Reverse Shoulder Arthroplasty. J Shoulder Elbow Surg. 2025.

4. Meisterhans M, Bouaicha S, Meyer DC. Posterior and Inferior Glenosphere Position in Reverse Total Shoulder Arthroplasty Supports Deltoid Efficiency for Shoulder Flexion and Elevation. J Shoulder Elbow Surg. 2019.

5. Rai AA, LeVasseur CM, Kane GE, et al. Glenosphere Tilt and Size Predict Shoulder Kinematics During the Hand-to-Back Motion After Reverse Shoulder Arthroplasty. J Orthop Res. 2025.

6. Berton A, Longo UG, Gulotta LV, et al. Humeral and Glenoid Version in Reverse Total Shoulder Arthroplasty: A Systematic Review. J Clin Med. 2022.

7. Keener JD, Patterson BM, Orvets N, Aleem AW, Chamberlain AM. Optimizing Reverse Shoulder Arthroplasty Component Position in the Setting of Advanced Arthritis With Posterior Glenoid Erosion: A Computer-Enhanced Range of Motion Analysis. J Shoulder Elbow Surg. 2018.

8. Lee HH, Park SE, Ji JH, Jun HS. Mid-Term Comparative Study Between the Glenoid and Humerus Lateralization Designs for Reverse Total Shoulder Arthroplasty. BMC Musculoskelet Disord. 2023.

9. Wright MA, Murthi AM. Offset in Reverse Shoulder Arthroplasty: Where, When, and How Much. J Am Acad Orthop Surg. 2021.

10. Wolf GJ, Reid JJ, Rabinowitz JR, et al. Does Glenohumeral Offset Affect Clinical Outcomes in a Lateralized Reverse Total Shoulder Arthroplasty? J Shoulder Elbow Surg. 2022.

11. Nunes B, Linhares D, Costa F, et al. Lateralized Versus Nonlateralized Glenospheres in Reverse Shoulder Arthroplasty: A Systematic Review With Meta-Analysis. J Shoulder Elbow Surg. 2021.

12. Neyton L, Nigues A, McBride AP, Giovannetti de Sanctis E. Neck Shaft Angle in Reverse Shoulder Arthroplasty: 135 vs. 145 Degrees at Minimum 2-Year Follow-Up. J Shoulder Elbow Surg. 2023.

13. Baumgarten KM, Max C. Reverse Total Shoulder Arthroplasty Using Lateralized Glenoid Baseplates Has Superior Patient-Determined Outcome Scores at Short-Term Follow-Up. J Am Acad Orthop Surg. 2024.

14. Kawashima I, King JJ, Wright JO, et al. Shoulder Geometry After Reverse Total Shoulder Arthroplasty With a Medialized Glenoid and a Lateralized Humerus Predicts Subacromial Notching and Acromial or Scapular Spine Fractures. J Shoulder Elbow Surg. 2025.

15. Burden EG, Batten TJ, Smith CD, Evans JP. Reverse Total Shoulder Arthroplasty. Bone Joint J. 2021.

16. Nelson R, Lowe JT, Lawler SM, et al. Lateralized Center of Rotation and Lower Neck-Shaft Angle Are Associated With Lower Rates of Scapular Notching and Heterotopic Ossification and Improved Pain for Reverse Shoulder Arthroplasty at 1 Year. Orthopedics. 2018.

17. Ameziane Y, Audigé L, Schoch C, et al. Mid-Term Outcomes of a Rectangular Stem Design With Metadiaphyseal Fixation and a 135° Neck-Shaft Angle in Reverse Total Shoulder Arthroplasty. J Clin Med. 2025.

18. Jang YH, Lee JH, Kim SH. Effect of Scapular Notching on Clinical Outcomes After Reverse Total Shoulder Arthroplasty. Bone Joint J. 2020.

19. Mollon B, Mahure SA, Roche CP, Zuckerman JD. Impact of Scapular Notching on Clinical Outcomes After Reverse Total Shoulder Arthroplasty: An Analysis of 476 Shoulders. J Shoulder Elbow Surg. 2017.

20. Simovitch R, Flurin PH, Wright TW, Zuckerman JD, Roche C. Impact of Scapular Notching on Reverse Total Shoulder Arthroplasty Midterm Outcomes: 5-Year Minimum Follow-Up. J Shoulder Elbow Surg. 2019.

21. Spiry C, Berhouet J, Agout C, Bacle G, Favard L. Long-Term Impact of Scapular Notching After Reverse Shoulder Arthroplasty. Int Orthop. 2021.

22. Erickson BJ, Werner BC, Griffin JW, et al. A Comprehensive Evaluation of the Association of Radiographic Measures of Lateralization on Clinical Outcomes Following Reverse Total Shoulder Arthroplasty. J Shoulder Elbow Surg. 2022;31:963–970. [Multiple authors report financial relationships with Arthrex, Inc., manufacturer of the implant system used in this study.]

23. Werner BC, Lederman E, Gobezie R, Denard PJ. Glenoid Lateralization Influences Active Internal Rotation After Reverse Shoulder Arthroplasty. J Shoulder Elbow Surg. 2021;30:2498–2505.

24. Southam BR, Bedeir YH, Johnson BM, et al. Clinical and Radiological Outcomes in Lateralized Versus Nonlateralized and Distalized Glenospheres in Reverse Total Shoulder Arthroplasty: A Randomized Control Trial. J Shoulder Elbow Surg. 2023.

25. Longo UG, Gulotta LV, De Salvatore S, et al. The Role of Humeral Neck-Shaft Angle in Reverse Total Shoulder Arthroplasty: 155° Versus <155° — A Systematic Review. J Clin Med. 2022.

Complications and revisions following reverse total shoulder: doing the math.

Summary: Three takeaway points

(1) The commonly used and easy to measure "revision rate" is an inadequate endpoint for evaluating rTSA outcomes. The 8–20 percentage point gap that exists between the complication rate and the revision rate includes a substantial number of patients with failed but unrevised reverse shoulder arthroplasties. These patients are not considered in determining the failure rate when measured by the percentage having revision. Complication-free survival or patient-reported outcome measures should supplement or replace revision rate or "implant survival" as the primary outcome measure for rTSA.

(2) The most common complications of rTSA, acromial and scapular spine fractures, are unique and difficult to solve problems for patients having reverse arthroplasty. In contrast, the most common complications from aTSA - glenoid component loosening and rotator cuff tear - can be effectively managed by revision to a rTSA

(3) The salvage pathway for patients with rTSA failure is poor. Revision rTSA carries a 31% complication rate and 27% re-revision rate, with outcomes that are significantly worse than those for primary rTSA. This contrasts sharply with aTSA-to-rTSA conversion, the outcome of which approaches the outcomes for primary rTSA. 

The details

Complications vs revisions

Meta-analytic evidence reports that anatomic total shoulder (aTSA) has a higher rate of revision compared to reverse total shoulder (rTSA). This observation may have driven much of the shift toward rTSA for patients with cuff-intact arthritis.  However, it may neglect the fact that surgeons and patients can decide against revision of a failed rTSA because of the low rate of success. Thus, the absence of a revision does not indicate a good outcome.

Consider the references below

Complications and further surgery after reverse total shoulder arthroplasty : report of 854 primary cases reported an overall complication rate of 18–22%, yet a revision rate of only ~10%. That 8–12 percentage point gap likely represents patients who are living with a failing implant — either because revision was technically not feasible, or because expected outcomes were too poor to justify reoperation

Incidence, radiographic predictors, and clinical outcome of acromial stress reaction and acromial fractures in reverse total shoulder arthroplasty found 46 acromial stress fractures (5.4%) in 44 patients and 44 acromial stress reactions (5.2%) in 43 patients. The overall union rate was 55% but was significantly higher following operative treatment compared with nonoperative treatment. However, facture consolidation did not result in better clinical outcomes compared with nonunion; this may question the value of attempted fixation.

Comparison of complication types and rates associated with anatomic and reverse total shoulder arthroplasty found that the top 3 complications for rTSA were acromial/scapular fracture/pain ( complication rate 2.5%, revision rate 0.0%), instability (complication rate 1.4%, revision rate 1.0%), pain (1.2%, revision rate 0.2%). Compare the relationship of complications to revisions for rTSA to that for aTSA in the two charts below constructed from the data in this article. Note that a high percentage of the rTSA complications were not associated with surgical revision. Absence of revision ≠ absence of complication,

The Gap Between Revision-Free and Complication-Free Survival 

Long-Term Outcomes of Reverse Total Shoulder Arthroplasty
A Follow-up of a Previous Study The 93% ten-year revision-free survival rate for rTSA did not capture the 29% of patients who experienced complications.

Long term clinical and radiological outcomes of primary reverse total shoulder arthroplasty at a minimum follow-up of 15 years: Norwegian registry data showed 10-year complication-free rates of only 76–80%, despite revision-free survival of 91–95%. That 15–20 percentage point gap represents patients living with failed implants who are invisible in the revision statistics. Again, this gap exists because many rTSA complications—particularly acromial fractures, low-grade infection, and baseplate loosening with severe bone loss—either cannot be addressed surgically or carry such poor expected revision outcomes that conservative management is chosen.

Revision rates alone remain an inadequate measure of rTSA performance; complication-free survival is the more appropriate endpoint.

Outcomes of Revision rTSA

Revision of reverse total shoulder arthroplasty: a scoping review of indications for revision, and revision outcomes, complications, and rerevisions and Revision of failed shoulder arthroplasty: epidemiology, etiology, and surgical options found the most common indications for revision to be instability/dislocation (28–30%), baseplate or glenoid complications (20–25%), and infection (15–23%). Notably low on the list is revision for the most common complication: acromial/spine fractures.

Outcomes After Revision

Revision reverse total shoulder arthroplasty: clinical and radiographic outcomes compared to primary reverse total shoulder arthroplasty found a 31% complication rate and 27% re-revision rate of a failed rTSA: over five times these rates after primary rTSA. The most common complications after revision are recurrent instability (22%), fractures (18%), and baseplate issues (12%). Patients requiring revision rTSA experience markedly worse comfort and function in comparison to primary rTSA

The Salvage Pathway Asymmetry:  revision for aTSA Failure compared to revision for rTSA Failure

Conversion of failed anatomic total shoulder arthroplasty to reverse shoulder arthroplasty yields similar pain and functional outcomes to primary reverse shoulder arthroplasty but has a higher baseplate failure rate: a matched cohort study. In contrast to the often grim outcomes for revision of a failed rTSA, revision of a failed aTSA to a rTSA yields outcomes approaching those of primary rTSA. 

Summary: Three takeaway points (again)

(1) The commonly used and easy to measure "revision rate" is an inadequate endpoint for evaluating rTSA outcomes. The 8–20 percentage point gap that exists between the complication rate and the revision rate includes a substantial number of patients with failed but unrevised reverse shoulder arthroplasties. These patients are not considered in determining the failure rate when measured by the percentage having revision. Complication-free survival or patient-reported outcome measures should supplement or replace revision rate or "implant survival" as the primary outcome measure for rTSA.

(2) The most common complications of rTSA, acromial and scapular spine fractures, are unique and difficult to solve problems for patients having reverse arthroplasty. In contrast, the most common complications from aTSA - glenoid component loosening and rotator cuff tear - can be effectively managed by revision to a rTSA

(3) The salvage pathway for patients with rTSA failure is poor. Revision rTSA carries a 31% complication rate and 27% re-revision rate, with outcomes that are significantly worse than those for primary rTSA. This contrasts sharply with aTSA-to-rTSA conversion, the outcome of which approaches the outcomes for primary rTSA. 

Swans reversing course

Trumpeter Swans

Union Bay Natural Area

Seattle

Follow on twitter/X: https://x.com/RickMatsen

Follow on facebook: https://www.facebook.com/shoulder.arthritis

Follow on LinkedIn: https://www.linkedin.com/in/rickmatsenmd