Kaveh Sajadi, MD
When it comes to shoulder arthritis, shoulder arthroplasty is considered when symptoms are significant to the patient and conservative (nonoperative) management has failed to provide adequate relief. The general consensus is to delay shoulder replacement when possible due to concerns about implant longevity, particularly in younger individuals in whom a future revision would be anticipated. But delay may also come at a cost.
Over time, a patient may lose range of motion, muscle quality may deteriorate from limited use, and bony destruction may progress due to the arthritis. All of those may impact the outcome of shoulder replacement, as well as make the surgical procedure more difficult or necessitate different techniques, such as glenoid augments, bone grafts, or reverse shoulder arthroplasty. The rate of progression of arthritis has not been clearly established, so Drs. Logli and associates sought to better understand the progression of bone loss in osteoarthritis.
The authors performed a retrospective evaluation of patients who underwent shoulder arthroplasty and had high-quality radiographs taken at least once between 5 and 15 years prior to undergoing surgery.1 Forty-eight (48) patients were included with a mean interval time period of 8.9 years. They used the modified Walch classification to classify glenoid morphology on initial and preoperative axillary radiographs.
Overall, progression occurred but was not always consistent. Walch A1 patterns progressed to A2 41% of the time. Walch A patterns also progressed to become eccentric patterns (B or D patterns) approximately 20% of the time. Those shoulders presenting with subluxation (B) remained that way, with the only B1 becoming a B2, and 56% of B2s becoming B3.
In other words, those presenting initially with concentric alignment remained that way approximately 80% of the time, but the majority (60%) of those with eccentric alignment or posterior subluxation (B1/2) progressed to the next classification of wear.
Interestingly, the initial age of the first radiograph was highest in the B types (67 v. 60 for A and 64 for D), though this did not reach statistical significance but may reinforce the general progression shown in the data.
Limitations of the study include the relatively small patient numbers and the use of plain axillary radiographs. Plain radiographs may underestimate the amount of glenoid wear. In addition, they may underestimate the amount of humeral head subluxation. However, plain radiographs are far more practical and available in the outpatient setting to serve as a guide for patients and surgeons.
The authors are to be commended for this addition to the shoulder arthritis literature. They cite the only other significant article describing progression, but this current paper has a much longer time frame between imaging for comparison, providing a larger perspective. Surgeons can apply this paper as part of their patient counseling regarding the timing of shoulder arthroplasty.
Patients with concentric wear (A patterns) are much less likely to significantly progress over time than patients with eccentric wear. Those with posterior wear and subluxation can be counseled that the majority will continue to progress and worsen their glenoid wear.
Management of patients with posterior glenoid wear and subluxation is challenging. Proper soft-tissue balancing (carefully avoiding posterior capsule releases) is mandatory, and reconstruction of the glenoid vault as well as access to a retroverted glenoid is difficult, even in experienced hands.
Bone graft reconstruction of the glenoid, metallic glenoid augments, reverse shoulder arthroplasty, or even combinations of these may be necessary to provide an optimal outcome. The high percentage of eccentric patterns in this study (27% B and D) highlights the need for an implant system that includes augment options.
Three-dimensional (3-D) planning software is becoming widely available and is almost necessary for these types of cases. Finally, intraoperative navigation can make preparing the glenoid and inserting the implant more accurate,2 which is critical to restoring alignment.
Ultimately, shoulder replacement should only be performed when the patient has sufficient symptoms, not based solely on the radiographs. This article provides important information for surgeons to educate their patients on expected progression and the consequences of that progression.
- Logli AL, Pareek A, Nguyen NTV, Sanchez-Sotelo J. Natural history of glenoid bone loss in primary glenohumeral osteoarthritis: how does bone loss progress over a decade? J Shoulder Elbow Surg. 2021 Feb;30(2):324-330. doi: 10.1016/j.jse.2020.05.021. Epub 2020 Jun 9.
- Nashikkar PS, Scholes CJ, Haber MD. Computer navigation re-creates planned glenoid placement and reduces correction variability in total shoulder arthroplasty: an in vivo case-control study. J Shoulder Elbow Surg. 2019 Dec;28(12):e398-e409. doi: 10.1016/j.jse.2019.04.037. Epub 2019 Jul 26.
Kaveh Sajadi, MD, practices orthopaedics with Kentucky Bone and Joint Surgeons and is an instructor in the University of Kentucky’s residency program. He completed his residency at the Campbell Clinic and his fellowship at the NYU Langone Hospital for Joint Diseases. Dr. Sajadi is a frequent instructor at Exactech domestic and international shoulder courses.