Lessons are to be learned from success and failures of shoulder arthroplasty

Winston J. Warme, MD, answers 4 Questions about research on failed shoulder arthroplasties.

Issue: December 2008

We all can usually benefit from studying and reviewing our own failures. When we do not perform a large volume of a particular surgery, it is helpful to learn from the experience of others, particularly when learning how to best avoid failures. This month I asked Winston J. Warme, MD, to share his insights gained from the Shoulder Arthroplasty Failure Experience (SAFE), a review of the failure modes for 353 shoulders arthroplasties done over a 10-year period that was published in the Journal of Shoulder and Elbow Surgery in 2007. I invited him to share some of the aspects that might enhance our patient care with shoulder arthroplasties.

Douglas W. Jackson, MD
Chief Medical Editor

Douglas W. Jackson, MD: Which specific operations are included in the SAFE study?

4 Questions

Winston J. Warme, MD: In the SAFE study, we reviewed the failure modes for 353 shoulders entered into a prospective database over a 10-year period at a busy referral center. The patients presented with hemiarthroplasties or total shoulder arthroplasties, performed for the diagnoses of rheumatoid or osteoarthritis, fracture, post-traumatic arthrosis, avascular necrosis or capsulorraphy arthropathy.

Jackson: While failures are multifactorial, did you see a difference in low- vs. high-volume surgeons?

Warme: We did not specifically research the patients’ previous surgeon’s case volumes, so we cannot categorically answer the question from our database. However, there are articles in the total hip, knee and shoulder literature that show higher complication rates and longer lengths of hospital stay in low-volume vs. high-volume surgeons.

While somewhat arbitrary, more than 30 arthroplasty procedures a year is an accepted definition for a high-volume shoulder arthroplasty surgeon. A low-volume surgeon performs five or fewer shoulder arthroplasties annually.

Winston J. Warme, MD
Winston J. Warme

Jackson: What constitutes and/or contributes to a failure in terms of a shoulder arthroplasty?

Warme: A failure is defined as an arthroplasty procedure that fails to meet the patient’s expectations. This is not necessarily what we as surgeons might consider a failure, such as a re-operation, or a complication such as an intraoperative humeral shaft fracture. However patients may or may not see it in the same light.

Patients may consider an arthroplasty a failure, despite good implant positioning and the absence of infection, if their residual stiffness does not allow them to do their own hair, or other activities of daily living. Despite nice postoperative X-rays, if they still have a level of discomfort that they find lifestyle-altering, they may consider the operation a failure. This highlights the importance of setting appropriate expectations, careful patient selection, technically excellent surgery, meticulous postoperative therapy, and close follow-up in the global period.

Jackson: What practical findings and outcomes can we draw from the SAFE investigation?

Warme: While failure modes were commonly multifactorial, component malposition, glenohumeral malalignment and glenoid failure were all prevalent features among patients studied with unsatisfactory outcomes. For trauma patients, tuberosity malunion or nonunion were commonly associated with clinical failure.

Shoulder arthritis requiring arthroplasty is less prevalent than coxarthrosis and gonarthrosis, such that most community orthopedic surgeons have less experience or familiarity with it.

Postoperative X-ray showing restored "register"

Postoperative X-rays showing restored “register” with the center of rotation of the implant at the center of rotation of the glenoid, on both Grashey and axillary views, with reduced tuberosities and proper implant height.

Images: Warme WJ

Postoperative X-ray showing restored "register"

There are a few technical pearls that we can share that may help:

To ensure 30° of retroversion: With the shoulder dislocated, simply rotate the forearm to 120° of external rotation. Cut the anatomic neck and insert the implant pointing straight forward. This is the simplest way we know to get the version right. There are many complex jigs that have been designed to help, but keeping it simple is always best.
To set the correct “register:” This refers to the relationship of the humeral head to the glenoid. The centers of rotation should match. It is important to recognize that the center of the humeral head prosthesis is often not the center of rotation, as many prosthetic designs do not utilize hemispherical implants. In an arthritic shoulder, the register is often disrupted in both the axial (anteroposterior) and coronal (superoinferior) planes. As the humeral head subluxates posteriorly, it brings the center of rotation posterior to the glenoid center, and progressive soft-tissue contracture often leads to the superior displacement of the center of rotation. At the time of surgery, we seek to ensure that the humeral head center of rotation remains centered in the glenoid when the arm is adducted to the side, when it is externally rotated 40· with the arm at the side, and when it is internally rotated 60· while in 90· of abduction. If the register is incorrect in either plane, adjustments in the position of the humeral component can be performed. Appropriate register must be assured before the procedure is concluded to optimize load transfer and stability by centering the humeral articular surface in the corresponding glenoid surface.
Insertion of a glenoid implant must be done meticulously to minimize the chance of loosening. Choose the largest size that will be fully supported by bone. Be cognizant of posterior wear to avoid excessive retroversion that can lead to posterior instability and adjust the reaming to normalize this somewhat. Use irrigation to avoid excessive heat production while reaming and avoid cement on the back of the implant; thin PMMA will crack and lead to loosening.
With fracture cases, strongly consider stem designs that preserve the maximal amount of bone and that optimize tuberosity healing by maximizing bone contact and fixation with north-south as well as east-west sutures. Intraoperative imaging of hemiarthroplasties for fracture is recommended to ensure implant and tuberosity height are perfect prior to closing.
Finally, consider referral if a case seems excessively challenging. These are often tough, challenging cases in anyone’s hands and are commonly not surgical emergencies.

For more information:

Winston J. Warme, MD, chief, shoulder and elbow surgery and associate professor in the Department of Orthopaedics and Sports Medicine, University of Washington Medical Center, can be reached at 1959 NE Pacific St., Box 356500, Seattle, WA 98195-6500; 206-543-3690; e-mail: armewj@u.washington.edu.

Reference:

Franta AK, Lenters TR, Mounce D, et al. The complex characteristics of 282 unsatisfactory shoulder arthroplasties. J Shoulder Elbow Surg. 2007:16(5):555-562.