“What are the issues in preventing and dealing with complications in reverse shoulder arthroplasty (RSA)?” asked Pascal Boileau, MD, professor of orthopaedic surgery, Université Nice Sophia Antipolis. “In this session, we will attempt to compare the American and French experiences, and perhaps learn from each other.”

Dr. Boileau moderated Instructional Course Lecture 147 at the 2014 AAOS Annual Meeting, which was a collaborative effort between AAOS and the Société Française de Chirurgie Orthopédique et Traumatologique (SOFCOT).

Indications and limits
“Although RSA was designed to treat arthritis in cuff-deficient shoulders, its use has expanded to treat a variety of conditions, including fractures, posttraumatic arthritis, tumors, and revision arthroplasty,” said Gilles Walch, MD, of the Centre Orthopédique Santy, in Lyon, France.

This expansion has resulted in an increasing number of RSAs being performed in France. In his practice, Dr. Walch reported a steady increase in the use of reverse shoulder prostheses over nearly 2 decades, with RSAs exceeding the number of anatomic prostheses in 2011.

Dr. Walch said that traditional contraindications for RSA include infection, deltoid damage, glenoid bone loss, young age, and intact rotator cuffs. However, he called a few of those assumptions into question.

“Complete palsy of the deltoid is a contraindication,” he agreed. “But patients with deltoid damage are not uncommon. Preoperative deltoid damage may not be a contraindication if the remaining deltoid has adequate strength. However, the results definitely are less predictable in the case of global deltoid insufficiency.”

According to Dr. Walch, young age is often considered to be a contraindication, because there is little long-term data on RSAs, but recent studies offering data with 10-year follow-up suggest that outcomes may be similar to anatomic shoulder arthroplasty.

“Age is no longer a contraindication for RSA, if no other good option is available,” he said.

Nor should glenoid deficiency be considered a contraindication, at least in primary RSA, Dr. Walch explained.

“It is always possible to do a glenoid reconstruction using bone graft from either the humeral head or the iliac crest,” he said. “In shoulder revision, bone deficiency may be very significant. We know that autograft or allograft are always possible, and therefore revision is not a contraindication for performing an RSA.”

Regarding humeral deficiency, Dr. Walch pointed out that use of allograft or a custom prosthesis may be options. However, if the deltoid insertion is absent, RSA is contraindicated.

Longer life for RSAs?
“Can we make RSAs last longer?” asked Luc Favard, MD, of the department of orthopaedic surgery, University of Tours.

Dr. Favard pointed out that although 10-year survival of RSA may run higher than 90 percent, the procedure may be associated with increase in pain and reduced mobility over time.

“We compared follow-up at 5, 7, and 9 years,” he said. “We found a significant difference in pain at 5 years. At 7 years, we found a significant difference in relative constant score. At 9 years, there was a very significant difference for pain, mobility, relative constant score, and constant score.”

According to Dr. Favard, a study of 68 shoulders with minimum 8-year follow-up found that, on the glenoid side, 88 percent of cases had scapular notching, 75 percent had bony scapular spurs, and 16 percent had radiolucency. Scapular notching and spurs were more frequently associated with a superolateral approach.

“Now the problem is to understand what influence scapular notching has on constant score,” he said. “We found no significant difference in constant score at 5, 7, or 9 years in the case of grade 3 or 4 notching.”

Regarding the humeral side, a separate study found subsidence in 8.8 percent of cases, more frequently with cemented stems. In contrast, uncemented stems had higher incidences of stress shielding (26 percent of cases) and lysis of tuberosities (85 percent of cases).

“We can say that the pain with RSA could be related to potential loosening on the glenoid side in the case of radiolucencies, or due to potential loosening on the humeral side in the case of subsidence or radiolucencies.

“A decrease of mobility with time could be related to the aging of patients, of course,” he continued, “but probably and mainly because of lysis of tuberosity, especially with uncemented stems. A decrease with the deltoid strength and extension of the cuff tear are other possible reasons.”

Nuts and bolts
Gregory P. Nicholson, MD, associate professor of orthopaedic surgery at Chicago’s Rush University, offered some “nuts-and-bolts” strategies for performing RSA. He agreed with other presenters that an intact deltoid and adequate glenoid bone stock are important factors in the success of RSA. In addition, he cited an informed patient, proper prosthetic design, and an experienced shoulder surgeon as additional keys to success.

“A good anteroposterior radiograph can give you a lot of information before heading to the operating room,” he advised. “The inferior rim is usually intact, even in glenoids with massive bone loss.”

In addition, he recommended assigning someone in the operating room to hold onto the humeral head during sawing.

“You need to save the head for morselized graft. Even in primary reverses, you may need to have some bone graft available,” he said.

“In addition, bony increased offset (BIO)-RSA has the potential to provide a more uniform graft, but you can also tailor that graft at an angle to fill up some defect.”

Dr. Nicholson reviewed data from a study at his institution of 137 shoulders (111 primary and 26 revisions). The research team found that revision RSAs were linked to a 69 percent complication rate, compared to a 25 percent rate for primary RSA.

“The most common predictive factor for a surgical complication when performing an RSA is the fact that it is a revision,” he said. “The most predictive factor for a medical complication is a condition very typical in the United States: increased body mass index. The bigger you are, the more complications you have.”

In addition, Dr. Nicholson addressed the assessment of glenoid bone loss. Plain radiographs can help, he explained, but computed tomography (CT) and magnetic resonance imaging are limited because there may be an existing implant.

“You’re probably going to have to do an intraoperative assessment,” he said. “When you’re in the operating room, you should have a primary plan and several backup plans. You also must have the equipment, preferably in the room with you. Even if you think your radiograph and CT showed you exactly what you need, you’re going to need long-post baseplates; you might need eccentric or lateralized glenospheres; you need multiple bone graft options available.”

Revising the reverse
When performing a revision, Dr. Boileau said that retention or removal of the device depends on the cause for revision: instability, infection, humeral problems, or glenoid problems.

“Treatment of instability after an RSA may include closed reduction, use of an additional humeral spacer, exchange of the prosthesis for lengthening or lateralization, transformation of the reverse into a hemiarthroplasty, or temporary removal of the humeral component to generate scar tissue and achieve latent stability. Think about doing a closed reduction if you have an early instability, because it works in 50 percent of cases, but you should immobilize your patient for at least 4 weeks.”

Dr. Boileau pointed out that infection is often a problem with RSA.

“Why? Because we operate on more and more patients with failed trauma and failed fracture sequelae, and these patients have multiple prior surgeries, which presents a high risk for low-grade infection.

“In our experience, infection occurs in 2 percent of primary RSAs and 7 percent of revision cases.”

According to Dr. Boileau, if there is any doubt about infection, a two-stage procedure is safer.

“Conflict of ego”
Jon J.P. Warner, MD, professor of orthopaedic surgery at Harvard Medical Center, spoke about the learning curve for performing RSA, and what he called the “conflict of ego.”

“We talk about conflict of interest and conflict of commitment,” he said, “but we don’t talk about conflict of ego. As surgeons, most of us have egos, and that’s necessary to make good decisions for our patients, but sometimes, that can interfere with what we do. It’s very important to be self-critical and analyze. If you want to do one thing to reduce your complication rates, measure your outcomes and look at them periodically and systematically.”

Dr. Warner said that orthopaedic surgeons can learn more from their failures than their successes, and he encouraged his colleagues to track and analyze their own progress as surgeons, so they can continue to improve their techniques. The first step, he explained, is to define the factors within the surgeon’s control.

“Some things are avoidable and some things are not,” he said. “We would agree that infection, fracture, instability, nerve problems, loosening, and mechanical component problems are generally things within our control, or at least we want to believe that. But conditions such as scapular notching may or may not be.”

Dr. Warner displayed graphs of complication rates at his practice, which declined over time.

“When you measure and critically analyze, your conflict of ego becomes less of a factor, because you learn how to avoid problems. For me, it’s very sobering issue. I’m a better surgeon now than when I started, because of my experience.”

Disclosure information: Dr. Boileau—Tornier; Smith & Nephew; Mitek; Dr. Walch—Tornier; Journal of Shoulder and Elbow Surgery; Journal of Bone and Joint Surgery – American; OTSR-RCO; CORR; Dr. Favard—Tornier; Mathys Ltd; Dr. Nicholson—Innomed; Tornier; Zimmer; SLACK Incorporated; Dr. Warner—Tornier; Arthrocare; DJ Orthopaedics; Arthrex, Inc; Mitek; Breg; Smith & Nephew; American Shoulder and Elbow Surgeons; San Diego Shoulder Institute

Peter Pollack is electronic content specialist for AAOS Now. He can be reached at ppollack@aaos.org