Monday, June 27, 2011

Failed, Unsatisfactory Shoulder Joint Arthroplasty

We will now begin a series of posts on the causes of failure of shoulder joint replacement and then move on to how these problems may be managed.

Our shoulder fellows Hasan and Leith presented our first article on the subject, Characteristics of Unsatisfactory Shoulder Arthroplasties. Failure of shoulder arthroplasty is often defined as a complication or the need for revision, but it may also be viewed as a result that does not meet the expectations of the patient. To enhance our understanding of failed shoulder arthroplasties, we identified the characteristics common to a series of 139 consecutive patients who came to our shoulder consultation service because of dissatisfaction with the result of their shoulder arthroplasty. Primary osteoarthritis (28%) and proximal humeral fractures (26%) were the most common indications for the initial arthroplasty. Seventy-three shoulders (fifty-two percent) had at least one surgery before arthroplasty was performed. Seventy-four percent of the shoulders were stiff, 35% were unstable, and in the total shoulders, 59% of the glenoids were loose. Components were substantially malpositioned in 23%. Forty-two percent of shoulders with a failed hemiarthroplasty had substantial glenoid erosion, and 43% of shoulders that had undergone a hemiarthroplasty for fracture had nonunion of the tuberosities. Patients demonstrated impaired shoulder function; on average, they could perform only 2 of 12 shoulder functions. The rate of revision underestimated the rate of failure, as 23% of arthroplasties did not undergo revision. The challenge of achieving patient satisfaction after arthroplasty may be greater than previously recognized. Many of these unsatisfactory shoulder arthroplasties did not meet the criteria for failure used in previously published series. These observations suggest that greater attention to achieving proper component position, postoperative motion, and in fracture cases, fixation of the tuberosities may lead to increased patient satisfaction after shoulder arthroplasty.

Our shoulder fellows Amy Franta and Tim Lenters later published an update on this topic, The Complex Characteristics of 282 Unsatisfactory Shoulder Arthroplasties. We defined an 'unsatisfactory arthroplasty' as a prior shoulder joint replacement with a result that prompted the patient to consult with us about the possibility of a revision surgery. We found that pain was a prominent feature in 241 cases. Shoulder function was seriously compromised: the average patient was able to perform only 2.6 of the 12 functions of the Simple Shoulder Test. Component malpositioning and glenohumeral malalignment were the most technical failures observed in these shoulders. 85 of 136 total shoulders had loose glenoid components. Glenoid erosion was found in 51 of 80 hemiathroplasties.

These studies have motivated our research into ways to avoid unsatisfactory arthroplasties on one hand and ways to manage these unsatisfactory arthroplasties on the other.