George Athwal did a nice study (J Shoulder Elbow Surg 2018) where he showed if you’re going to do an entirely metaphyseal implant, you want to work towards the edges where the bone is a little bit stronger. If you’re going to have something to center, it’s very valuable to use the cortical bone also for your fixation. So, the ideal stemless device is going to have metaphyseal and epiphyseal fixation, anatomic head options and ease of instrumentation. Some devices that have now been out in Europe for many years, much longer than the United States, report results that are equal to the third or fourth generation stems. (Habermeyer, et al., J Shoulder Elbow Surg 2015) They are particularly useful in post-traumatic cases.
What about long-term outcomes? In the United States we’ve had these devices for less than two years. In Europe they’ve had them for more than ten years. Nine-year outcomes on a stemless device and the results are outstanding. (Hawl, et al., J Shoulder Elbow Surg 2017)
You have to careful with these stemless devices. It’s sold as an easier operation, but surgeons have a tendency when they are first using it to overstuff the joint. You have to be careful of your subscapularis repair. And you have to have a short stem available when you’re first learning in case you mess things up. Be careful about cutting in too much varus. That creates problems. (Collin, et al., International Orthopaedics 2017)
So, in summary, correct head size and position is critical. Short stems may lead to some stress shielding. Stemless may have less of that. Good outcomes are seen in both. Stemless—less bone removal, less blood loss, less pain, probably a better outpatient procedure. Revision from stemless is easiest. The key principle is put the head in the right spot. Why do you need all that extra metal to keep it there?
Dr. Cordasco: I would say that the short-stem implants are the way to go.
So, we’ll start with a paper from Mike Worth and colleagues, which was published last year in JBJS 2017 (“Complications of Shoulder Arthroplasty”). Glenoid wear and loosening remains a common cause of anatomic shoulder arthroplasty failure, followed by instability, rotator cuff tear, periprosthetic fracture, neural injury, infection, hematoma, deltoid injury and deep venous thrombosis. Humeral loosening is quite rare.
Neer first reported the outcomes of total shoulder over 45 years ago in 1974. And at the time, if you think about those first ten years, most initial implant designs had few variables of key features. Over the subsequent 20-25 years there’s been a slow but steady shortening of the humeral stem. Currently we can get as short as 55mm, with convertible platforms added most recently.
The latest generation of short-stem components can accommodate the variability of the proximal humerus with respect to head size, version, offset, and inclination. The humeral canal also facilitates proper component positioning. We can recreate pre-morbid anatomy, improve fixation, preserve bone, and also provide convertible options.
Churchill and Athwal (Curr Rev Musculoskelet Med 2016) reviewed the history of the stemless shoulder arthroplasty. It developed in 2004, some 30 years after the first anatomic reports. And it’s often referred to as canal sparing. It’s not to be confused with humeral resurfacing, such as the Copeland CTA. A standard humeral neck cut is performed with stemless total shoulders. They were designed to avoid stem-related complications including intraoperative humeral fractures, postoperative periprosthetic fractures—again unusual—proximal humeral bone loss with stress shielding, loosening—which is rare—and osteolysis. And to provide ease of revision and maintain optimum bone quality.
There are contraindications. Clearly, the proximal humeral bone has to be substantive enough to warrant this. Metabolic bone disease, inflammatory arthritis, and acute fractures are contraindications.
The stemless implant designs have four variables regarding these key features. The first is insertion technique. Most require some type of impaction, but one does have a threaded post with a screw-in application. The presence or absence of a collar is a variable. The surface coating, whether it’s ingrowth or ongrowth, is also a variable. And finally, the degree to which the metaphysis bone is contacted is an issue.
