The hemiarthroplasty for a cuff-deficient shoulder is technically not challenging when compared to reverse arthroplasty or conventional total shoulder arthroplasty in shoulders with an intact cuff. Goldberg and Bigliani provide an excellent description of the surgical technique [2]. A modified beach chair position is used. A mechanical arm holder is useful. A deltopectoral approach is used protecting the cephalic vein. Blunt dissection is used to expose the subacromial, subdeltoid, and subconjoined spaces. A self-retaining retractor is used to expose the subscapularis with the blades beneath the conjoined tendon and the deltoid. The biceps is tenotomized for later tenodesis if present. A subscapularis tenotomy is performed carefully tagging the tendon for later repair. The anterior humeral circumflex vessels are ligated or coagulated at the muscular portion of the inferior subscapularis.

Progressive extension and external rotation are used to dislocate the humeral head. Due to the massive tearing of the rotator cuff, exposure of the humerus is easily gained. An intramedullary guide can be used to guide a humeral cut in anatomic retroversion (about 30°). The height of the cut should aim to exit superiorly at the level of the greater tuberosity. After the humeral head cut, we mobilize any remaining supraspinatus and infraspinatus for later repair to the greater tuberosity.

Broaches are used to prepare the canal in standard fashion, and the final broach is used to trial extended coverage humeral heads that are designed to articulate with the acromion. Final tension should allow the humerus to glide about fifty percent anteriorly and posteriorly without instability. After ensuring an adequate fit of the prosthesis, definitive implants are inserted in standard fashion. Prior to placement of the definitive press-fit humeral stem, we pass heavy #5 suture through the lesser tuberosity to facilitate subscapularis repair. In addition, at this point, sutures can be passed through the greater tuberosity to aid in repairing as much rotator cuff as possible.

Currently, some modern shoulder systems allow for conversion of a hemiarthroplasty to reverse arthroplasty without removal of the stem. At present, we recommend using this type of convertible or modular stem to avoid having to extract a stem if the patient requires conversion to reverse arthroplasty at some point in the future. A partial rotator cuff repair is undertaken then with the sutures passed through the greater tuberosity. The subscapularis is then closed through drill holes in the lesser tuberosity or by using a tendon-to-tendon stitch (see Pearls and Pitfalls and Table 9.3).

Several classic articles must be mentioned when discussing the outcomes of hemiarthroplasty in the cuff-deficient shoulder. Williams et al. reviewed the results of 21 hemiarthroplasties performed in the setting of rotator cuff deficiency [3]. Using Neer’s grading scale for shoulder arthroplasty, there were no excellent results, 14 satisfactory results, and 7 unsatisfactory results. Active forward flexion improved from 70° preoperatively to 120° postoperatively. Similarly, Zuckerman et al. reviewed the results of 15 hemiarthroplasty for rotator cuff arthropathy. At a mean follow-up of 28.2 months, patients gained a modest improvement in forward elevation from 69° to 86°; 14 of 15 patients had significant pain relief [4].

Sanchez Sotelo and the group from the Mayo Clinic evaluated the results of hemiarthroplasty for glenohumeral arthritis associated with rotator cuff deficiency. At an average of 5-year follow-up, 33 shoulders showed an average gain of forward elevation to 91° from 72°. There were 11 unsuccessful results [5].

Taken together, these historical results of hemiarthroplasty in the setting of CTA show that while hemiarthroplasty clearly provides pain relief, functional improvement is less predictable with many patients improving their range of motion only marginally; also, there is a subset of patients (up to a third) who remain dissatisfied with the operation.

More recent results echo these older results. Vistosky et al. retrospectively reviewed the results of 60 patients that received hemiarthroplasty with an extended coverage humeral head for treatment of rotator cuff tear arthropathy [6]. These authors showed the most dramatic improvement in range of motion across the previous studies mentioned with forward elevation improving from 56° to 116° in patients with 2-year follow-up. Importantly, the visual analog pain scale improved from 9.3 preoperatively to 1.9 postoperatively. To date, there have been no comparative trials evaluating the effectiveness of the extended coverage humeral head (CTA head) in comparison to a conventional hemiarthroplasty.

Goldberg et al. retrospectively evaluated the results of hemiarthroplasty in 31 patients with 34 cuff-deficient shoulders [7]. Average forward elevation improved from 78° to 111° postoperatively. The most important finding of this study was that patients that had preoperative forward elevation of greater than 90° had higher final ASES scores (both total and functional) and better pain relief when compared to patients with worse preoperative motion. In addition to hemiarthroplasty in the study, all patients in these series had an attempt at partial or complete cuff repair with a trend toward improved results in those patients in which a repair was possible.

While it is clear that hemiarthroplasty can provide some benefit to certain patients, there are data that show that among indications for hemiarthroplasty, CTA is one of the least favorable. Hettrich et al. retrospectively studied 71 hemiarthroplasties in an attempt to identify preoperative factors associated with a good functional result [8]. An intact rotator cuff was a predictor for improved function postoperatively and a diagnosis of cuff tear arthropathy (along with capsulorrhaphy arthropathy and rheumatoid arthritis) showed the least functional shoulder improvement.

In a large study (272 shoulders) evaluating the outcomes and long-term survival analysis according to etiology, Gadea et al. showed that the rotator cuff arthropathy as a preoperative diagnosis resulted in a 10-year survival rate of 81.5 % but a low Constant-Murley score (46.2) [9]. The authors concluded that the best indications for hemiarthroplasty in their group of patients was avascular necrosis and that the worst indications were cuff tear arthropathy and fracture sequelae.

Although there are few prospectively collected comparative data regarding CTA for hemiarthroplasty, these two aforementioned studies reveal that the results of hemiarthroplasty for CTA are likely to inferior to the results of hemiarthroplasty for other indications in the cuff-intact shoulder such as avascular necrosis. Due to the relatively poor results with hemiarthroplasty in CTA, shoulder arthroplasty surgeons have turned increasingly to the reverse arthroplasty in the cuff-deficient shoulder [10]. However, since the advent of the reverse replacement, there have been very few studies that actually directly compare the results of reverse arthroplasty with conventional arthroplasty.