Total shoulder arthroplasty has gone through several generations, as instruments and implant designs have given surgeons both more options in the alignment of the components and more guidance in the best choices to make. However, while the measurement of alignment has become more sophisticated, the importance of particular aspects of alignment to actual patient comfort and function has been less completely characterised.

Overstuffing of the joint and proud humeral heads have been most associated with clinical failure. The efforts to avoid this can be divided into two camps: 1.) The anatomic school, who believe an experienced surgeon can divine the correct anatomy that existed before the distortions of arthritis began, and that the surgeon should make free-hand cuts and alignments to restore the normal anatomy. 2.) The cutting-guide school, who believe that average versions and positions avoid error and that soft-tissue balancing requires occasional deviations from “normal” anatomy.

Reverse total shoulder replacement in contrast is a semi-constrained implant, with built-in “internal impingement” at the extremes of motion, which can cause notching and/or instability (levering out). Initial European experience favored placing the humeral component in 0 degrees, but most surgeons have gravitated toward 15–20 degrees of retroversion to allow easy conversion from/to a hemiarthroplasty as needed. Increased retroversion may block internal rotation, and increased anteversion limits external rotation.

Displaced proximal humerus fractures remain a difficult clinical problem, and techniques as diverse as percutaneous pinning, locked plating, intramedullary nailing, and shoulder arthroplasty have been proposed. In recent years, reverse total shoulder arthroplasty (RTSA) has become a very popular option to fix just about any fracture. However, RTSA is not without risk, with complications ranging from infection, instability, acromial stress fractures, aseptic loosening, notching and more. In a 2017 study on 39 patients, Tokish et al. compared non-operative treatment to reverse shoulder arthroplasty for displaced 3- and 4-part fractures. There was no difference in pain, range of motion or outcome scores between the two groups. Among the patients who underwent RTSA, there was no difference between early (<30 days) and late (>30 days) surgery suggesting that it could be safe to attempt a non-operative trial in most patients and see how they do. This is also supported by a 2016 study by Sanchez-Sotelo et al. in which they compared 18 patients with primary RTSA to 26 patients with failed ORIF who underwent salvage RTSA. There was no difference in ASES score, ROM and overall satisfaction between both groups suggesting that an ORIF can be attempted in many patients without the fear of compromising a revision RTSA. And although RTSA may provide more predictable results, in a properly selected patient, a well-executed hemiarthroplasty can outperform an RTSA. In a study from Molé et al., 38 patients were randomised to either RTSA or to a hemiarthroplasty. In the hemiarthroplasty group, half of the patients had <90 degrees of forward elevation and half the patients had >120 degrees of forward elevation showing a bi-modal distribution dependent on tuberosity healing. In the RTSA group, however, while having an average of 115 degrees of forward elevation, 68% of patients had less than 120 degrees of forward elevation. While RTSA is a great tool to treat complex displaced comminuted fractures in elderly patients with poor bone quality, it should not be blindly applied to all fractures types and all patients.

Shoulder arthroplasty procedures continue to increase in prevalence and controversy still remains about the optimal method to manage the subscapularis. Scalise et al. performed an analysis of 20 osteotomies and 15 tenotomy procedures, and found the tenotomy group had a higher rate of abnormal subscapularis tendons on ultrasound examination. There was one tendon rupture in the subscapularis tenotomy group and no ruptures in the osteotomy group. Jandhyala et al. retrospectively examined 26 lesser tuberosity osteotomies and 10 subscapularis tenotomies for arthroplasty, and their study demonstrated a significant improvement in the belly press test for the osteotomy group. Lapner et al. performed a randomised controlled trial assigning patients to either a lesser tuberosity osteotomy or a subscapularis peel procedure. They evaluated 36 osteotomies and 37 subscapularis peels. The outcomes evaluated were Dynamometer internal rotation strength, the Western Ontario Osteoarthritis of the Shoulder Index (WOOS) score and American Shoulder and Elbow Surgeons (ASES) score, and in a subsequent paper they evaluated the healing rates and Goutallier grade. Their studies illustrated no difference in the internal rotation strength between groups. Both groups significantly improved WOOS and ASES scores postoperatively, but the difference was not significant between groups. Goutallier grade increased significantly in both groups, but there was no significant difference between the groups. Overall, the different approaches have not demonstrated a meaningful clinical difference. Further studies are needed to help understand issues leading to subscapularis complications after arthroplasty.

Challenging cases in shoulder surgery emphasizing joint reconstruction will be presented to a distinguished panel of experts. Audience participation will be encouraged.

Preoperative assessment, imaging, operative techniques, and postoperative care will be emphasised. Special focus will be on shoulder replacement, especially reverse shoulder arthroplasty.

Reverse TSA initially followed Grammont's dictum that the center of rotation (COR) must be in the bone (“medial” COR). Others have argued for a more lateral COR, which can be a challenge if glenoid bone stock has been medially eroded. When bone loss must be made up, and/or the COR lateralised, the options include use of bone graft or use of metal.

Metal constructs produce a cantilever-loading situation, with substantial bending moments applied to the bone-implant junction. Use of bone graft allows remodeling with living bone, so that ultimately the forces are applied to the bone-implant junction in a more compressive pattern.

The author's preference is to have at least 30% of the circumference of the baseplate contact living bone while the rest may be made up with bone graft which can remodel. It is important to have a deep keel penetrate the cortex medially.

Challenging shoulder cases will be presented and discussed by a panel of experts in the field. Emphasis will be on restoring glenohumeral anatomy, repairing or reconstructing the rotator cuff, and supervising rehabilitation. Different surgical options and new emerging technologies will be reviewed while highlighting the pros and cons of each.

Total shoulder arthroplasty has gone through several generations, as instruments and implant designs have given surgeons both more options in the alignment of the components and more guidance in the best choices to make. However, while the measurement of alignment has become more sophisticated, the importance of particular aspects of alignment to actual patient comfort and function has been less completely characterised.

Overstuffing of the joint and proud humeral heads have been most associated with clinical failure. The efforts to avoid this can be divided into two camps:

The anatomic school, who believe an experienced surgeon can divine the correct anatomy that existed before the distortions of arthritis began, and that the surgeon should make free-hand cuts and alignments to restore the normal anatomy.

The cutting-guide school, who believe that average versions and positions avoid error and that soft-tissue balancing requires occasional deviations from “normal” anatomy.

Total shoulder arthroplasty with a cemented polyethylene glenoid is a proven technology with excellent survivorship, even in young, active patients followed for over 20 years.

Despite numerous efforts to avoid cement fixation, uncemented technology has failed to improve survival even in the short and middle term, and in fact has usually yielded worse results at early follow-up – hardly promising that further follow will reveal an advantage.

There have been two approaches – metal-backed designs and all-poly “magic peg” designs. There are two major problems with metal backing: there can be technical issues and fretting failure with snap-fit mechanisms, and the stiff metal backing induces early polyethylene wear. Although some early studies report cautious optimism, larger series with careful, long follow-up have reported poor results. Some have attempted to use less-stiff metal but the value is still unclear. While early results from all-poly uncemented designs have less early failure, the wisdom of having long-term close apposition of polyethylene with bone, the large bone-removal for most large-peg designs, and the unknown long-term mechanical failure mechanisms suggest caution.

Further, early work on eliminating cement was prompted by a desire to avoid “cement disease”, while more recent studies across multiple joints suggest that loosening and bone resorption are more likely due to polyethylene particles than any effect of the cement. Since most uncemented designs stiffen the basic polyethylene surface with larger pegs or metal-backing, the wisdom of the entire effort is debatable. In 2016 optimal glenoid fixation requires cement.

Analysis of orthopaedic malpractice claims has shown that highest impact allegations (highest payment dollars per claim) were those that were related to failure to protect anatomic structures in surgical fields. The prevalence of subclinical peripheral neurologic deficit following reverse and anatomic shoulder arthroplasty has been reported to be 47% and 4%, respectively. We propose the following five rules in order to avoid neurovascular injury during shoulder arthroplasty cases:

Pre-operative planning would assure a smooth operation without intra-operative difficulties. Adequate planning would include appropriate imaging, obtaining previous operative reports, complete pre-operative neurovascular examination and requesting the necessary operative equipment.

Neuromonitoring has been utilised in shoulder surgery in the past. Nagda et al showed that nerve alerts during shoulder arthroplasty occurred 56.7% of the time and 50% of the events were with the arm in abduction, external rotation and extension; 76.7% of signals returned to normal with retractor removal and change in arm positioning. We recommend removing all retractors and returning the arm to neutral position several times during surgery, especially during the glenoid exposure when the arm is in abduction and external rotation.

Newer commercially available nerve stimulators are extremely useful in locating and protecting neurovascular structures. We recommend brachial plexus exploration and axillary nerve dissection with the aid of a nerve stimulator in all revision cases.

Availability of a nerve/microvascular surgeon as an assistant in revision cases for brachial plexus exploration using a microscope is crucial for successful revision surgery.

Total shoulder arthroplasty has gone through several generations as instruments and implant designs have given surgeons both more options in the alignment of the components and more guidance in the best choices to make. However, while the measurement of alignment has become more sophisticated, the importance of particular aspects of alignment to actual patient comfort and function has been less completely characterised.

Overstuffing of the joint and proud humeral heads have been most associated with clinical failure. The efforts to avoid this can be divided into two camps:

The anatomic school, who believe an experienced surgeon can divine the correct anatomy that existed before the distortions of arthritis began, and that the surgeon should make free-hand cuts and alignments to restore the normal anatomy

Shoulder arthroplasty procedures continue to increase in prevalence, and controversy still remains about the optimal method to manage the subscapularis. Scalise et al. performed an analysis of 20 osteotomies and 15 tenotomy procedures, and found the tenotomy group had a higher rate of abnormal subscapularis tendons on ultrasound examination. There was one tendon rupture in the subscapularis tenotomy group and no ruptures in the osteotomy group. Jandhyala et al. retrospectively examined 26 lesser tuberosity osteotomies and 10 subscapularis tenotomies for arthroplasty, and their study demonstrated a significant improvement in the belly press test for the osteotomy group. Lapner et al. performed a randomised controlled trial assigning patient to either a lesser tuberosity osteotomy or a subscapularis peel procedure. They evaluated 36 osteotomies and 37 subscapularis peels. The outcomes evaluated were Dynamometer internal rotation strength, the Western Ontario Osteoarthritis of the Shoulder Index (WOOS) score and American Shoulder and Elbow Surgeons (ASES) score, and in a subsequent paper they evaluated the healing rates and Goutallier grade. Their studies illustrated no difference in the internal rotation strength between groups. Both groups significantly improved WOOS and ASES scores post-operatively, but the difference was not significant between groups. Goutallier grade increased significantly in both groups, but there was no significant difference between the groups. Overall, the different approaches have not demonstrated a meaningful clinical difference. Further studies are needed to help understand issues leading to subscapularis complications after arthroplasty.

The incidence of total shoulder arthroplasty continues to increase. The most common reason for failure of a total shoulder arthroplasty is the glenoid component. Two styles of cemented all-polyethylene components are commonly implanted. These two styles are pegged glenoid and keel glenoid components. Data regarding the superiority of the styles has focused on radiolucent lines, complete loosening and need for revision procedures. Several retrospective and randomised controlled trials have been published to examine these endpoints. There is a trend in the literature to demonstrate decreased rates of radiolucent lines with pegged glenoid components, but a recent systematic review of available trials did not demonstrate a significant difference in the rate of radiolucent lines between the two styles. A slightly increased rate of revision was noted for the keel components. Overall pegged and keel glenoids both still represent good options in total shoulder arthroplasty.