Glenohumeral Osteoarthritis: Anatomic Total Shoulder Arthroplasty With Lesser Tuberosity Osteotomy and Anteroinferior Capsulectomy



Glenohumeral Osteoarthritis: Anatomic Total Shoulder Arthroplasty With Lesser Tuberosity Osteotomy and Anteroinferior Capsulectomy


Grant E. Garrigues

Gerald R. Williams Jr



INTRODUCTION

Anatomic total shoulder arthroplasty (aTSA) involves replacing the humeral head with a convex bearing that approximates the humeral head dimensions and resurfacing the glenoid with a concave component that roughly approximates the glenoid size and shape. Within that generic definition, there is an immense variety of designs and even some variety of materials.

There is some debate about the implantation of the first total shoulder arthroplasty (TSA), with credit frequently given to the Parisian surgeon Jules Pean who, in 1893, implanted a platinum and rubber prosthesis for destructive tuberculosis of the glenohumeral joint and published his case. However, Pean himself gave credit for the inspiration to Themistocles Gluck, a German surgeon, who was using ivory prostheses at that time for a variety of joint replacements and may have preceded Pean.1 These designs, however, were far from anatomic. More modern designs through the end of the 20th century began to approximate our current technology with advances such as the all-polyethylene glenoid by Neer, humeral modularity by Fenlin, and offset tapers by Walch and Boileau.2

Current areas of design innovation include alternative bearing surfaces (eg, ceramic, pyrocarbon, treated titanium, and vitamin E cross-linked polyethylene),3 alternative glenoid fixation strategies (eg, metal-backed, platform, and hybrid components), and alternative humeral designs (eg, mini-stem, stemless, and platform stems).4




INDICATIONS AND CONTRAINDICATIONS

The indications and contraindications for the anatomic total shoulder are enumerated in Table 46-1. They can be summarized as requiring an infection-free shoulder with bone and soft tissues competent to support and power a stable prosthesis in a patient able to follow a postoperative rehabilitation course.









A robust debate currently rages regarding when to perform reverse total shoulder arthroplasty (rTSA) and when to perform aTSA, given the significant overlap in the accepted indications. The rTSA predated the work of Grammont, but Grammont’s Delta design, with results first published in 1993, was the first with reasonable results and complication profile.5 This revolutionized the field of orthopedic surgery as the reversed articulation and semiconstrained bearing allowed patients without a functioning rotator cuff to regain active motion with a low incidence of glenoid failure, while aTSA in this same setting of massive rotator cuff tear failed due to the “rocking horse” edge loading and poor function.6 Initially, rTSA was viewed as a salvage operation, and reports of these complex cases showed a predictably high complication rate (up to 50%) in these challenging scenarios.7

Since that time, however, the complication rate of rTSA has fallen significantly, to the point where it is now comparable with aTSA.8 Furthermore, the indications for rTSA have expanded dramatically to include not just rotator cuff tear arthropathy but also revision shoulder arthroplasty, arthroplasty for tumor resection, proximal humerus fractures, and glenohumeral instability.9 With excellent results, the indications for rTSA have further expanded to include conditions with “cuff at risk” such as inflammatory arthropathy including rheumatoid arthritis and patients with thin or partially torn rotator cuffs. Finally, while US Food and Drug Administration approved, in the minds of many surgeons, rTSA indications have been expanded further to include patients with a completely intact rotator cuff and no glenoid bony erosion—heretofore the ideal indication for aTSA. Thus, as of the writing of this chapter, there is no indication for aTSA that is not also a potential indication for rTSA.

So, what are we to make of this overlap? The best available data currently suggest that implant longevity between aTSA and rTSA is roughly the same when metal-backed and non-cross-linked components are excluded.10,11 Registry data from the Australian registry show a similar revision rate, with rTSA having a higher revision rate in the first 3 months (typically for the complication of instability), but at 4 years the revision rates were statistically equivalent.8 These data are especially interesting given that the registry outcomes are nonrandomized, with implants chosen by the surgeon, and the rTSA group tended to be older and female and had lower body mass index, which might indicate a higher demand in the aTSA group.

Many surgeons postulate that aTSA requires a pristine rotator cuff for full function and longevity of the implant. However, it should be noted that postoperative rotator cuff failure is a very uncommon cause of failure of aTSA—as low as 0.8% at 12 years in the Australian registry.12 Furthermore, even small tears—if isolated to the supraspinatus and without fatty infiltration of the infraspinatus—have shown no negative ramifications even if noted and not repaired at the time of surgery.13

Many comparative outcome studies consistently show the outcome of aTSA is better concerning range of motion—particularly rotational motion. It is not clear whether the loss of rotation in rTSA is a function of impingement-free range of motion, humeral version, rotator cuff integrity, scapular control, or some other factor, but, regardless of the cause, studies support better rotational motion with aTSA.14, 15 and 16

Despite the similar or increased longevity of aTSA and the greater range of motion, usage data show that the rate of rTSA is increasing two times faster than that of aTSA.17 This may be because, due to the diminished importance of soft tissue healing, rTSA does not require postoperative formal physical therapy and may allow a more rapid return to activities of daily living.18, 19 and 20 rTSA is a technically much easier and, in many ways, a more forgiving operation given a diminished need for preservation of bone and soft tissue integrity, and it is possible that this explains at least some of its
popularity.14 Exposure can be achieved more easily by removing extra bone or releasing portions of the rotator cuff. The implant position for rTSA is more forgiving as rTSA seems less sensitive to nonanatomic version than aTSA, reaming through the subchondral bone is not as problematic, full backside contact is not required, and the joint line medialization can be accounted for with prosthetic adjustments through the glenosphere and the humeral liner.21,22 Furthermore, soft tissue balancing is less relevant given the semiconstrained nature of the bearing.

Regardless, the fact remains that a well-done anatomic total shoulder has the potential to recreate a normal shoulder range of motion and kinematics. In this chapter, we will thoroughly outline the senior author’s (GRW) technique for aTSA. This technique is designed to improve exposure and maximize the potential benefits of aTSA including postoperative range of motion and minimize complications including subscapularis insufficiency, posterior instability, and glenoid loosening.


Rationale

aTSA requires intact soft tissues to afford both power and stability to the joint, sufficient bone for stable prosthesis implantation in an anatomic position on both the humerus and glenoid, and a supple joint to allow motion. The technique of lesser tuberosity osteotomy (LTO) and anteroinferior capsulectomy has reliably allowed for early range of motion and a more complete return range of motion for the authors.

Anatomic reconstruction of the humerus requires respecting the normal anatomic relationships of the proximal humerus.23 Notably, while the version, offset between the epiphysis and metaphysis, and inclination vary widely in native shoulders, the ratio of head height to circumference is well conserved.24 The authors do not utilize humeral heads with nonanatomic ratios as we believe that reconstructing the bony anatomy and releasing or resecting pathologic soft tissues is the ideal way to balance the shoulder—in distinction to the use of nonanatomic humeral heads. In fact, we prefer a humeral reconstruction consistent with the circle method of Iannotti (Figure 46-1).25 Assuming adequate bony fixation can be achieved, a stemless anatomic humeral head is frequently the ideal choice as it allows humeral head prosthetic reconstruction independent from the variable metaphyseal and diaphyseal alignment.






Implanting the glenoid component is frequently the most challenging aspect of aTSA. We recommend correcting the version to within 10° of neutral, the joint line to native, and the inclination to not more than 10°.26 For Walch grade B2 and B3 glenoids, augmented anatomic glenoids have shown excellent short-term results and avoid the problems of excessive retroversion, over medialization, and peg penetration that plagued the use of nonaugmented glenoids in cases of posterior glenoid erosion.27, 28 and 29



Anteroinferior Capsulectomy

The anteroinferior capsule is pathologic in glenohumeral osteoarthritis. Basic science data have shown that the capsule plays an active role in the pathogenesis of glenohumeral osteoarthritis with clear pathological evidence of increased collagen deposition, contractile elements (myofibroblasts), inflammation, and fibrosis (M2 macrophages and synovitis).30 Cadaveric models have shown that a tightened anterior capsule leads to posterior humeral head subluxation and increased joint reactive forces.31 This is in keeping with imaging data from computed tomography (CT) and magnetic resonance imaging (MRI) showing that increased thickening of the anterior capsule is correlated with posterior humeral head subluxation.14 Furthermore, evaluation of retrieved humeral heads from patients with osteoarthritis shows chondral loss proceeds from the central portion of the humeral head with the peripheral cartilage frequently spared in cases with limited range of motion.32 All these data indicate that the anteroinferior capsule pathologically contributes to increased joint reactive forces, posterior humeral head subluxation, and inflammation that leads to glenohumeral osteoarthritis. For these reasons, we favor removing this tissue.

Some have questioned whether the removal of the capsule will lead to glenohumeral instability; however, it must be recalled that the primary stabilizer of the shoulder is the rotator cuff.33 To this end, we believe, just as with any rotator cuff repair, that the complete rotator cuff releases affected by capsulectomy minimize tension on the LTO/subscapularis repair. Data have shown a very low rate of LTO failure with 6.7% early in GEG’s career34 and no cases of LTO failure in the last 2000 cases. Lastly, we believe that anteroinferior capsulectomy leads to an improved range of motion and exposure, but the authors are unwilling to randomize patients to perform the former study and the latter is difficult to quantify.


Lesser Tuberosity Osteotomy

There are multiple methods for the management of the subscapularis during TSA, including tenotomy, peel, LTO, and subscapularis-sparing techniques. Historically, multiple studies have shown a high incidence of subscapularis insufficiency—up to 46%.35, 36 and 37 More recently, multiple studies have shown excellent results with LTO.38,39 However, randomized trials have shown a high rate of statistically equivalent success with multiple methods.40,41 We prefer LTO because it is a subscapularis-sparing technique that does not violate the subscapularis muscle or tendon in any way while giving excellent exposure.42 In addition, studies show that LTO does not compromise fixation in stemless TSA.43 Additionally, while fibrous healing would be acceptable and equivalent to peel or tenotomy, the vast majority of cases achieve bone-to-bone healing. Bone is the only tissue in the body that heals without scar—the remodeled bone is eventually histologically equivalent in structure, function, and biomechanics to the original.44 Furthermore, we find the ability to monitor the integrity of the LTO on routine postoperative radiographs incredibly helpful should the patient or therapist have a concern about the repair integrity or healing.


PREOPERATIVE PREPARATION

The preoperative history and physical examination will show the classic signs and symptoms of end-stage glenohumeral osteoarthritis—shoulder pain and limited range of motion. Eliciting a history of prior surgical procedures on the shoulder is critical as many surgical approaches to the shoulder can alter the integrity of the rotator cuff or the anatomic relationships. This is especially true for a history of prior rotator cuff repairs, which is uncommon for patients with primary glenohumeral osteoarthritis, and when this history is present, it can be concerning for rotator cuff integrity. Furthermore, anterior instability surgery, frequently employed in younger patients, can develop into instability arthropathy or capsulorrhaphy arthropathy and may involve either subscapularis dysfunction or altered anatomy in the case of Latarjet, Magnuson-Stack, or Putti-Platt operations, for example. In addition to prior surgery, a history of prior shoulder fractures or infections should also be taken.

Note that some patients will complain of instability. Rotational motion and translational motion are coupled in the shoulder.31 Thus, if the rotational range of motion is significantly decreased, as it typically is, true glenohumeral instability is not possible. We refer to this phenomenon as “saltatory motion” of the glenohumeral joint as the higher coefficient of static friction of the arthritic joint causes the affected shoulder to seize up and then suddenly “jump” leading to the perception of instability.


Physical examination should include an assessment of range of motion, neurovascular integrity, and rotator cuff strength. Often, the tests for subscapularis function, belly press, lift-off, and modified lift-off,45 are not possible as the internal range of motion may be decreased. The bear hug test and advanced imaging are helpful in this scenario.

Plain radiographs should include the Grashey view, axillary lateral, and scapular Y view (Figure 46-2). For patients proceeding to surgery, we prefer a CT scan for preoperative planning. The CT scan allows assessment of the relevant metric of fatty infiltration of the rotator cuff13 (note the Goutallier scale is a CT-based grading scale46). Furthermore, the excellent bony definition allows the use of not only coronal and sagittal reformats but also 3D volume rendering and proprietary implant templating software.26