Abstract
Resurfacing is a type of shoulder arthroplasty that involves replacing the humeral joint surface with a metal covering, or cap, thus preserving the bone of the proximal part of the humerus. If the glenoid is also replaced, a current conventional polyethylene glenoid replacement prosthesis or an interposed soft-tissue graft is used.
The potential advantages of humeral resurfacing, as compared with conventional shoulder arthroplasty, are: (1) no osteotomy is performed (and thus the head-shaft angle does not have to be addressed); (2) minimal bone resection; (3) a short operative time; (4) a low prevalence of humeral periprosthetic fractures; and (5) ease of revision to a conventional total shoulder replacement, if needed.
Outcomes of surface replacement arthroplasty have been comparable with those of arthroplasties with a stemmed prosthesis in numerous short and mid-term follow-up studies.
Future studies are required to assess the long-term outcomes of humeral resurfacing and to evaluate alternative surface bearing materials, especially on the glenoid side.
Resurfacing appears to be a viable option for shoulder replacement, especially in young patients.
Over the past twenty years, shoulder resurfacing arthroplasty has gained popularity as an alternative to conventional shoulder arthroplasty for the treatment of glenohumeral arthropathy. In contrast to conventional shoulder arthroplasty, which involves removal of the entire humeral head followed by placement of an intramedullary stem into the proximal aspect of the humerus, shoulder resurfacing consists of reaming the proximal portion of the humeral head and fitting a metal-alloy cap over the remainder of the head. This cap may or may not be mated against a glenoid component1,2.
The early clinical results, developments, and new innovations of shoulder resurfacing have paralleled similar advances in hip resurfacing arthroplasty3. Resurfacing of the proximal part of the femur (hemiresurfacing or total hip resurfacing arthroplasty) has been performed for more than thirty years, and there has been a resurgence in its use over the past ten years with advances in metallurgy and design4. However, there are functional and anatomic differences between the two joints: the shoulder joint generally supports lower loads, has a greater range of motion, and has a decreased offset (the displacement of the head of the long bone in the medial/lateral, anterior/posterior, and superior/inferior directions).
In this article, we will review shoulder resurfacing by describing its indications and contraindications, the history of its development, the biomechanical rationale for the various resurfacing designs, the operative techniques for shoulder resurfacing procedures, and the short and mid-term results of clinical series. Finally, we will also review various resurfacing innovations and assess the future of this procedure.
To the best of our knowledge, Steffee and Moore reported on the first humeral resurfacing procedures, which were performed in the late 1970s with use of a resurfacing hip implant5. The radius of curvature of the prosthesis was soon modified to more precisely fit the dimensions of the humeral head. This and other early prostheses were made of stainless-steel and had no central stem. They were fixed to the proximal aspect of the humerus with methylmethacrylate cement5,6. All of the procedures were performed as hemiarthroplasties, with the humerus, but not the glenoid, being resurfaced.
During the same time period, Copeland began researching and developing a cementless surface replacement7. The prosthesis that he developed consisted of a central pegged humeral component that was secured by a screw through the lateral cortex combined with a polyethylene glenoid element that was secured by a peg. The humeral screw often loosened, and in vitro testing suggested that it was not contributing to fixation. The screw was eliminated in a later version of this prosthesis, and a metal-backed glenoid component was added. In 1993, a hydroxyapatite coating was added to the humeral and glenoid components in an attempt to decrease the prevalence of loosening.
Other prosthetic designs have been used for humeral resurfacing. Current designs all have a central humeral stem, and these stems have variable shapes, diameters, and lengths. Most humeral resurfacing components are composed of cobalt-chromium alloy, although some are made of titanium alloy. Most prostheses are implanted into the humerus with use of a press-fit method and have a hydroxyapatite or ceramic porous coating to encourage bone ingrowth, but some designs are fixed with cement. Figures 1-A, 1-B, and 1-C show some examples of modern humeral prosthetic designs.
Partial resurfacing of the humeral head is a newer concept that has been recently incorporated into a prosthetic design. This implant consists of two parts: a tapered post, which is a headless titanium-alloy cannulated screw, and an articular cobalt-chromium-alloy surface component, which has a small central peg that mates with the post (Fig. 2). This may be useful for the treatment of asymmetric chondral defects of various sizes. The key feature is that the component can be matched by size and shape to the articular surface for partial resurfacing to address lesions of various sizes.
Normal shoulder anatomy can vary considerably among different individuals as well as between the left and right shoulders of the same individual8,9. The normal humeral head is retroverted and is inclined medially relative to the humeral shaft. However, normal humeral head retroversion may range from 0° to 55°, and the inclination may range from 30° to 55°9. In addition, the offset of the humerus in relation to the glenoid may vary in three dimensions, and the radius of curvature ranges from 20 to 30 mm9. All of these variations must be considered when an arthroplasty is performed.
Changes in the angle or offset of the humerus during a shoulder arthroplasty can adversely affect the biomechanics of the joint. Alterations of the retroversion or inclination of the humeral head might change the tension and/or the lever arm of the deltoid and rotator cuff muscles, which may lead to a decreased range of motion, weaker flexion, or instability of the joint. Changes in the offset of the humeral head may result in impingement with the acromion or the glenoid rim, increased tension of the rotator cuff tendons, or a decreased range of motion10. An increase or decrease in the radius of curvature of the humeral head by 5 mm may decrease the range of motion by 20° to 30°, which may increase the extent of glenohumeral translation during movement11,12. These factors could negatively impact the function of the joint as well as patient satisfaction.
With resurfacing arthroplasty, the humeral neck and >50% of the humeral head are retained, which is beneficial with regard to restoration of the biomechanics of the shoulder joint (Fig. 3). The native head-shaft angle remains intact because no osteotomy of the neck is performed. In addition, some authors believe that it is simple to restore the anatomic offset as well as the height of the center of instant rotation with resurfacing13. If the center of rotation and radius of curvature are accurately determined during preoperative planning, then the optimal prosthesis size can be selected and the resurfaced humeral head will resemble the original anatomy. Accurate placement of the guide-pin during the procedure will help to recreate the anatomic version and inclination. Thomas et al.13 reported that resurfacing of the humerus increased the humeral offset by a mean of 5 mm (from 23 to 28 mm) but there was a mean preoperative erosion of 6 mm of the lateral offset, so the resurfacing reestablished the anatomic offset that had been lost as a result of erosion. Many authors have reported that the inclination, version, offset, and head-shaft angles were maintained with resurfacing1,2,7,13-15. Finally, partial resurfacing prostheses do not change the offset or center of rotation at all, as they are inserted into the bone defect16; this might optimize the biomechanics, although we are not aware of any published studies in which this was examined.
Indications for shoulder resurfacing include pain and/or decreased function that are not treated successfully with nonoperative means. Etiologies of this pain include osteoarthritis, rheumatoid arthritis, osteonecrosis, posttraumatic arthritis, cuff tear arthropathy, and chronic instability of the joint3,6,7,16.
In addition, shoulder resurfacing may be a relatively simple method for treating patients for whom an arthroplasty is indicated but who have extra-articular deformities of the proximal part of the humerus that can make the placement of a conventional stemmed prosthesis difficult or impossible7. Examples of such extra-articular deformities include angulations or curvatures of the humeral metaphysis or diaphysis secondary to congenital defects, metabolic bone disorders, or a malunited proximal humeral fracture. A resurfacing prosthesis can be easily placed on the humeral head in these patients, whereas a conventional stem would not pass through the curvature or angulation of the humerus. Similarly, a resurfacing prosthesis can be easily placed on a humerus that has hardware, such as an intramedullary nail or screws, that would impede the passage of a conventional stem.
Contraindications for shoulder resurfacing include four-part fractures of the humerus and inadequate bone stock1. On the basis of their experience, Copeland7 and Thomas et al.13 reported that humeral resurfacing requires a minimum of 60% of normal humeral head bone stock, so patients who have less may need a conventional stemmed arthroplasty. Our experience has confirmed that 60% is sufficient for resurfacing with bone-grafting, although, to our knowledge, there are no published data to support this. The presence of an extensive Hill-Sachs lesion17 may also preclude humeral resurfacing.
Preoperative evaluation includes anteroposterior, axillary, and lateral radiographs, which are used to measure the humeral head diameter and cartilage wear. Occasionally, computed tomography scans may be helpful, especially for the assessment of glenoid erosion.
The resurfacing procedure can be performed with general anesthesia, regional anesthesia, or a combination of both. The patient should be placed in a supine, beach-chair position. The deltopectoral approach5 and the anterosuperior approach18 have both been utilized successfully for this procedure. The anterosuperior approach is reported to allow easier access to the glenoid as well as the superior and posterior aspects of the rotator cuff1. Glenoid exposure is much more difficult during resurfacing than it is during a conventional stemmed humeral arthroplasty because the humeral head is not resected and therefore visualization of the glenoid is decreased. We have found that some additional releases, as described below, may help with exposure of the glenoid.
We use the anterosuperior approach, with a skin incision that runs 3 to 4 in (7.6 to 10.2 cm) distally from just posterior to the acromioclavicular joint (Fig. 4). The fibers of the anterior deltoid muscle are then split longitudinally, and the deltoid muscle is detached at its origin on the anterior part of the acromion. Next, the rotator cuff should be assessed. If it is intact or repairable, an anterior acromioplasty with partial resection of the coracoacromial ligament is performed. This may prevent future impingement problems. If the rotator cuff is severely torn and not repairable, the coracoacromial arch should be left intact because it acts to prevent further superior migration in the rotator cuff-deficient shoulder. If the patient has symptoms and radiographic signs of osteoarthritis of the acromioclavicular joint, an excision arthroplasty can be performed, which improves the exposure. Next, the rotator interval is divided longitudinally and the subscapularis muscle is detached at its insertion. Next, the inferior aspect of the capsule is released along the humeral neck and for 1 to 2 cm down the anterior aspect of the humeral shaft while the shoulder is progressively externally rotated to approximately 80° to 90°. This often allows sufficient exposure of the glenoid (Fig. 5), but if further exposure is needed a partial release or z-plasty of the pectoralis major tendon is performed. Any osteophytes on the glenoid rim must be removed to improve visualization. With the biceps tendon and the posterosuperior part of the rotator cuff retracted posteriorly, the humeral head is then dislocated anteriorly.
Preparation of the humeral head begins with the removal of all osteophytes. This is essential for complete visualization of the head and neck as well as accurate guide-pin placement. The key to precise anatomic positioning of the humeral component is proper identification of the anatomic neck and the head-shaft angle. The humeral drill guide is adjusted for proper anterior-posterior positioning and is centered on the humeral head, perpendicular to the plane of the anatomic neck, so that the anatomic retroversion and inclination will be maintained (Fig. 6). The guide-pin is then drilled into the center of the humeral head, the guide is removed, and the pin is checked for proper positioning (Fig. 7). Next, the head is shaped with an appropriately sized reamer to facilitate component fit and bone ingrowth (Fig. 8). Bone fragments obtained from the reaming at this stage may be used to fill minor osseous defects. After all remaining osteophytes have been removed, the trial component is placed over the guide-pin. The component-bone interface should be examined to ensure proper fit. If a hemiarthroplasty is being performed, the trial component is reduced into the glenoid and the humerus is moved through a full range of motion to assess stability and soft-tissue balance.
If glenoid resurfacing is to be performed, the humerus must be retracted gently to avoid damage to the humeral bone. A guidewire is drilled into the center of the glenoid. This wire is then used to guide a central taper hole drill followed by a glenoid surface shaper. Trial components are then inserted, and the construct is assessed for motion and stability. The trial components are then removed, and the final component(s) are impacted or cemented into place (Fig. 9).
Closure may involve a subscapularis step-cut lengthening if a glenoid component has been placed, because of the potential for compromise of rotator cuff function secondary to lateralization of the glenohumeral center of rotation1. The tensioning of the subscapularis muscle is important because, with overtightening, external rotation is limited and the risk of posterior subluxation or dislocation may be increased. Any rotator cuff tears may then be repaired, and final closure can begin. Suction drains may be placed, depending on the surgeon's preference. A compression dressing and a postoperative shoulder immobilizer are then applied.
Various postoperative rehabilitation protocols following humeral resurfacing have been described, with some authors allowing early mobilization and many recommending limited external rotation with no internal rotation for four to six weeks to allow the subscapularis tendon to heal1,14,15,19,20. We utilize both principles. Our patients immediately begin performing elevation and external rotation range-of-motion exercises. External rotation is limited to a maximum of 30° during the first four weeks. Internal rotation is avoided for the first two weeks and then is limited to a maximum of 30°. Four weeks after the surgery, the patients begin internal and external rotation exercises through a full range of motion. Additionally, the patients begin working with light-resistance exercise bands. Eight weeks following the surgery, the patients are allowed to begin exercise with heavier resistance as tolerated.
There have been a number of studies describing the outcomes of shoulder resurfacing arthroplasties performed over the past twenty years (see Appendix).
Results of Early Designs
In 1984, Steffee and Moore5 reported on humeral resurfacing in fifty-three patients (fifty-six shoulders) who had a mean age of sixty-five years (range, forty-eight to eighty-three years) and were followed for a mean of twenty-two months (range, one to sixty-two months). Initially, a hip resurfacing prosthesis was used, as the authors believed that it was similar in shape and size to the humeral head, but they later designed a humeral prosthesis that had a radius of curvature that was closer to that of the normal shoulder. The authors did not specify how many patients were treated with the hip prosthesis and how many were managed with the shoulder design. At the time of that study, there were no clinical functional scores available for the assessment of outcomes. However, the active range of motion improved markedly in almost all patients, especially those who had osteoarthritis. Seven of ten patients who had irreparable rotator cuff arthropathy achieved a nearly normal range of motion and had little pain.
In 2001, Levy and Copeland1 reported on ninety-four patients (103 shoulders) who had undergone shoulder resurfacing arthroplasty with use of a cementless cobalt-chromium humeral resurfacing prosthesis that had a fluted, tapered central stem. There were sixty-eight total replacements and thirty-five hemiarthroplasties. The patients had a mean age of sixty-four years (range, twenty-two to eighty-eight years) and were followed for a mean of seven years (range, five to ten years). At the time of final follow-up, 94% of the patients felt that the shoulder was improved. The authors concluded that the results of cementless surface replacement arthroplasty were comparable with published outcomes of conventional total shoulder arthroplasty, with decreased rates of complications involving the humeral shaft and periprosthetic fractures. In 2004, Levy and Copeland21 reported similar results for seventy-one patients (seventy-nine shoulders) who had undergone cementless shoulder resurfacing to treat osteoarthritis; forty-two total replacements and thirty-seven hemiarthroplasties were performed. Mean shoulder elevation improved from 60° to 126°, age-adjusted Constant scores (a 100-point scoring system that incorporates measures of pain, function, range of motion, and strength)22 improved from <25 points to >60 points, and 90% of the patients subjectively considered the shoulder to be better.
Newer-Generation Devices
Multiple recent studies have demonstrated successful results following the use of newer shoulder resurfacing prostheses. In 2005, in a study of fifty-two patients (fifty-six shoulders) who had undergone shoulder resurfacing, Thomas et al. reported that the mean Constant score improved from 16 points (range, 8 to 36 points) preoperatively to 54 points (range, 20 to 83 points) at the time of final follow-up at a mean of thirty-four months (range, twenty-four to sixty-three months)23. The authors concluded that these results were comparable with outcomes following use of modern stemmed prostheses reported in the literature. In another study, Thomas et al.13 examined the humeral head geometries and offsets of thirty-nine patients (thirty-nine shoulders) who had undergone shoulder resurfacing and concluded that surface replacement restores humeral offset as well as the lever arms of the deltoid and supraspinatus muscles while otherwise mimicking individual anatomy.
Buchner et al.15 compared twenty-two patients treated with shoulder surface replacement with a matched group of twenty-two patients who had undergone a conventional stemmed total shoulder replacement. The mean operative time, estimated blood loss, and length of the hospital stay were significantly decreased in the resurfacing group, but the total shoulder replacement group had a higher mean Constant score at the twelve-month follow-up evaluation (67 ± 12 points compared with 59 ± 15 points, p = 0.056). The patients who had undergone total shoulder arthroplasty had significantly better mobility, although the pain, strength, and range-of-motion scores of the two groups were similar. The authors concluded that the outcomes of surface replacement at one year were slightly inferior to those of arthroplasty with a stemmed prosthesis but that, because of the superior perioperative parameters as well as the preservation of bone stock, resurfacing could be a therapeutic option for younger patients who have primary arthritis. A case report by Crowther and Bell24 described the use of an uncemented cobalt-chromium humeral prosthesis along with soft-tissue resurfacing of the glenoid to treat neuropathic arthropathy in a forty-seven-year-old woman who had syringomyelia. At the twenty-four-month follow-up evaluation, the patient was entirely satisfied and reported no pain. Active elevation was 140°, external rotation was 50°, and the American Shoulder and Elbow Surgeons (ASES) score25 was 81.5 points. We are not aware of any other reports describing the use of shoulder resurfacing for treatment of neuropathic arthropathy, and longer follow-up of a larger cohort of patients will be necessary to determine the usefulness of this procedure in this patient group.
Resurfacing for Various Age Groups
Recent studies have revealed that humeral resurfacing can be used in the young, active population as well as in the elderly. Ellenbecker et al.19 reported the use of humeral resurfacing in a twenty-eight-year-old man who had sustained a shoulder injury while playing football at the age of sixteen years. The Single Assessment Numeric Evaluation (SANE) score26 improved from 2 points preoperatively to 90 points at the ten-month follow-up evaluation but decreased to 60 points at the two-year follow-up evaluation. Similarly, the ASES score improved from 17 points preoperatively to 85 points at the ten-month follow-up evaluation but decreased to 68 points at the two-year follow-up evaluation. The patient was able to return to his job as a laborer but without performing any overhead work or heavy lifting.
Bailie et al.14 reported the results of a prospective study on the use of an uncemented, cobalt-chromium prosthesis in thirty-six patients (thirty-six shoulders) who were younger than fifty-five years of age and had end-stage glenohumeral arthritis but not osteonecrosis. After a mean final follow-up time of thirty-eight months (range, twenty-four to sixty months), the clinical findings, as measured with a visual analog pain scale, the SANE scale, and the ASES scale, had improved significantly when compared with the preoperative assessments. Thirty-five of the patients reported that they had been able to return to their desired athletic activities at a satisfactory level, although six patients had voluntarily decreased their level of activity because of a fear of implant failure. The patients reported participating in resistance training, yoga, or Pilates exercises (thirty-four patients); mountain biking (four patients); golf (sixteen patients); hockey (two patients); tennis (four patients); power-lifting (three patients); bobsled push at the Olympic level (one patient); and competitive recreational basketball (six patients). All of the patients who were participating in sports activities were satisfied with the result, and the athletic activity levels were not associated with early failure. These results regarding the return to athletic activity are very promising, but the follow-up time was short and there was no comparison group. Longer follow-up will be necessary to better understand the impact of sports on the longevity of these prostheses.
Resurfacing has two other potential advantages for younger patients. Periprosthetic fractures, which are a concern in this more active population, are less likely to occur than they are with total shoulder replacement because the stem does not pass through the surgical neck. Another advantage is that bone stock is preserved with resurfacing. This is potentially beneficial for the younger patient, who may require revision to a total shoulder arthroplasty with a stemmed prosthesis during his or her lifetime.
Mullett et al.27 performed a prospective study of twenty-nine patients (twenty-nine shoulders) treated with an uncemented cobalt-chromium prosthesis when they were older than eighty years of age. The mean Constant score, adjusted for age and sex, improved from 15 to 77 points. There were no perioperative deaths or serious complications. The authors concluded that surface replacement may be performed for elderly patients with minimal morbidity and rapid rehabilitation.
Resurfacing in Patients Who Have Rheumatoid Arthritis
Many studies have specifically focused on the effects of shoulder resurfacing in patients who have rheumatoid arthritis. Jonsson et al. used hemiarthroplasty to treat rheumatoid arthritis in twenty-five patients (twenty-six shoulders) who had a mean age of sixty years28. All patients had reduced pain and an improved range of motion at a mean of two years. However, three patients had radiolucent lines that exceeded 1 mm in thickness, although none of those patients had pain at the final follow-up evaluation. Additionally, three patients initially had pain that resolved without treatment, and one patient had pain that decreased after an anterior acromioplasty. Rydholm and Sjogren6 reported on the use of a stainless-steel, cemented resurfacing prosthesis with no central stem in fifty-nine patients (seventy-two shoulders) with a mean age of fifty-three years who had rheumatoid arthritis. At a mean of four years, 94% of the patients were pleased with their pain relief and 82% reported improved shoulder mobility. Of note, 25% of the shoulders exhibited cup migration, loosening, or central glenoid wear, without an apparent effect on the clinical outcome. Alund et al.29 followed thirty-four patients (forty shoulders) with rheumatoid arthritis who had been treated with humeral resurfacing with use of a stainless-steel, stemless, cemented resurfacing prosthesis at a mean age of fifty-five years. At a mean of 4.4 years postoperatively, the mean Constant score was 30 points and one shoulder resurfacing had been converted to a stemmed prosthesis. The median visual analog pain score decreased from 80 mm preoperatively to 16 mm at the time of final follow-up, and twenty-six patients were satisfied with the result. However, glenoid erosion (mean, 2.6 mm) and proximal humeral migration (mean, 5.2 mm) occurred in most of the shoulders, possibly as a result of progressive rotator cuff destruction, although these factors did not correlate with shoulder pain or function.
Levy et al.2 reported on sixty-two patients (seventy-five shoulders) who had rheumatoid arthritis and who had been treated with a cementless, hydroxyapatite-coated cobalt-chromium prosthesis; there were thirty-three hemiarthroplasties and forty-two total shoulder arthroplasties. The mean Constant scores were 48 points (age and sex-adjusted score, 71%) in the hemiarthroplasty group and 53 points (age and sex-adjusted score, 76%) in the total shoulder replacement group. Fink et al.30 reported on a prospective study in which a cemented cobalt-chromium prosthetic cup had been used in thirty-nine patients (forty-five shoulders) who had rheumatoid arthritis, with different degrees of rotator cuff disease. Fifteen patients had an intact cuff, eighteen patients had a partial tear or a repaired rotator cuff, and twelve patients had a massive cuff tear. The mean Constant score for all groups increased from 20 points preoperatively to 63 points at the time of the latest follow-up. All of the patients who had undergone the procedure were pain-free at the time of the latest follow-up. The authors concluded that cup arthroplasty is a good alternative to other kinds of shoulder endoprostheses in patients who have rheumatoid arthritis with or without a massive rotator cuff tear. In a similar study, Fuerst et al.31 reported the mid-term results of the use of a cemented cobalt-chromium prosthesis in twenty-nine patients (thirty-five shoulders) who had rheumatoid arthritis. The patients had a mean age of sixty-one years and were followed for a mean of seventy-three months. The mean Constant score increased from 21 points preoperatively to 64 points at the latest follow-up evaluation. Proximal migration of the cup occurred in 63% of the shoulders, and glenoid depth increased in 31%. However, no difference in clinical outcome was noted between the patients who had a massive rotator cuff tear and those who had a smaller or no tear. In another recent study, Raiss et al. reported the clinical and radiographic results in ten patients who had rheumatoid arthritis at a mean of twenty-four months postoperatively20. The mean age at the time of surgery was sixty-seven years. The mean Constant score in this cohort of patients increased from 20 points to 61 points, and there were no signs of implant loosening.
In summary, patients who have rheumatoid arthritis typically have, despite their more extensive rotator cuff disease, good functional results after shoulder resurfacing that are comparable with the outcomes of patients who have osteoarthritis. Glenoid erosion and proximal migration of the cup appear to be common sequelae that are not associated with pain or loss of function at the time of mid-term follow-up, but these findings require further study to determine whether they may affect longer-term clinical outcomes.
Partial Resurfacing
Partial resurfacing of the humeral head has recently been used to treat smaller humeral chondral lesions, but the results of this method have been examined in only one published study, to our knowledge. Scalise et al.16 reported short-term multicenter results of the use of a partial resurfacing prosthesis in sixty-two patients (sixty-two shoulders) who had a mean age of sixty years (range, twenty-five to eighty-four years) and were followed for a mean of eight months (range, three to twenty-four months). The indications for partial resurfacing included osteoarthritis that was limited to part of the humeral head (forty-five patients), osteonecrosis (eight patients), focal chondral defects (four patients), cuff tear arthropathy (four patients), and rheumatoid arthritis (one patient). At the latest short-term follow-up evaluation, 95% of the patients reported a good-to-excellent result. The mean ASES score improved from 38 points preoperatively to 75 points at the time of follow-up, and the mean Constant score improved from 55 to 78 points. There was no evidence of radiolucencies or osteolysis.
Humeral resurfacing has been associated with low complication rates in studies with follow-up times ranging from eight months to seven years1,2,5,6,13-16,19-21,23,24,27-31. To our knowledge, only one intraoperative periprosthetic fracture has been reported, and it was managed nonoperatively23. In contrast, the prevalence of periprosthetic fractures of stemmed prostheses is approximately 3% and accounts for almost 20% of all complications associated with total shoulder arthroplasty1. Loosening, especially on the glenoid side, has been the most frequently encountered complication of resurfacing. Recent changes in the prostheses, such as the addition of a hydroxyapatite coating, have been made to address this complication, but additional studies are needed to assess the long-term effects of these changes. Proximal migration of the cup and glenoid wear have been noted in patients with rheumatoid arthritis, but these findings have had no correlation with the clinical outcome6,29,31.
The revision rates of humeral resurfacing prostheses appear to be comparable with, or better than, those of stemmed prostheses2,21. In addition, conversion to a stemmed component has been reported to be an uncomplicated procedure, since bone stock and humeral length are conserved with resurfacing1. Compared with revision of a traditional stemmed component, removal of the humeral component may require less manipulation and may be associated with a smaller risk of complications. Reported indications for revision include loosening of the humeral and/or the glenoid component, glenoid erosion, infection, fracture, and improper implant size. Multiple types of revision procedures are possible, including arthroscopy, arthrodesis, and conversion to a conventional total shoulder replacement with a stemmed humeral component.
There is controversy regarding whether the surgeon should resurface the humeral side only (hemiarthroplasty) or resurface the glenoid as well (total arthroplasty). Glenoid wear is a concern with hemiarthroplasty, and numerous studies have shown that the results of stemmed shoulder replacements are improved if the glenoid is resurfaced32. A number of studies of hemiresurfacing have demonstrated no problems with glenoid wear, but they all had short follow-up times (less than five years)1,2,14-16,23,27. Two studies with longer follow-up revealed that the clinical outcomes of glenoid resurfacing were better than those of hemiresurfacing1,21. Levy and Copeland1 assessed the clinical outcomes at a mean of seven years (range, five to ten years) in a cohort of ninety-four patients (103 shoulders) treated with either hemiarthroplasty or total shoulder resurfacing. They reported that, for patients who had osteoarthritis, the age-adjusted Constant score increased from 33.8 points preoperatively to 93.7 points following total shoulder resurfacing compared with an increase from 40.0 points preoperatively to 73.5 points following hemiresurfacing. Glenoid resurfacing can be performed with several different materials: local tissue interposition (by mobilizing ligaments from the joint capsule, reflecting them across the glenoid surface, and securing them to the glenoid labrum)24,33, soft-tissue grafts (fascia lata, Achilles tendon, or knee meniscus)33,34, polyethylene, and metal-backed polyethylene. Metal-backed polyethylene components may be thicker, which may lateralize the glenoid and subsequently decrease the range of motion. These components are also associated with an increased risk of loosening. In a resurfacing procedure, the humeral offset cannot be lowered by decreasing the head and neck length, as can be done with conventional total shoulder arthroplasty. Therefore, other options for glenoid resurfacing may have better results.
The decision whether to perform a hemiarthroplasty instead of a total arthroplasty depends on various factors. Resurfacing hemiarthroplasty is indicated when the adequacy of the glenoid bone stock is questionable or when there is minimal damage to the glenoid articular cartilage. It is also indicated for arthritic shoulders with irreparable rotator cuff damage7. Concerns about the difficulty of glenoid exposure with the humeral head intact may also affect the decision, but we have been able to achieve good glenoid exposure by using the technique described above. If a hemiresurfacing is performed, soft-tissue balancing is important to reduce glenoid wear. Studies with longer follow-up are needed to determine whether glenoid wear is an important factor in the failure or clinical results of hemiresurfacing.
Shoulder resurfacing may become more popular as the advantages become clearer and materials improve. Newer surgical approaches may become less invasive. One of us (P.M.B.) has utilized the posterior deltoid-splitting approach that was described by Wirth et al.35 for the treatment of shoulder instability. This approach preserves the deltoid origin and may decrease the risk of injury of the axillary nerve, the cephalic vein, and other neurovascular tissues. A deltopectoral approach with use of a 6-cm incision has also been described36. Both of these approaches can be considered to be minimally invasive because of decreased tissue disruption. Also, arthroscopic visualization might be utilized to assist with the procedure, especially for partial resurfacing. In addition, one of us (P.M.B.) has described the concept of using transosseous instrumentation through the humeral head to ream the glenoid surface as well as the humeral surface37.
Additional methods of operative treatment may improve the outcomes of resurfacing. For example, the step-cut lengthening of the subscapularis muscle has been associated with tearing, weakening, and detachment38. One potential technique for avoiding overtightening involves peeling the capsule and subscapularis together, in one layer, from the biceps attachment medially and attaching them to the osteotomy site. This procedure may lengthen the subscapularis muscle by 1 cm and increase external rotation by 20% to 30%39.
A potential problem with shoulder resurfacing may be difficulty with resurfacing of the glenoid without lateralization of the humeral head. This might cause increased tension on the rotator cuff and may decrease the range of motion. Thinner glenoid component designs or biologic materials may solve this potential problem.
Shoulder resurfacing is a viable alternative to conventional shoulder arthroplasty, as shown by multiple short and mid-term follow-up studies. Potential advantages of humeral resurfacing are decreased bone resection, shorter operative times, a lower prevalence of humeral periprosthetic fractures, and the potential for straightforward revision to a conventional total shoulder replacement. In addition, it may be simple to restore normal offset, inclination, and version of the glenohumeral joint because no osteotomy is performed and the head-neck angle remains intact. A potential disadvantage of resurfacing is the limited ability to correct for lateralization of the glenoid. However, new surgical approaches and materials might lessen these risks and improve results.
Partial humeral resurfacing is a procedure that involves even less bone resection. The indications for partial resurfacing include osteoarthritis, osteonecrosis, and focal chondral defects. We believe that only lesions that are limited in size and involve no substantial bone loss (for example, osteonecrotic lesions that involve less than 10% to 20% of the humeral head) can be treated with partial resurfacing. In addition, the glenoid must be undamaged because the bone-metal junction of a partial resurfacing implant could potentially rub against any incongruity of the glenoid, causing pain and further destruction.
In conclusion, the success rates of shoulder surface replacement arthroplasty are comparable with, or superior to, those associated with stemmed prostheses at the time of short and mid-term follow-up. Modern designs that include a hydroxyapatite coating may help to reduce the prevalence of radiolucent lines and prosthetic loosening, but additional studies to further assess the rate of this complication at the time of long-term follow-up are needed. In addition, future studies are necessary to evaluate alternative surface bearing materials, especially on the glenoid side, and to determine the long-term success rates.
A table listing many of the clinical series of patients treated with humeral resurfacing arthroplasty is available with the electronic versions of this article, on our web site at (go to the article citation and click on "Supplementary Material") and on our quarterly CD/DVD (call our subscription department, at 781-449-9780, to order the CD or DVD). 
Note: The authors thank Joy Marlowe, MA, CMI, for her illustrations and graphic design expertise.
Levy O, Copeland SA. Cementless surface replacement arthroplasty of the shoulder. 5- to 10-year results with the Copeland Mark-2 prosthesis. J Bone Joint Surg Br.2001;83:213-21.83213Â
2001Â
[PubMed][CrossRef] Â
Levy O, Funk L, Sforza G, Copeland SA. Copeland surface replacement arthroplasty of the shoulder in rheumatoid arthritis. J Bone Joint Surg Am.2004;86:512-8.86512Â
2004Â
Â
Jensen KL. Humeral resurfacing arthroplasty: rationale, indications, technique, and results. Am J Orthop.2007;36(12 Suppl 1):4-8.364Â
2007Â
Â
Mont MA, Ragland PS, Etienne G, Seyler TM, Schmalzried TP. Hip resurfacing arthroplasty. J Am Acad Orthop Surg.2006;14:454-63.14454Â
2006Â
Â
Steffee AD, Moore RW. Hemi-resurfacing arthroplasty of the shoulder. Contemp Orthop.1984;9:51-9.951Â
1984Â
Â
Rydholm U, Sjogren J. Surface replacement of the humeral head in the rheumatoid shoulder. J Shoulder Elbow Surg.1993;2:286-95.2286Â
1993Â
[CrossRef] Â
Copeland S. The continuing development of shoulder replacement: "reaching the surface". J Bone Joint Surg Am.2006;88:900-5.88900Â
2006Â
[CrossRef] Â
Pearl ML. Proximal humeral anatomy in shoulder arthroplasty: implications for prosthetic design and surgical technique. J Shoulder Elbow Surg.2005;14(1 Suppl S):99S-104S.1499SÂ
2005Â
[CrossRef] Â
Pearl ML, Volk AG. Coronal plane geometry of the proximal humerus relevant to prosthetic arthroplasty. J Shoulder Elbow Surg.1996;5:320-6.5320Â
1996Â
[CrossRef] Â
Williams GR Jr, Wong KL, Pepe MD, Tan V, Silverberg D, Ramsey ML, Karduna A, Iannotti JP. The effect of articular malposition after total shoulder arthroplasty on glenohumeral translations, range of motion, and subacromial impingement. J Shoulder Elbow Surg.2001;10:399-409.10399Â
2001Â
[CrossRef] Â
Harryman DT, Sidles JA, Harris SL, Lippitt SB, Matsen FA 3rd. The effect of articular conformity and the size of the humeral head component on laxity and motion after glenohumeral arthroplasty. A study in cadavera. J Bone Joint Surg Am.1995;77:555-63.77555Â
1995Â
Â
Jobe CM, Iannotti JP. Limits imposed on glenohumeral motion by joint geometry. J Shoulder Elbow Surg.1995;4:281-5.4281Â
1995Â
[CrossRef] Â
Thomas SR, Sforza G, Levy O, Copeland SA. Geometrical analysis of Copeland surface replacement shoulder arthroplasty in relation to normal anatomy. J Shoulder Elbow Surg.2005;14:186-92.14186Â
2005Â
[CrossRef] Â
Bailie DS, Llinas PJ, Ellenbecker TS. Cementless humeral resurfacing arthroplasty in active patients less than fifty-five years of age. J Bone Joint Surg Am.2008;90:110-7.90110Â
2008Â
[CrossRef] Â
Buchner M, Eschbach N, Loew M. Comparison of the short-term functional results after surface replacement and total shoulder arthroplasty for osteoarthritis of the shoulder: a matched-pair analysis. Arch Orthop Trauma Surg.2008;128:347-54.128347Â
2008Â
[CrossRef] Â
Scalise JJ, Miniaci A, Iannotti JP. Resurfacing arthroplasty of the humerus: indications, surgical techniques and clinical results. Tech Shoulder Elbow Surg.2007;8:152-60.8152Â
2007Â
[CrossRef] Â
Hill HA, Sachs MD. The grooved defect of the humeral head. A frequently unrecognized complication of dislocations of the shoulder joint. Radiology.1940;35:690-700.35690Â
1940Â
Â
Mackenzie DB. The antero-superior exposure for total shoulder replacement. Orthop Traumatol.1993;2:71-7.271Â
1993Â
[CrossRef] Â
Ellenbecker TS, Bailie DS, Lamprecht D. Humeral resurfacing hemiarthroplasty with meniscal allograft in a young patient with glenohumeral osteoarthritis. J Orthop Sports Phys Ther.2008;38:277-86.38277Â
2008Â
Â
Raiss P, Aldinger PR, Kasten P, Rickert M, Loew M. Humeral head resurfacing for fixed anterior glenohumeral dislocation. Int Orthop. 2007 Dec 19 [Epub ahead of print].Â
2007Â
Â
Levy O, Copeland SA. Cementless surface replacement arthroplasty (Copeland CSRA) for osteoarthritis of the shoulder. J Shoulder Elbow Surg.2004;13:266-71.13266Â
2004Â
[CrossRef] Â
Constant CR, Murley AH. A clinical method of functional assessment of the shoulder. Clin Orthop Relat Res.1987;214:160-4.214160Â
1987Â
Â
Thomas SR, Wilson AJ, Chambler A, Harding I, Thomas M. Outcome of Copeland surface replacement shoulder arthroplasty. J Shoulder Elbow Surg.2005;14:485-91.14485Â
2005Â
[CrossRef] Â
Crowther MA, Bell SN. Neuropathic shoulder in syringomyelia treated with resurfacing arthroplasty of humeral head and soft-tissue lining of glenoid: a case report. J Shoulder Elbow Surg.2007;16:e38-40.16e38Â
2007Â
Â
Richards RR, An KN, Bigliani LU, Friedman RJ, Gartsman GM, Gristina AG, Iannotti JP, Mow VC, Siodles JA, Zuckerman JD. A standardized method for the assessment of shoulder function. J Shoulder Elbow Surg.1994;3:347-52.3347Â
1994Â
[CrossRef] Â
Williams GN, Gangel TJ, Arciero RA, Uhorchak JM, Taylor DC. Comparison of the Single Assessment Numeric Evaluation method and two shoulder rating scales. Outcomes measures after shoulder surgery. Am J Sports Med.1999;27:214-21.27214Â
1999Â
Â
Mullett H, Levy O, Raj D, Even T, Abraham R, Copeland SA. Copeland surface replacement of the shoulder: results of an hydroxyapatite-coated cementless implant in patients over 80 years of age. J Bone Joint Surg Br.2007;89:1466-9.891466Â
2007Â
[CrossRef] Â
Jonsson E, Egund N, Kelly I, Rydholm U, Lidgren L. Cup arthroplasty of the rheumatoid shoulder. Acta Orthop Scand.1986;57:542-6.57542Â
1986Â
[CrossRef] Â
Alund M, Hoe-Hansen C, Tillander B, Heden BA, Norlin R. Outcome after cup hemiarthroplasty in the rheumatoid shoulder: a retrospective evaluation of 39 patients followed for 2-6 years. Acta Orthop Scand.2000;71:180-4.71180Â
2000Â
[CrossRef] Â
Fink B, Singer J, Lamla U, Ruther W. Surface replacement of the humeral head in rheumatoid arthritis. Arch Orthop Trauma Surg.2004;124:366-73.124366Â
2004Â
Â
Fuerst M, Fink B, Ruther W. The DUROM cup humeral surface replacement in patients with rheumatoid arthritis. J Bone Joint Surg Am.2007;89:1756-62.891756Â
2007Â
[CrossRef] Â
Radnay CS, Setter KJ, Chambers L, Levine WN, Bigliani LU, Ahmad CS. Total shoulder replacement compared with humeral head replacement for the treatment of primary glenohumeral osteoarthritis: a systematic review. J Shoulder Elbow Surg.2007;16:396-402.16396Â
2007Â
[CrossRef] Â
Burkhead WZ Jr, Krishnan SG, Lin KC. Biologic resurfacing of the arthritic glenohumeral joint: historical review and current applications. J Shoulder Elbow Surg.2007;16(5 Suppl):S248-53.16S248Â
2007Â
[CrossRef] Â
Pennington WT, Bartz BA. Arthroscopic glenoid resurfacing with meniscal allograft: a minimally invasive alternative for treating glenohumeral arthritis. Arthroscopy.2005;21:1517-20.211517Â
2005Â
[CrossRef] Â
Wirth MA, Butters KP, Rockwood CA Jr. The posterior deltoid-splitting approach to the shoulder. Clin Orthop Relat Res.1993;296:92-8.29692Â
1993Â
Â
Schneider JA, Hill JD, Cuomo F, McCann PD. Minimal-incision total shoulder arthroplasty: a cadaveric study. Am J Orthop.2007;36:596-9.36596Â
2007Â
Â
Bonutti PM, Hawkins RJ, Saddemi S. Arthroscopic assessment of glenoid component loosening after total shoulder arthroplasty. Arthroscopy.1993;9:272-6.9272Â
1993Â
[CrossRef] Â
Miller BS, Joseph TA, Noonan TJ, Horan MP, Hawkins RJ. Rupture of the subscapularis tendon after shoulder arthroplasty: diagnosis, treatment, and outcome. J Shoulder Elbow Surg.2005;14:492-6.14492Â
2005Â
[CrossRef] Â
Bonutti PM, Mont MA. Shoulder resurfacing technique for lengthening the subscapularis muscle. Unpublished data.Â
Â