This annual update on shoulder and elbow surgery is a review of the most relevant studies from July 2009 through June 2010. It includes clinical and basic science articles from The Journal of Bone and Joint Surgery (American Volume), The Journal of Bone and Joint Surgery (British Volume), the Journal of Shoulder and Elbow Surgery, and Arthroscopy: The Journal of Arthroscopic and Related Surgery. Relevant level-I and level-II studies from other medical journals are included where appropriate.
The level of evidence is indicated at the end of each review when it is known. Particular attention should be paid to the level-I and level-II studies as they represent randomized controlled studies. Additional level-III and level-IV studies representing important topics in shoulder and elbow surgery are also included in the review.
General
Infection following shoulder surgery is a devastating complication. While Propionibacterium acnes is a part of the normal skin flora, it is a frequent cause of wound infection after shoulder surgery. Patel et al.1 obtained skin samples from the shoulder, hip, and knee and characterized the colonization of various anatomic locations with Propionibacterium acnes to explain this clinical observation. There was a significantly higher prevalence of Propionibacterium acnes at the shoulder as compared with either the knee or the hip. Anterior and posterior acromial sites had a greater prevalence of Propionibacterium acnes than did the hip and knee, whereas the axilla had a significantly greater prevalence of Propionibacterium acnes than did the knee. Males had a greater burden of Propionibacterium acnes at the acromial sites than females did, but there was no significant difference between males and females with regard to the prevalence of Propionibacterium acnes at the axilla.
As the skin is colonized with organisms known to cause surgical infections, the efficacy of surgical preparation solutions in reducing native skin flora prior to shoulder surgery is critical. Saltzman et al.2 evaluated three different shoulder surgical preparations to determine which was most effective against common skin flora about the shoulder. One hundred and fifty shoulders were randomized to undergo preparation with either ChloraPrep (2% chlorhexidine gluconate with 70% isopropyl alcohol), DuraPrep (0.7% iodophor with 70% isopropyl alcohol), or povidone-iodine scrub and paint (0.75% iodine scrub and 1.0% iodine paint). Culture specimens were obtained prior to the surgical preparation in the cases of twenty patients and prior to the surgical incision in all cases. The most common native skin flora prior to skin preparation were Staphylococcus epidermidis and Propionibacterium acnes. The overall rate of positive cultures following skin preparation was 31% in the povidone-iodine group, 19% in the DuraPrep group, and 7% in the ChloraPrep group. All agents were equally effective for reducing Propionibacterium acnes. However, ChloraPrep was most effective for reducing the presence of coagulase-negative Staphylococcus, whereas the iodine scrub and paint were the least effective. Despite the results of this study, no patient developed clinical evidence of infection in the postoperative period. (Level I)
Adhesive Capsulitis
The nonoperative treatment of adhesive capsulitis of the shoulder often includes the use of corticosteroids. Lorbach et al.3 performed a randomized prospective trial in which oral corticosteroid treatment was compared with three intra-articular corticosteroid injections for the treatment of idiopathic adhesive capsulitis of the shoulder. Significant improvements were found in both treatment groups in terms of the Constant-Murley score, the Simple Shoulder Test score, the visual analog scale (VAS) pain score, and shoulder motion at the time of the four-week follow-up. These improvements were noted after longer follow-up only in the injection group. Intra-articular injections of corticosteroids showed superior results in terms of objective shoulder scores, range of motion, and patient satisfaction compared with a short course of oral corticosteroids.
Rotator Cuff
Impingement Syndrome/Rotator Cuff Tendinitis
Rotator cuff symptoms in the absence of a full-thickness tear are often the source of considerable disability. Chou et al.4 performed a prospective, randomized, double-blind, placebo-controlled study to evaluate the effect of sodium hyaluronate injections for the treatment of rotator cuff lesions without full-thickness tears. Twenty-five patients had a weekly subacromial injection of sodium hyaluronate (25 mg/wk for five consecutive weeks). An additional twenty-six patients (the placebo group) were given 2.5 mL of normal saline solution according to the same injection protocol as was used in the sodium hyaluronate group. There was no difference between the two groups one week following injection. However, the sodium hyaluronate group had significantly better Constant scores and VAS pain scores than the placebo group did at six weeks after treatment. Following the completion of the study, patients in the placebo group were offered sodium hyaluronate injections. In the forty-one patients who subsequently underwent sodium hyaluronate injection, the Constant scores and VAS pain scores were significantly improved after a mean duration of follow-up of 33.1 months. The authors concluded that subacromial injections of sodium hyaluronate are effective for treating rotator cuff lesions without a complete tear. (Level I)
Full-Thickness Tears
The external rotation lag sign was originally described5 as a way to diagnose full-thickness tears of the supraspinatus tendon. Castoldi and coauthors6, including the author who originally described the external rotation lag sign, reassessed the sensitivity and specificity of the external rotation lag sign for the diagnosis of supraspinatus tears. For isolated full-thickness supraspinatus tears, the external rotation lag sign had a sensitivity of 56% and a specificity of 98%. When the rotator cuff tear extended to the infraspinatus and the teres minor, the sensitivity improved significantly. There was a strong correlation between the extension of the tear and the amount of the lag. The lag increased from 7° for an isolated rupture of the supraspinatus tendon to 26° when the tear extended to the teres minor. The external rotation lag sign is highly specific and acceptably sensitive for the diagnosis of full-thickness tears, even in cases of an isolated lesion of the supraspinatus tendon. (Level II)
Shoulder surgery to address rotator cuff symptoms is often accompanied by severe pain. Two studies evaluated pain management following rotator cuff surgery. In the first study, Coghlan et al.7 performed a randomized, blinded, placebo-controlled trial of patients undergoing either arthroscopic decompression or rotator cuff repair. All patients were managed with preemptive 1% ropivacaine (20 mL) and intraoperative intravenous parecoxib (40 mg). Patients were randomly assigned to the administration of 0.75% ropivacaine or placebo by means of an elastomeric pump at a rate of 5 mL/hr. No difference was detected between the groups with regard to the maximum pain in the first twelve hours or the average or maximum pain in the second twelve hours, with or without adjustment for opioid use. In addition, no difference was found between the groups with regard to the amount of oral analgesia used. The authors of this study found little evidence to support the use of ropivacaine infusion for improving analgesia outcomes following rotator cuff surgery in the setting of preemptive ropivacaine and intraoperative parecoxib. (Level I)
The second study evaluated the effect of low-dose gabapentin on postoperative pain in patients undergoing arthroscopic rotator cuff repair. Bang et al.8 performed a randomized, double-blinded, placebo-controlled study. The patients were divided into two groups according to the drug administered two hours before surgery (either 300 mg of gabapentin or placebo). The VAS pain scores at two, six, and twelve hours postoperatively were significantly lower in the gabapentin group than in the placebo group. The consumption of fentanyl over a period of twenty-four hours and the prevalence of side effects were similar in the two groups. The authors concluded that a single dose of 300 mg of gabapentin reduced the VAS score during the first twenty-four hours postoperatively in patients undergoing arthroscopic rotator cuff repair, with no significant difference in the rate of side effects when compared with placebo. (Level I)
Arthroscopic rotator cuff repairs are performed with use of suture anchors to fix the tendon to the tuberosity. Suture anchors are made from a variety of materials, and claims of superiority of one over another are not scientifically supported. Milano et al.9 performed a randomized prospective trial in which 110 patients were managed with arthroscopic repair of a full-thickness rotator cuff tear with use of either metal suture anchors (fifty-five patients) or biodegradable suture anchors (fifty-five patients). There were significant differences between the groups in terms of the mean Disabilities of the Arm, Shoulder and Hand (DASH) score, Work-DASH score, or Constant score. Only the baseline score, age, tear location, and fatty degeneration significantly and independently influenced the outcome. The authors concluded that, at the time of short-term follow-up, differences between arthroscopic repair of full-thickness rotator cuff tears with metal and biodegradable suture anchors were not significant.
Basic Science
In a rat rotator cuff tear model, Galatz et al.10 evaluated the effect of the mechanical environment on the healing rotator cuff. The rats in the experimental group had paralysis of the supraspinatus muscle with use of Botulinum toxin, whereas the control rats were injected with saline solution. Following surgery, the animals either were immobilized or were allowed free cage activity. The authors demonstrated that paralysis of the supraspinatus muscle with Botulinum toxin, thereby removing all loading effects on the repair construct, resulted in inferior scar volume and structural properties compared with controls. Alteration of the mechanical environment through complete paralysis of the repaired supraspinatus muscle may be detrimental to rotator cuff healing.
Studies have demonstrated the role of matrix metalloproteinases (MMPs) and tissue inhibitors of matrix metalloproteinases (TIMPs) in the pathophysiology of rotator cuff tears. Bedi et al.11 hypothesized that local delivery of an MMP inhibitor to the rotator cuff repair site in a rat rotator cuff repair model would improve healing at the tendon-bone interface. In the control group, the supraspinatus was repaired to its anatomic footprint. In the experimental group, recombinant α-2-macroglobulin (A2M) protein, a universal MMP inhibitor, was applied at the tendon-bone interface with an identical surgical repair. Significantly greater fibrocartilage was seen at the repair site in the treatment group as compared with the control group at two weeks. Significantly greater collagen organization was observed in the treatment group as compared with the control group at four weeks. A significant reduction in collagen degradation was observed at both two and four weeks in the experimental group. Biomechanical testing revealed no significant differences in terms of stiffness or ultimate load to failure.
Glenohumeral Instability
Multiple techniques have been described for closed reduction of anterior shoulder dislocations. Sayegh et al.12 compared two common techniques, the Kocher technique and the traction-countertraction (Hippocratic) technique, with a newly described FARES (Fast, Reliable, and Safe) technique. The FARES technique employs traction, but no countertraction, to the extended arm. The involved arm is slowly abducted by the physician until reduction. In this study, patients were randomly assigned to one of the three reduction techniques, and no sedation was given for two attempts at reduction. The authors found that the mean duration of the reduction maneuver was significantly shorter for the FARES method than for the Hippocratic or the Kocher method and that the mean visual analog pain score was significantly lower for the FARES method. No complications were noted in any group. In addition, the FARES technique was significantly more successful than either the Hippocratic or the Kocher technique (88.7%, 72.5%, and 68%, respectively). (Level I)
Loss of proprioception is known to occur in the face of glenohumeral instability. When surgery is required for the treatment of recurrent glenohumeral instability, the influence of the surgical approach on the recovery of strength and proprioception is not known. Rokito et al.13 performed a prospective analysis in which thirty patients who underwent open anterior stabilization with subscapularis takedown and repair were compared with twenty-five patients who underwent open stabilization through a subscapularis split. Strength and proprioception normalized in the subscapularis-splitting group at six months. The patients with subscapularis takedown still had abnormal strength and proprioception at six months; however, both strength and proprioception normalized by one year. (Level II)
Recurrent instability is a known complication of arthroscopic repair for the treatment of anterior dislocation of the shoulder. Porcellini et al.14 hypothesized that patients at higher risk for redislocation following surgery could be identified preoperatively on the basis of their clinical history. The authors investigated a cohort of patients with traumatic unidirectional instability in order to identify the risk factors for recurrence. Thirty-six months postoperatively, the patients with recurrent instability were compared with a matched cohort of patients who had successful surgery. A younger age at the time of first dislocation, male sex, and time from the initial injury were all significantly associated with higher rates of recurrent instability. (Level II)
Glenohumeral Arthritis
Osteonecrosis
The surgical treatment of advanced humeral head osteonecrosis traditionally has been performed with replacement or resurfacing of the entire humeral head. Uribe and Botto-van Bemden15 presented their experience with partial resurfacing of the humeral head for the treatment of advanced osteonecrosis of the humeral head. In that prospective study, the results for twelve shoulders with humeral head osteonecrosis were reported after an average duration of follow-up of thirty months. Postoperatively, all patients reported significant pain relief, with the visual analog scale for pain improving from 75 preoperatively to 16 postoperatively. Physical examination showed significant improvements in functional outcomes as well. Forward elevation improved from a mean of 94° preoperatively to 142° postoperatively. This prospective series demonstrated that partial resurfacing of the humeral head was effective for relieving pain and restoring function. (Level II)
Outcomes of Arthroplasty
The need to address increased glenoid retroversion when resurfacing the glenoid during total shoulder arthroplasty is well known. Asymmetric reaming to correct glenoid retroversion is commonly performed but is thought to decrease the glenoid bone available for component implantation. Nowak et al.16 used a computer model to evaluate the amount of glenoid retroversion that could be corrected before a peripheral pegged component would penetrate the glenoid vault. A glenoid with <12° of preoperative retroversion could always be implanted in neutral version without violation of the glenoid vault. Conversely, a glenoid with >18° of retroversion could not be implanted in neutral version because of glenoid vault violation from the pegs. Smaller-sized glenoid components allow for greater version correction and less residual retroversion when an in-line pegged component is used. (Level II)
Two studies evaluated the ability to improve the accuracy of glenoid implantation with use of computer-assisted techniques. In the first study, Nguyen et al.17 placed sixteen glenoid components in eight matched pairs of cadaveric shoulders; computer assistance was used for one set of shoulders and traditional techniques were used for the other set. The computer-assisted technique was significantly more accurate in terms of achieving the correct version as measured on the post-implantation computed tomography scan. The largest errors associated with traditional glenoid implantation were observed during drilling and reaming, with a trend to place the glenoid in excessive retroversion. In the second study, Kircher et al.18 performed a prospective randomized clinical study evaluating the improvement in intraoperative glenoid placement with use of surgical navigation. Two groups of ten patients each underwent total shoulder arthroplasty with or without intraoperative navigation for the treatment of glenohumeral osteoarthritis. Glenoid version was measured on axial computed tomography scans preoperatively and at six weeks postoperatively. Retroversion was corrected to 3.7° in the navigation group, compared with 10.9° in the control group (p = 0.021), although operative time was significantly longer when navigation was used. (Level II)
Radiolucency around cemented glenoid components as seen on radiographs immediately following surgery is well recognized. Newer designs and improved cement techniques are thought to reduce the prevalence of radiolucency. Edwards et al.19 studied the effect of glenoid design on immediate and follow-up radiolucency around pegged and keeled glenoid components that were inserted with use of modern cementing techniques. Patients were prospectively randomized to receive either a pegged or keeled glenoid component. On immediate postoperative radiographs, there was no significant difference in the rate of glenoid radiolucency between pegged (0%) and keeled (15%) glenoid components. However, after an average of twenty-six months, the rate of glenoid radiolucency was significantly higher in patients with keeled components (46%) as compared with pegged components (15%). The authors concluded that, even with modern cementing techniques, pegged glenoid components remain superior to keeled glenoid components when radiographs are used for assessment.
Proximal Humeral Fractures
The majority of proximal humeral fractures are treated nonoperatively. Parameters for the degree of displacement and angulation that can be treated nonoperatively are well established. However, the alignment of proximal humeral fractures can be difficult to interpret, particularly on initial shoulder radiographs. Poeze et al.20 reviewed the initial and one-week-postoperative anteroposterior and Y-view radiographs for patients who were managed nonoperatively. The final functional scores were strongly influenced by the angulation seen on the initial Y-view radiograph, with worse outcomes reported in association with angulation of >55° at the time of the first or second visit. In this series, angulation on the Y-view radiographs was more closely related to the functional outcome than was angulation on the anteroposterior radiographs.
Locking plate technology for the treatment of proximal humeral fractures was introduced to address the difficulties of stabilizing these injuries. Thanasas et al.21 performed a systematic review of the literature to evaluate the efficacy of and functional results associated with the use of locking plates for the stabilization of proximal humeral fractures. The authors identified twelve studies that included 791 patients. The patients in those studies continued to improve for up to one year, achieving a mean Constant score of 74.3. The reported complications included osteonecrosis (7.9%) and screw cut-out (11.6%), and the reoperation rate was 13.7%. The high rate of screw cut-out may have been secondary to the rigidity of the implant in combination with inadequate medial support in cases compromised by severe underlying osteoporotic bone. The authors reported that a definition of indications for the use of locking plates and attention to the technical aspects of applying them would help to optimize the results.
Lateral Epicondylitis
The nonoperative treatment of lateral epicondylitis of the distal part of the humerus often involves bracing. Garg et al.22 evaluated the efficacy of two common bracing methods for the treatment of acute lateral epicondylitis in a randomized, prospective trial. In their series of forty-four patients, twenty-four patients were randomized to receive a wrist extension splint and twenty patients were randomized to receive a counterforce forearm strap. There was no significant difference between the groups in terms of overall outcome as determined with the American Shoulder and Elbow Surgeons (ASES) and Mayo Elbow Performance (MEP) scores. However, within the ASES-derived score, pain relief was significantly better in the wrist extension splint group. No other variables were significantly different. As a pain-relieving modality, the wrist extension splint allows a greater degree of pain relief than does the forearm strap brace for patients with lateral epicondylitis of the elbow.
Fractures
Distal Humeral Fractures
The ability to achieve stable fixation following fractures of the distal part of the humerus in patients with osteoporosis is dictated by the regional cortical and trabecular bone architecture. Park et al.23 analyzed the cortical and trabecular bone in the distal part of the humerus to find regional variations and differences according to age. In the distal part of the humerus, the greatest bone volume was in the anterior part of the lateral condyle and the least was in the posterior part of the lateral condyle. The cortical thickness of the distal part of the humerus from thickest to thinnest was the posteromedial aspect followed by the lateral aspect, with the thinnest portion being the anterior aspect. The changes in cortical thickness noted with aging were most apparent in the posterior aspect of the distal part of the humerus. Changes with aging in the trabecular bone were most apparent in the medial aspect of the distal part of the humerus. The authors found a potential weakness of plate fixation in the posterolateral aspect of the distal part of the humerus because of relative insufficient osseous microarchitecture, which may affect the treatment of distal humeral fractures in elderly patients.
Several plating techniques have been advocated for the treatment of intra-articular fractures of the distal part of the humerus. Several biomechanical studies have compared perpendicular plate constructs with parallel plate constructs. However, clinical information comparing these plating techniques is lacking. In the study by Shin et al.24, the results for seventeen patients who had undergone perpendicular plating for the fixation of intra-articular distal humeral fractures were compared with the results for eighteen patients who had undergone parallel plating. No significant differences were found between the two groups with regard to the clinical outcomes. With the exception of two patients in the perpendicular plating group, all patients obtained bone union.
Collateral Ligament Injuries
Much attention has been paid to the static stabilizers of the elbow, but little is known about the contribution of dynamic stabilizers to elbow stability. Udall et al.25 used a cadaveric model to simulate the contraction of the flexor-pronator muscles to test elbow mechanics. The loads were serially changed to assess the effect of the flexor-pronator muscles on valgus stability. The elbow was then torqued in various degrees of flexion with the medial collateral ligament intact, attenuated, or cut. The flexor digitorum superficialis appeared to provide the most dynamic stability as the elbow alignment significantly changed under all testing conditions when the flexor digitorum superficialis was flaccid. The flexor carpi ulnaris and pronator teres also appeared to provide support, but to a lesser extent.
Arthritis
Total Elbow Arthroplasty
Anatomic implant alignment in total elbow arthroplasty is a challenging process with use of conventional techniques. The ability to define anatomic landmarks is complicated by deformity or bone loss. McDonald et al.26 investigated the use of an image-based navigation system to improve humeral component positioning. A modified commercial total elbow arthroplasty humeral component was implanted, with and without navigation, into the distal part of eleven cadaveric humeri. The authors demonstrated that the use of image-based navigation to guide humeral implant placement during total elbow arthroplasty significantly improved implant alignment compared with standard implantation techniques. The difference in alignment during humeral implantation between the image-based implantation technique and the standard implantation technique was more pronounced if distal humeral bone loss was present. Interestingly, the implant used in the study was modified to allow proper placement. A commercially available implant might not allow anatomic implantation with a navigated or conventional implantation technique.
Fellowships in shoulder and elbow surgery continue to gain in popularity. There are currently twenty-three shoulder and elbow fellowships available to interested applicants. A formalized match process has been successfully conducted for the past several years. The match process allows the fellowship applicants to interview at a number of programs without being under pressure to make a choice prior to completing the entire interview process. This past year, all thirty-eight available positions in the twenty-three fellowships matched for the 2012-2013 academic year.