What's New in Shoulder and Elbow Surgery.

Abstract

This update is a review of the most impactful studies related to shoulder and elbow surgery from April 2018 to April 2019. Included are clinical studies from The Journal of Bone & Joint Surgery, The American Journal of Sports Medicine, and the Journal of Shoulder and Elbow Surgery. Specific emphasis has been placed on higher-quality research (Level-I and II studies). Rotator Cuff Repair Nonoperative management still plays a pivotal role in the management of chronic, symptomatic, full-thickness rotator cuff tears, especially in older patients. There has been recent concern about an increased risk for a revision surgical procedure for patients with full-thickness tears with subacromial cortisone injections who subsequently undergo rotator cuff repair1-3. Coory et al.4 randomized rotator cuff tears treated nonoperatively to ropivacaine and betamethasone suprascapular nerve blocks compared with subacromial cortisone injections and found a greater improvement in Constant scores and higher final Constant scores at 12 weeks in the suprascapular nerve block group, with the improvements exceeding the minimal clinically important difference5. Suprascapular nerve blocks may offer a method to avoid the potential negative effects of subacromial cortisone. Pain management after rotator cuff repair continues be an important focus of research. Syed et al. evaluated the effects of preoperative narcotic education on narcotic use after rotator cuff repair6. One hundred and forty patients were randomized to either viewing a video on the risks of narcotic overuse and abuse or not receiving any narcotic education. At a 3-month follow-up, the education group consumed significantly fewer narcotics and were 2.2 times more likely to discontinue using narcotics. It is thus recommended that surgeons set up an education program to reduce postoperative narcotic consumption. Kim et al. randomized patients to a single-bolus dose interscalene brachial plexus blockade or to a patient-controlled interscalene indwelling catheter after rotator cuff repair7. The authors determined that, at 2 hours postoperatively, the single-dose bolus group had a significantly lower pain level compared with the catheter group, but, at 12 and 24 hours, the pain level of the bolus group was significantly higher than that of the catheter group. At 48 hours postoperatively, there was no difference in pain between groups. A combination of an early bolus and an indwelling catheter may be the best option to have optimal early and late postoperative pain control. Optimal pain relief and functional improvement after rotator cuff repair has been linked to psychosocial factors such as preoperative expectations and overall mental health8,9. In a longitudinal cohort study, Jain et al. determined that higher fear-avoidance behavior and alcohol use of 1 to 2 times per week or more compared with alcohol use 2 to 3 times per month or less negatively impacted shoulder pain and function via Shoulder Pain and Disability Index (SPADI) scores at 18 months postoperatively10. Surgical factors, patient demographic characteristics, and tear characteristics had no effect. Ravindra et al. showed that higher levels of postoperative pain correlated with poor emotional health11. Preoperative narcotic usage was also correlated, emphasizing the need for narcotic education. Partial-thickness rotator cuff tears can present a dilemma with regard to the timing of treatment. Currently, partial tears are often treated nonoperatively initially. Kim et al. compared immediate surgical repair with delayed surgical repair in a randomized controlled trial and did not find any differences in retear rates, suggesting that a trial of nonoperative treatment remains appropriate for partial-thickness rotator cuff tears12. Various biologic augmentation strategies have been investigated with regard to improving healing after rotator cuff repair. Oh et al.13 randomized patients to weekly human growth hormone injections for 3 months or to no injections after rotator cuff repair for a large tear and found no difference in healing rates. Walsh et al. evaluated the effects of platelet-rich plasma in fibrin matrix on rotator cuff repair healing in a prospective randomized study and found no improvement14. Similarly, Malavolta et al.15 randomized patients undergoing single-row rotator cuff repair of the supraspinatus to either control or platelet-rich plasma obtained by apheresis and applied in a liquid form with the addition of thrombin, and the authors found no differences in either clinical outcomes or healing rates. Cai et al.16 randomized 112 patients undergoing rotator cuff repair of medium to large tears to either the suture bridge technique (the control group) or the suture bridge technique with a 3-dimensional collagen graft placed between the tendon and the bone (the graft group); the authors found a lower retear rate in the graft group (13.7%) compared with the control group (34%) at 2 years postoperatively. Consequently, aligned 3-dimensional collagen grafts utilized as an interposition can reliably improve rotator cuff repair healing as well promote new tendon-bone attachment and human growth hormone and platelet-rich fibrin matrix do not appear to have a positive effect. Postoperative rehabilitation has been investigated extensively with regard to its impact on function and tendon healing. There are limited long-term effects of delaying passive motion exercises after rotator cuff repair for small and medium-sized tears17,18. Tirefort et al.19 randomized patients who had tears of <3 cm in width and were undergoing rotator cuff repair to either sling use and passive mobilization for 4 weeks postoperatively or passive mobilization only without sling use; the authors found that not using a sling resulted in improved elevation and internal rotation at 3 months and equal tendon healing at 6 months. These data would also support no clinically important benefit of sling use with respect to tendon healing for tears of <3 cm in width20,21. Shoulder Arthroplasty Perioperative Management Similar to rotator cuff repair, improvement in pain control after shoulder arthroplasty has been high priority. Sicard et al.22 randomized patients after shoulder arthroplasty to either an interscalene block using a catheter for 48-hour infusion or local infiltration at the time of the surgical procedure and a glenohumeral catheter that was dosed the morning after the surgical procedure; the authors found no difference between groups in pain levels at 48 hours but found less overall pain and opioid consumption in the recovery room with the local infiltration group. Namdari et al.23 randomized patients after shoulder arthroplasty to an interscalene brachial plexus blockade with or without the addition of liposomal bupivacaine injection and found that the addition of liposomal bupivacaine increased narcotic consumption over the first 24 hours postoperatively and had no effect on pain levels during the first 3 days postoperatively. McLaughlin et al.24 prospectively studied a multimodal postoperative analgesia regimen with preoperative and postoperative nonsteroidal anti-inflammatory drugs, gabapentin, and acetaminophen and compared it with a standard regimen that included postoperative opioids and a preoperative interscalene regional block with regard to pain, opioid consumption, length of stay, and readmission rates after shoulder arthroplasty. Postoperative opioids were between 37% and 47% lower on days 0 to 2 and the length of stay was 0.5 day less in the multimodal regimen group compared with the standard regimen group. Consequently, multimodal anesthesia combined with an interscalene brachial plexus blockade with or without the addition of local infiltration should be utilized to optimize pain control after shoulder arthroplasty. Shoulder immobilization after shoulder surgery has historically been utilized to protect the surgical repair and to improve comfort. Various techniques have been used with varying results, including a simple sling, a small or large abduction wedge, and external rotation wedges. Baumgarten et al. performed a randomized, single-blinded, prospective study comparing a simple sling without a pillow (internal rotation) with a sling with an abduction-wedge pillow placing the arm into a more neutral rotation (neutral rotation sling)25. A full external rotation wedge was not utilized in this study. The authors determined that the neutral wedge sling group had greater improvements in active and passive external rotation and less night pain at 2 weeks postoperatively. The addition of a simple abduction-wedge pillow should be considered after anatomic total shoulder arthroplasty as there appear to be clear advantages with limited morbidity. Finally, refining perioperative treatments for patients to allow early discharge or outpatient arthroplasty may include limiting the usage of drains. Trofa et al. performed a prospective randomized study evaluating the use of drains postoperatively after shoulder arthroplasty26. One hundred patients were randomized to drain use or no drain use after anatomic total shoulder arthroplasty or reverse shoulder arthroplasty in the setting of the routine use of tranexamic acid. The authors determined no advantage or disadvantage of using a drain with regard to postoperative anemia, length of stay, or cost when complemented with tranexamic acid. Infection Diagnosis and Prevention The prevention of infection continues to be of paramount interest as it is one of the most common early and devastating complications in shoulder surgery, specifically arthroplasty. In a prospective randomized controlled trial, Kolakowski et al.27 compared, with skin culture specimens obtained at the time of the surgical procedure, the effect of the preoperative application of a solution of either 5% benzoyl peroxide or 4% chlorhexidine gluconate to the skin for 3 days prior to shoulder surgery. The authors found fewer positive cultures in the benzoyl peroxide-treated shoulders compared with the contralateral control shoulders, whereas there was no difference in the chlorhexidine gluconate-treated shoulders compared with the contralateral shoulders. The benzoyl peroxide group showed a significant reduction in Cutibacterium acnes counts compared with the chlorhexidine gluconate group. Using 5% benzoyl peroxide preoperatively is a reasonable strategy to reduce topical C. acnes and to potentially reduce the quantity of bacteria introduced into the wound at the time of the surgical procedure. As an alternative to topical antibiotic treatments, preoperative oral antibiotics directed at C. acnes may be a method to potentially reduce bacterial load at the time of shoulder arthroplasty. Rao et al. performed a randomized controlled trial investigating the utility of preoperative intravenous cefazolin compared with cefazolin plus intravenous doxycycline in reducing the bacterial load of C. acnes from the skin, dermis, and joint capsule during primary shoulder arthroplasty28. The authors reported that 38% of patients undergoing arthroplasty had ≥1 positive cultures for C. acnes, with no differences between treatment groups. Although preoperative doxycycline within 1 hour of the surgical procedure does not reduce positive C. acnes cultures taken at the time of the surgical procedure, doxycycline taken orally days or weeks prior to shoulder arthroplasty has not been investigated and may have a role in reducing bacterial burden. Reverse Total Shoulder Arthroplasty Neurologic injury remains a concerning postoperative complication after reverse total shoulder arthroplasty. Several authors have shown increased neurologic injury in reverse total shoulder arthroplasty using a Grammont-style implant compared with an anatomic total shoulder arthroplasty, most likely due to arm lengthening with reverse total shoulder arthroplasty29,30. Lowe et al. reported on a series of patients undergoing reverse shoulder arthroplasty using a lateralized glenosphere and a 135° neck-shaft angle, instead of a Grammont-style 155° neck-shaft angle, and compared the incidence of neurologic injury using intraoperative neuromonitoring to a series of patients undergoing anatomic total shoulder arthroplasty31. With mean changes in arm length of 3 mm for anatomic total shoulder arthroplasty and 14 mm for reverse total shoulder arthroplasty, the authors found an equal number of nerve alerts and temporary postoperative neurologic changes between implants, concluding that a lateralized design with less arm lengthening eliminates the historical increased risk of nerve injury with reverse shoulder arthroplasty relative to anatomic total shoulder arthroplasty. Reverse shoulder arthroplasty reliably improves forward elevation and external rotation in those with cuff tear arthropathy, but internal rotation remains challenging. Dedy et al. evaluated the impact of subscapularis repair integrity on shoulder function, specifically internal rotation motion, after reverse shoulder arthroplasty32. Forty-three patients were evaluated at a median of 19 months after Grammont-style reverse shoulder arthroplasty with sonography to determine the healing status of the subscapularis. Forty-six percent of patients had an intact or mildly attenuated repair, whereas 54% of patients had a severely attenuated repair or a repair that was not intact. Those with an intact or mildly attenuated repair had significantly better internal rotation motion compared with those with severe attenuation or an absent tendon. On the basis of these data, if subscapularis repair is performed and remains intact, it may improve the overall internal rotation range of motion. Anatomic Total Shoulder Arthroplasty Preoperative planning utilizing computed tomography (CT) has become commonplace prior to anatomic total shoulder arthroplasty. Most implant manufacturers have a specific tool created to allow either 3-dimensional preoperative templating alone or templating with the addition of patient-specific instrumentation using either a custom or reusable intraoperative guide. Iannotti et al. analyzed 173 patients with end-stage glenohumeral arthritis who were enrolled in 3 prospective trials evaluating 3-dimensional preoperative planning and patient-specific instrumentation with anatomic total shoulder arthroplasty33. The authors determined that there were no significant differences in the deviation from the preoperative plan for the glenoid implant orientation or location when using 3-dimensional planning and standard instruments; 3-dimensional planning and a single-use, patient-specific guide; or 3-dimensional planning and reusable patient-specific instrumentation. Consequently, patient-specific instrumentation appears to afford no advantage compared with 3-dimensional planning alone. Patient-specific instrumentation may still offer an advantage to less experienced surgeons in glenoid component placement. Glenoid component loosening and failure remain the most common reason for late failure of anatomic total shoulder arthroplasty. Advancement in glenoid component fixation techniques has included ingrowth peg designs. Ricchetti et al. performed a follow-up study evaluating patients immediately postoperatively and at a minimum of 2 years postoperatively with a 3-dimensional CT scan to evaluate peg osteolysis and component shift34. The authors noted that 30% of components shifted position at 2 years. Significantly more patients with glenoid component shift (83%) had central peg osteolysis compared with patients without glenoid component shift (7%) (p = 0.002). Ingrowth central peg glenoid components have a relatively high incidence of early shift, which is likely associated with limited ingrowth of the central peg. Besides postoperative infection, subscapularis failure after anatomic total shoulder arthroplasty remains the most common early complication that may lead to weakness, instability, and a poor result. Multiple studies have previously evaluated the 3 primary subscapularis takedown techniques, tenotomy, peel, and osteotomy, both clinically and biomechanically. Levine et al. performed a prospective randomized trial comparing 2 common techniques, tenotomy and osteotomy, for tendon healing, case duration, and repair time35. The authors reported better healing rates with osteotomy (93%) compared with tenotomy (87%) but a shorter case duration by 23 minutes and a shorter subscapularis repair time by 5 minutes for the tenotomy technique. Despite improved healing in the osteotomy group, there were no differences in range of motion, strength, or functional outcomes between groups at 1 year postoperatively. The study supported improved healing with osteotomy, as supported by other authors36, but no advantage in outcomes. Further studies with larger sample sizes are required to determine if improved healing results in improved range of motion or strength. Glenohumeral Instability Arthroscopic labral repair is a standard treatment for patients with recurrent anterior shoulder instability without severe bone loss. Various stitch configurations from suture anchors have been implemented, including simple and mattress stitches or a combination of both, with unclear advantage of one over another. Park et al. performed a prospective randomized controlled study comparing the vertical simple stitch with a modified Mason-Allen method (combination of vertical simple and horizontal mattress stitches from the same anchor), evaluating labral healing, height, and width at 6 months postoperatively and recurrence at 2 years postoperatively37. The authors reported no differences in recurrence rates, apprehension rates, or clinical outcomes at ≥2 years postoperatively. Similarly, at 6 months postoperatively, on the basis of CT arthrography, there were no differences in labral height, width, or healing. Either method is reasonable, with no advantage of one over the other when performing an arthroscopic labral repair. Anterior and posterior shoulder instability have a different injury mechanism, clinical presentation, and response to treatment. Bernhardson et al. attempted to quantify each of these characteristics in a matched cohort examination and surgical outcome analysis in patients with recurrent anterior or posterior shoulder instability38. The authors determined that the primary mechanism of injury in the anterior cohort was a formal dislocation (82.5%), whereas no identifiable injury mechanism was defined in 78% of the posterior cohort. The primary symptom was instability 80% of the time in the anterior cohort, whereas the primary symptom was pain 90.7% of the time in the posterior cohort. Postoperative clinical outcomes were all significantly better in the anterior cohort compared with the posterior cohort as measured by the American Shoulder and Elbow Surgeons (ASES) score, Single Assessment Numeric Evaluation (SANE), and Western Ontario Shoulder Index (WOSI). Controversy continues to exist with regard to the treatment of patients with a first-time anterior dislocation of the shoulder. Factors influencing the decision with regard to operative treatment compared with nonoperative treatment include age, activity or sport level, and concomitant osseous or soft-tissue injuries. Rugg et al. reported on a multicenter prospective cohort study of 172 patients undergoing surgical stabilization for anterior shoulder instability at multiple institutions39. Patients were categorized into groups based on the number of dislocations (1, 2 to 5, or >5). The authors reported that patients with 1 dislocation were more often treated arthroscopically, whereas those with recurrent dislocations were more frequently treated with a Bristow-Latarjet procedure. Recurrent dislocators were more likely to have an osseous Bankart lesion intraoperatively and increased glenoid bone loss on preoperative imaging. It is clear that patients with recurrent instability have higher risks of bone loss requiring osseous procedures for stabilization, although it is not clear from the study how many recurrences become the tipping point before these events occur. The same group evaluated early return to function and range of motion after arthroscopic anterior instability repair and determined that 74% of patients returned to baseline range of motion and 98% returned to baseline strength at a mean time of 5.3 months postoperatively40. Increased global hyperlaxity and number of preoperative dislocations were the only factors associated with a failure to return to baseline motion and strength, suggesting that early treatment (fewer dislocations) may also allow a quicker recovery to baseline. Elbow Lateral Epicondylitis A variety of nonoperative and operative treatments exist for recalcitrant elbow pain secondary to lateral epicondylitis, including physical therapy, corticosteroids, platelet-rich plasma, and surgical release. In a randomized double-blinded study, Creuzé et al. reported the efficacy of injection of botulinum toxin into the extensor carpi radialis brevis (ECRB) muscle compared with placebo, utilizing electromyography to isolate the muscle41. The overall pain levels were significantly lower, and the percentage of patients who received >50% reduction in initial pain was significantly greater in the botulinum toxin group (51.7%) compared with the control group (25%) at 90 days. Botulinum toxin is a reasonable nonoperative alternative, with relatively low risks, to surgical treatment. In a prospective, randomized, double-blinded, placebo-controlled trial, Kroslak et al. investigated the use of a counterforce brace compared with placebo bracing42. The authors determined that both braces improved pain frequency and severity at 6 months and 3 years, although the counterforce brace group had lower frequency of pain at rest at 6 and 12 weeks and better patient-related elbow function at 26 weeks compared with the placebo group. Consequently, counterforce bracing should also be included in nonoperative management, as there are clear benefits compared with placebo. If nonoperative treatment fails, surgical excision of the affected portion of the ECRB tendon has been performed with success. In a prospective, randomized, double-blinded, placebo-controlled trial, Kroslak and Murrell investigated the effectiveness of tendon excision compared with that of sham surgery for lateral epicondylitis43. The experimental group included a Nirschl mini-open technique without bone drilling where the ECRB degenerative tissue was excised without repair. The sham group had exposure of the ECRB tendon through elevation of the extensor carpi radialis longus muscle but without excision of the ECRB tendon. Both groups had similar improvements in all pain measures by postoperative week 26, including frequency of pain with activity, and maintenance over 1 to 4 years. The study clearly demonstrated that surgical treatment of lateral epicondylitis is an effective and lasting solution for chronic epicondylitis resistant to nonoperative treatment. What is unclear is the exact mechanism by which patients improve as excision of the diseased tissue does not seem to improve outcomes. Osteochondritis Dissecans of the Capitellum Optimal management of patients with osteochondritis dissecans of the capitellum is dependent on multiple factors, including patient age, elbow range of motion, and radiographic findings including the status of the growth plate closure and fragment stability. Niu et al. evaluated the success of nonoperatively treating patients with osteochondritis dissecans of the capitellum without fluid underneath the fragment44. The authors reported that 53.8% of lesions healed at a mean time of 8.3 months. Factors associated with lesion healing included no clear margins of the fragment on magnetic resonance imaging, no cyst-like lesions, and overall smaller-sized lesions; other factors including age and physeal status had no effect. The authors presented a nomogram that can be used by clinicians to predict healing, based on lesion size and the presence of cyst-like lesions, and can assist in the treatment algorithm for these patients. Ulnar Collateral Ligament Insufficiency Various factors affect return to play after ulnar collateral ligament (UCL) reconstruction in baseball, including tear type, surgical technique, graft type, and player characteristics. Two recent studies evaluated these factors in U.S. Major League Baseball (MLB) players who underwent UCL reconstruction. Marshall et al. evaluated the effects of surgical technique, graft type, and tear characteristics on return to play in MLB players undergoing UCL reconstruction45. Although surgical technique did not affect performance, pitchers with a palmaris graft played longer and returned to play at a higher percentage than pitchers with a gracilis graft. Pitchers with complete tears played longer after reconstruction than those with partial tears, and tear location did not affect how patients performed after reconstruction. On the basis of these data, technique and graft type did not affect performance; therefore, the surgeon’s preference on technique and graft type can be expected to lead to comparable results. Camp et al. evaluated the effect of position on return to play and identified that professional baseball position players returned to play sooner than pitchers but demonstrated lower rates of return to play after UCL reconstruction46. The player position with the lowest likelihood of return to play is catcher, at 58.6% compared with 83.7% for the position of pitcher. The data from these studies can be used to set expectations for players undergoing UCL reconstruction in terms of who might have the greatest success after surgical treatment using the various surgical techniques and grafts. Evidence-Based Orthopaedics The editorial staff of JBJS reviewed a large number of recently published studies related to the musculoskeletal system that received a higher Level of Evidence grade. In addition to articles cited already in this update, 4 other articles relevant to shoulder and elbow surgery are appended to this review after the standard bibliography, with a brief commentary about each article to help guide your further reading, in an evidence-based fashion, in this subspecialty area. Evidence-Based Orthopaedics Hurley ET, Fat DL, Duigenan CM, Miller JC, Mullett H, Moran CJ. Biceps tenodesis versus labral repair for superior labrum anterior-to-posterior tears: a systematic review and meta-analysis. J Shoulder Elbow Surg. 2018 Oct;27(10):1913-9. Epub 2018 May 24. Comparative studies in the literature were systematically reviewed to determine if biceps tenodesis or labral repair results in superior outcomes when treating superior labrum anterior-to-posterior (SLAP) tears. Five studies with 234 patients were included in the analysis. When compared with SLAP repair, biceps tenodesis resulted in improved rates of patient satisfaction (95.6% compared with 76.2%) and rate of return to sport (81.3% compared with 64.3%). The study supports that biceps tenodesis results in higher rates of patient satisfaction and return to sport, although both repair techniques result in similar shoulder functional outcomes; therefore, tenodesis should be considered as the surgical procedure of choice for SLAP tears in a majority of patients, including most athletes. Moatshe G, Kruckeberg BM, Chahla J, Godin JA, Cinque ME, Provencher MT, LaPrade RF. Acromioclavicular and coracoclavicular ligament reconstruction for acromioclavicular joint instability: a systematic review of clinical and radiographic outcomes. Arthroscopy. 2018 Jun;34(6):1979-1995.e8. Epub 2018 Mar 21. The authors performed a systematic review of clinical and radiographic outcomes after surgical treatment of acromioclavicular joint instability using free tendon graft, suspensory devices, synthetic ligament devices, modified Weaver-Dunn technique, and hook plate and Kirschner wire techniques. There were 939 patients included in this study and there were no differences in Constant scores between techniques at the time of the latest follow-up. The hook plate and Kirschner wire techniques had the highest complication rates between techniques and the modified Weaver-Dunn technique had the highest unplanned reoperation rates. Based upon this information, consideration should be made for use of techniques with the lowest complication and reoperation rates, specifically the use of a free tendon graft. Patterson DC, Chi D, Parsons BO, Cagle PJ Jr. Acromial spine fracture after reverse total shoulder arthroplasty: a systematic review. J Shoulder Elbow Surg. 2019 Apr;28(4):792-801. Epub 2018 Nov 26. A systematic review was performed evaluating the occurrence and outcomes of acromial or scapular spine fractures after reverse total shoulder arthroplasty. Thirty-two studies including 3,838 reverse total shoulder arthroplasties identified acromial fractures in 159 patients for an incidence of 4.14%. The mean time from the surgical procedure to diagnosis was 9 months, and, regardless of operative or nonoperative treatment, patients experienced inferior results compared with initially after reverse total shoulder arthroplasty. The mean forward flexion was 95°, and the mean ASES score was 57 points. Acromial fractures occur frequently after reverse total shoulder arthroplasty and have a negative impact on range of motion and function independent of treatment. Surgeons should be aware of the relative frequency and should inform patients with regard to the potential complication. Sheps DM, Silveira A, Beaupre L, Styles-Tripp F, Balyk R, Lalani A, Glasgow R, Bergman J, Bouliane M; Shoulder and Upper Extremity Research Group of Edmonton (SURGE). Early active motion versus sling immobilization after arthroscopic rotator cuff repair: a randomized controlled trial. Arthroscopy. 2019 Mar;35(3):749-760.e2. The authors performed a randomized study of patients who underwent arthroscopic rotator cuff repair and compared patients who self-weaned from sling use and performed pain-free active range of motion during the first 6 weeks postoperatively and patients who wore a sling for 6 weeks postoperatively and had passive motion only. All tear sizes were included. At 6 weeks postoperatively, the early active motion group had better forward flexion and abduction, but, at 1 year and 2 years, there was no difference in range of motion, pain, strength, or health-related quality of life between groups. Healing rates were equal at 12 months postoperatively with 30% retear in the early motion group and 33% retear in the 6-week immobilization group. Retear rates were not affected by immobilization technique even in large tears. The data support no added benefit or detriment to sling immobilization after arthroscopic rotator cuff repair.