Enhanced Outcomes and Safety of Reverse Shoulder Arthroplasty in Rotator Cuff Deficiency and Proximal Humerus Fractures: A 22-Patient Retrospective Analysis

Introduction

Since its creation by Grammont and colleagues in 1987, reverse shoulder arthroplasty (RSA) has been extensively utilized in the treatment of patients with irreversible large rotator cuff tears and cuff tear arthropathy [1]. One of the most important technical developments in shoulder reconstructive surgery in the last 30 years is thought to be the reverse shoulder arthroplasty [2]. It can successfully help patients feel less pain and have better function when used as directed. The glenoid becomes a sphere that articulates with the socket of the humerus. By moving the center of rotation further medially and distally, the contemporary reverse prostheses increase the mechanical advantage of the deltoid. The quality of life and active shoulder elevation have been improved as a result of this design [3]. A rotator cuff injury typically originates from a traumatic event. Acute tears brought on by macro-trauma typically occur in younger patients and result in a full tear. Microtrauma results in degenerative tears when there is insufficient healing and tendon degradation. Younger patients typically experience acute tears, while elderly patients typically experience degenerative tears [4]. There are 3 options for imaging the shoulder: magnetic resonance imaging (MRI) or MR arthrography, ultrasonography, and plain radiography. The patient’s age, functional needs, and the degree of acuteness versus chronicity of the tear all influence the course of treatment. Nonoperative therapy of any patient with asymptomatic tears is recommended. The best course of action for healthy patients who continue to experience symptoms is repair. Debridement/partial repair and/or reconstruction may be necessary in cases of persistent, large tears. Lastly, an arthroplasty may be appropriate for someone who has an irreversible tear causing excruciating pseudoparalysis [5]. RSA has been shown in multiple studies to provide positive outcomes for patients with irreversible large rotator cuff tears or cuff tear arthropathy in terms of pain alleviation, function, and satisfaction [6,7]. Revision surgery for failed total shoulder arthroplasty prostheses, revision surgery for failed proximal humerus hemiarthroplasty, post-traumatic glenohumeral arthritis, chronic irreducible shoulder dislocation, pseudo shoulder paralysis due to irreparable massive rotator cuff tears, and inflammatory joint conditions such as rheumatoid arthritis are among the more recent additions to the list of indications [1].

Proximal humerus fractures (PHF) make up approximately 5–6% of all fractures in adults [8]. Elderly individuals, who are over 65 years old, frequently exhibit a condition following a fall caused by low energy, when they extend their arm to reduce the impact of the fall [9]. For every patient, radiographic imaging should be obtained. Orthoganol imaging is required for recommended views, which include the axial lateral, scapular Y, and true AP views. Preoperative planning is aided by computed tomography (CT) scanning, particularly in cases where intra-articular comminution and ambiguous humeral head or larger tuberosity position exist [10]. In an acute context, initial therapy consists of immobilization and pain management. Then, managerial objectives might be classified as operative or nonoperative. In cases of minimally displaced surgical and anatomic neck fractures, sling immobilization along with gradually increasing therapy is recommended [11]. Treatment with surgery for displaced proximal humeral fractures is still debatable. Operative management consists of several different options available, primarily in the form of various plates and screws, intramedullary nails, hemiarthroplasties, and reverse arthroplasties; however, there is not much data to favor one technique over another [11,12]. RSA has become the dominant surgical option for treating displaced proximal humeral fractures in the elderly over the past 10 years, and it is currently considered standard care [13]. According to Lanzetti et al, RSA is a better treatment option for proximal humerus fractures than open reduction and plating [14]. In a comparative study conducted by Keller et al, the 3-year RSA implant survivorship for both PHF and rotator cuff tear arthropathy (RTCA) patients showed comparable results. Outcomes improved in both groups [15].

Even though reverse shoulder prostheses have shown promising results, clinical investigations are not usually published and substantial patient series are uncommon in the literature. Therefore, this single-center study aimed to evaluate the outcomes from reverse shoulder arthroplasty for rotator cuff ınsufficiency and proximal humerus fractures in 22 patients from a single center.

Material and Methods

STUDY DESIGN AND ELIGIBILITY CRITERIA:

The Local Ethics Committee of Mardin Training and Research Hospital approved this study (Reference 2024/E-68051626). Before providing written agreement, qualified patients were informed about the study both verbally and in writing. The study was planned in compliance with STROBE guidelines and carried out following the Declaration of Helsinki’s tenets. In this investigation, 22 RSAs (22 patients) that were completed at our institutions between the start of 2014 and the end of 2022 were retrospectively reviewed. The median age of the patients was 66 years (range: 58–95). The surgeon treating each patient was the same. Before surgery, radiographic imaging was done on every patient. To evaluate the condition of the rotator cuff, computed tomography (CT) or magnetic resonance imaging (MRI) was used. The results of the physical examination (rotator cuff weakness, dynamic instability, and limited range of motion) and imaging were used to form the basis for this evaluation of the rotator cuff. When any of the following conditions were present, a cuff tear was deemed chronic or irreversible: proximal humeral migration with narrowing of the acromiohumeral space to < 6 mm on the AP view in neutral rotation; severe cuff muscle fatty infiltration (Goutallier stage 3 or 4) [16]. Patients who were judged to be non-compliant with rehabilitation or who had a history of dementia or neurological disease, alcohol or other substance abuse, concurrent injuries to the ipsilateral or contralateral upper extremity, glenoid fractures or deformities, or previous injuries or illnesses of the injured or contralateral shoulder were excluded.

SURGICAL TECHNIQUE AND IMPLANT:

Every patient had a reversible shoulder prosthesis Humelock™ II Cementless (FX Solutions, Viriat, France). The hydroxyapatite-coated and grit-blasted humeral stem comes in 4 sizes (8, 10, 12, and 14 mm) to aid with bone integration. Two screws inserted into the distal segment’s screw holes provided extra stability. Interlocking screws theoretically give initial fixation that is robust enough to allow osseous integration, but they also share minimal load, preventing proximal stress shielding. A humeral cup (insert) for reversed arthroplasty or a cobalt-chrome (CoCr) prosthetic head for hemiarthroplasty can be attached to the proximal portion of the humeral component. When different intraoperative discoveries emerge, the surgeon can proceed with alternative surgical procedures utilizing a single set of implants thanks to its modular architecture. There are 2 diameters of glenospheres: 36 and 40 mm. For added stability, a glenosphere can be screwed or tapered onto the metaglene. The metaglene itself has a coating of hydroxyapatite underneath its surface and 4 pores for attachment to the glenoid [17].

Interscalene brachial plexus block anesthesia was used while operating on each patient while they were in the beach chair position. For every patient, the deltopectoral surgical technique was employed. There was no acromioplasty done. The methods followed both the manufacturer’s product specifications and normal operating procedures for surgery. The glenosphere was positioned 5–10 mm below the glenoid inferior to avoid the scapular notch. Press-fit fixation was used to fasten the prosthesis; none of the study’s RTSA patients had a cemented construct. All patients received 3 ampoules (3×250 mg) of tranexamic acid intra-articularly to minimize bleeding after surgery. Our surgical procedure is shown step by step in Figure 1. The postoperative radiograph of a patient we operated on due to fracture is shown in Figure 2.

POSTOPERATIVE REHABILITATION:

For 4 weeks, the patients’ shoulders were immobilized in internal rotation using an arm sling and Velpeau bandage. On the first postoperative day after surgery, mild passive motion was allowed, except internal and external rotation and active-passive elbow and wrist movements. As soon as the patients could bear it, pendulum swings were started on the first postoperative day. Week 5 began with the introduction of somewhat active activities, then passive internal and exterior rotation exercises. Patients were given full range of motion (ROM) in all directions by the end of the second month. Aggressive ROM therapy and progressive resistance exercises were also initiated. Patients were discharged with care recommendations on the first postoperative day.

DATA COLLECTION AND OUTCOME MEASURES:

Data on baseline characteristics and adverse events were collected for every patient. Patients underwent follow-up at 1, 3, and 6 months, in addition to every 3 weeks. Patient-reported outcome measures (PROM) for joint function assessment included the Disabilities of Arm, Shoulder, and Hand (DASH) score, the American Shoulder and Elbow Surgeons Score (ASES), and the Constant-Murley Shoulder Score (CS). The Visual Analogue Scale (VAS) score was used to assess pain. A goniometer on top was used to measure active movement in degrees. When it was feasible to achieve a 90° abduction, internal and external rotations were assessed in abduction using forearm movement. In CS, indirect rotation was evaluated by looking at spots on the back that could be reached (internal rotation) or by reaching various points on or above the patient’s head (external rotation). Using a spring balance, the strength was evaluated in 90° flexion and, if feasible, 90° abduction. If 90° flexion and 90° abduction were not achieved, the score was zero.

RADIOGRAPHIC EVALUATION:

All patients had trauma-series radiography and a Toshiba Aquilion 16-slice CT scan of the afflicted shoulder performed before surgery. Preoperative pictures were acquired to design the procedure and use the Osteosynthesis/Orthopedic Trauma Association (AO/OTA) categorization system to classify the fracture pattern. Both anterior and posterior shoulder dislocation were noted, and preoperative radiographs and CT scans were utilized to classify injuries as three- or four-part fractures. Anteroposterior and axillary views were included in the postoperative radiographs that were taken during every visit. Radiographs obtained during the last follow-up visit were compared with the first visit to check for implant loosening, radiolucencies, scapular notching, and tubercle union. Any complications were reported.

STATISTICAL ANALYSIS:

IBM SPSS Statistics 25 (IBM Corporation, Armonk, NY, USA) was utilized for statistical analysis. Both parametric (the mean and standard deviation) and nonparametric (the median and interquartile range [25^th through 75^th percentiles]) statistics were used to summarize the quantitative data. For categorical variables, descriptive data were displayed as counts and percentages. To compare the preoperative and postoperative levels for outcome measures of treated shoulders, the t test for paired data was applied. Implant survival was evaluated using Kaplan-Meier (95% confidence interval) analysis. Revision surgery or implant failure for any reason was considered as failure. P values below 0.05 were regarded as significant.

Results

PARTICIPANTS’ BASELINE DATA:

The study included 22 patients with a median age of 66 years (range: 58–95). Of the patients, 36.4% were male (n=8) and 63.6% were female (n=14). The indications for surgery were primarily fractures (72.7%, n=16) and rotator cuff tears (27.3%, n=6). The average American Society of Anaesthesiologists (ASA) class was 2.18±0.5. According to the Arbeitsgemeinschaft für Osteosynthesefragen (AO) classification, the distribution of fractures was as follows: C1 (43.75%, n=7), C2 (31.25%, n=5), C3 (12.5%, n=2), and B3 (12.5%, n=2). Based on the Neer classification, 31.25% of patients had 3-part fractures (n=5) and 68.75% had 4-part fractures (n=11). The injured arm was the right arm in 63.6% of cases (n=14) and the left arm in 36.4% of cases (n=8). The dominant arm was the right arm for the majority of patients (95.5%, n=21), with only 1 patient having a dominant left arm (4.5%). The average time from injury to surgery for patients with fractures was 2 ±0.6 days. The average surgical time was 80.45±16.3 minutes, and the mean follow-up duration was 4.05±1.2 years (Table 1).

FUNCTIONAL OUTCOMES QUESTIONNAIRE SCORES:

The patient-reported outcomes of those who underwent surgery due to fractures or rotator cuff tears were evaluated separately. The preoperative and latest follow-up outcomes for patients who underwent surgery for rotator cuff tears were evaluated, showing significant improvements across all measured parameters (Table 2).

The mean forward flexion increased significantly from 61.3±4.3 degrees preoperatively to 137.1±5.7 degrees at the latest follow-up. This represents a mean difference of 75.8 degrees (95% CI, 69–82.6; P<0.001).

There was a significant improvement in mean abduction, from 57.1±4 degrees preoperatively to 113.6±7.5 degrees at the latest follow-up, with a mean difference of 56.5 degrees (95% CI, 46.4–66.5; P<0.001).

The mean external rotation increased from 18.1±2 degrees preoperatively to 36.8±5.4 degrees at the latest follow-up, showing a mean difference of 18.7 degrees (95% CI, 12.1–25.1; P<0.001).

ASES score showed a significant improvement, increasing from a mean of 35.8±2.8 preoperatively to 81.3±5.4 at the latest follow-up. The mean difference was 45.5 points (95% CI, 38.5–52.4; P<0.001).

The VAS score for pain decreased significantly from a mean of 7.3±1 preoperatively to 2.9±0.9 at the latest follow-up, with a mean difference of 4.5 points (95% CI, 3–5.9; P<0.001).

DASH score improved significantly from a mean of 66.3±4.3 preoperatively to 32.5±3.6 at the latest follow-up. This corresponds to a mean difference of 33.8 points (95% CI, 26.7–40.9; P<0.001).

The Constant-Murley score increased from a mean of 48.3±3.5 preoperatively to 74.6±7.7 at the latest follow-up, with a mean difference of 26.3 points (95% CI, 17.5–35.1; P<0.001).

The joint range of motion and patient-related outcome measures at the final evaluations of patients who underwent surgery for proximal humerus fractures are shown in Table 3.

These outcomes indicate substantial improvements in the range of motion, functional scores, and pain levels in patients following surgical treatment for proximal humerus fractures. In Figure 3, it can be observed that the range of motion deficits in a patient who presented with a fracture are satisfactory during follow-up appointments.

COMPLICATIONS:

No intraoperative complications occurred. An infection developed requiring the performance of a resection arthroplasty. No signs of scapular notching or humeral loosening were observed in any of the patients at the time of the last radiographic follow-up. The Kaplan-Meier survival analysis revealed that the mean implant survival was 6.7 years (95% CI, 6.3–7.2) (Figure 4).

Discussion

Initially designed for irreparable rotator cuff tears, the indications for the reverse shoulder prostheses have expanded in recent years to include proximal humerus fractures and other shoulder pathologies. We do not frequently encounter clinical studies related to reverse shoulder prostheses in the literature. In this regard, we believe that such studies are valuable. In the present study, 22 patients who underwent reverse shoulder prostheses due to rotator cuff tears and 3- or 4-part fractures of the proximal humerus were examined. In the final evaluations, the joint range of motion and patient-related outcome measures were found to be satisfactory for the patients. We observed very few complications in our study. We believe this is due to appropriate patient selection, advancements in implant technology, and adherence to the surgical procedure and implantation protocols. All patients received interscalene block anesthesia, which reduced the need for postoperative narcotic analgesics. All patients were administered 3×250 mg transamine intra-articularly and inside the wound while closing the incision. We think that this practice reduces the frequency of dressing changes and the risk of hematoma formation in the postoperative period. Patients were discharged 24 hours after the procedure. All these perioperative factors facilitate the application and follow-up of reverse shoulder prostheses for both the patient and the surgeon.

Treatment for complicated proximal humerus fractures is complicated. Patients over 65 who have osteoporosis and other comorbidities that limit their ability to mend their bones are usually the ones affected by these injuries [18]. With revision rates of 10% to 20% at 3 to 10 years of follow-up, ORIF with plate fixation has historically been the therapy of choice [19]; however, a systematic review [2] indicated that earlier reports had found rates of implant-related problems ranging from 9% to 26%. The optimal course of treatment for proximal humeral fractures is a topic of discussion in several recently concluded and ongoing trials. After 1 year, 2 randomized controlled trials of 4-part humeral fractures treated with hemiarthroplasty or without surgery were unable to demonstrate differences in the treatment groups [20,21]. Reverse shoulder arthroplasty (RSA) gained popularity as a treatment for complex proximal humerus fractures when an early case series showed that older patients having primary RSA had satisfactory pain alleviation, range of motion, and functional results [22,23].

A prospective study published in 2023 by Lanzetti et al reported that reverse shoulder arthroplasty yielded better patient-reported outcome scores and joint range of motion compared to open reduction and plating in patients with 3-part and 4-part proximal humerus fractures. They also noted that the risk of revision was similar [14]. In their study, the RSA group had a mean Constant score of 85±7, an ASES score of 46±4, and a DASH score of 41±4. In the ORIF group, the Constant score was 53±5, the ASES score was 30±4, and the DASH score was 31±3. Compared to the patients we operated on for fractures, our Constant score was found to be lower than their study, but our ASES and DASH scores were found to be superior. This difference may have been influenced by preoperative patient performance or postoperative rehabilitation. When looking at their ORIF group, the results of our study were found to be more advantageous in every aspect. In a recent randomized controlled trial, 65–85-year-old patients with AO Type B2 and C2 fractures were compared between RSA and open reduction and internal fixation (ORIF) [24]. Patients with RSA had a higher Constant score at a 2-year follow-up. On the other hand, there was no difference in Constant and ASES ratings across groups in Klug et al’s retrospective matched-pair research of ORIF and RSA in patients over 65 years of age, but a better DASH score that was comparable to the smallest clinically significant difference [25]. However, in Klug et al’s experiment, there were more humeral head split fractures in the RSA group. These fracture patterns are known to result in worse outcomes than those that do not impact the articular surface [26]. In a prospective trial involving 62 patients with 3- and 4-part proximal humeral fractures, Sebastiá-Forcada et al compared RSA with hemiarthroplasty. They discovered that the RSA group had a mean Constant score of 56 points, which is significantly lower than in our study. This difference may be due to variations in the postoperative training regimen. Nevertheless, the range of motion was better with RSA, except for rotation, and the Constant score was noticeably higher than with hemiarthroplasty [12].

Rotator cuff tear arthropathy affects around 2% of individuals in the elderly population (70 years and older). This condition can cause intense discomfort, resulting in a partially paralyzed arm and challenges in carrying out routine tasks [27]. It can be difficult to treat patients who have severe, irreversible rotator cuff injuries. Our current treatment options for this group of patients who have failed nonoperative management include: (1) arthroscopic debridement with biceps tenotomy, (2) partial or complete arthroscopic rotator cuff repair, and (3) reverse shoulder arthroplasty, although there are many surgical options available. Few studies have been published in the literature assessing the outcomes of reverse shoulder arthroplasty in this patient population, and it is unclear which patients might benefit more from one treatment than another [16,28].

Mulieri et al treated 69 patients with irreparable rotator cuff insufficiency using RSA [6]. They reported significant improvements in ASES and VAS scores and noted significant increases in flexion (from 54° to 136°), abduction (from 51° to 120°), external rotation (from 20° to 47°), and internal rotation. In their study, the overall implant survival rate following reverse shoulder arthroplasty for all patients was 90.7% after a mean follow-up duration of 52 months. These results are similar to the outcomes of our patients who underwent RSA for cuff tears. Mulieri et al also reported improvements in their patients’ Short-Form 36 (SF-36) Physical Component Summary and SF-36 Mental Component Summary scores. In our study, when all patients who underwent surgery for cuff tears were asked, “Would you want the same surgical procedure to be performed on you again?”, all responded positively. In another study comprising mostly patients with rotator cuff insufficiency and massive rotator cuff disorder with osteoarthritis, Muh et al. reported improvements in postoperative ASES and VAS scores, as well as increases in joint range of motion (ROM) in patients who underwent RSA [7]. Similarly to this, Sirveaux et al [29] found that at a mean follow-up of 44 months, 96% of 80 patients (mean age, 73 years) had little to no pain, with a mean Constant score of 65.6. Existing data indicate that reverse shoulder arthroplasty is an effective treatment for patients with extensive and irreversible rotator cuff injuries [30–32].

Previous studies of survival after reverse shoulder arthroplasty suggested that if revision or loosening of a component is the eventual goal, positive outcomes should be anticipated. Revision was the end aim for patients with rotator cuff arthropathy treated with reverse shoulder arthroplasty, and Guery et al reported an implant survival rate of 95% at 120 months [33]. When revision was taken into account as the outcome, Mulieri et al’s study yielded an overall survival rate of 90.7% [6]. In another study, the Kaplan-Meier survival rate estimates at 3 years were reported to be 90.5% in the PHF group and 85.2% in the rotator cuff tear group. The survival rate in our study was found to be higher compared to that study [15]. Variations in endpoint selection or follow-up periods could account for this outcome. On the other hand, we think that this result was more significantly impacted by our lower complication rate. We credit the use of modern implants, stringent attention to manufacturer instructions, and exact observance of surgical protocols for our low complication rate.

This study’s retrospective approach and small sample size were its main shortcomings. It might have been possible to get more objective results with a comparison group. Additionally, we lacked pre-injury functional data, which would have enhanced cohort matching and reduced confounding in our findings. Pre-injury data could not be obtained since patients were included consecutively and presented with acute fractures. The results may have been skewed since only 1 skilled shoulder surgeon at 1 institution carried out all of the treatments; less-skilled surgeons might not have produced the same results. The fact that the surgeries were performed with the same methodology and implant types restricts the applicability of our findings to alternative surgical methods. The observed complication rates might alter with an extended follow-up period. Our study’s strengths include the use of patient-reported outcome measures, the inclusion of functional testing, and the relative paucity of clinical research on reverse shoulder arthroplasty – metrics that have gained attention recently.

Conclusions

Reverse total shoulder arthroplasty is a dependable treatment approach for both proximal humerus fracture and rotator cuff tear arthropathy. Our study shows that both circumstances had a high rate of successful implant placement during the initial period of follow-up. Functional outcome scores, range of motion, and satisfaction levels improve significantly with RSA. While the occurrence of complications can be quite frequent, the rate of revising the surgery is comparatively lower than that of other treatments. Extended-duration researches are essential to ascertain whether these findings will persist over time. When dealing with older patients, it is crucial to take into account the long-term viability of implants, as they are more prone to medical comorbidities that restrict their capacity to endure several surgeries. Therefore, it is crucial to develop a meticulous therapeutic strategy that reduces the probability of undergoing further surgery. RSA is beneficial in facilitating this planning process when conducted with suitable guidance and practical expectations.