Functional and Efficacy Evaluation of Surgical Treatment for Osteonecrosis of The Humeral Head: A Systematic Review with Meta-Analysis

Objective: to investigate the results of the efficacy of surgical treatment in patients with osteonecrosis of the humeral head (ONHH) by means of a systematic review of comparative analysis of hemiarthroplasty (HA) and total shoulder arthroplasty (TSA) with meta-analysis. Materials and methods: a retrospective study using a systematic review with METANALYSIS.8 The search was carried out using the following search terms in the PubMed, EMBASE, Scopus, Ovid and Cochrane Library databases. A text search strategy was employed using the search string osteonecrosis of the humeral head OR avascular necrosis OR head replacement or hemiarthroplasty or total shoulder arthroplasty. Results: The 05 articles selected presented patients diagnosed with osteonecrosis of the humeral head who underwent treatment by total arthroplasty or hemiarthroplasty. The degrees of elevation, external and internal rotation of the shoulder after the established procedures were analyzed, in addition to the VAS (pain analogue assessment) and ASES (American Shoulder and Elbow Surgeons) scales present in some of these studies. A total of 271 patients were included, 144 of whom underwent hemiarthroplasty and 127 total shoulder arthroplasty. The studies showed that TSA was preferentially used in older patients, with HA being preferred in cases where the glenoid cartilage remains intact or almost intact. Both HA and TSA provided significant relief in pain and limb function. Conclusion: HA (hemiarthroplasty) should preferably be indicated in young patients without glenoid injury. TSA (total arthroplasty) is indicated for grade V or other stages in conjunction with a damaged glenoid.

The humeral head is a bone structure susceptible to avascular necrosis, an anatomical region with a higher prevalence of loss of blood supply and/or increased intraosseous pressure, with a higher prevalence after the femoral head [1,2]. Osteonecrosis of the humeral head leads to incongruence of the glenohumeral joint and consequent alterations due to the degenerative process, which can include severe complications such as loss of range of movement, intense pain and can lead to subchondral fractures and/or arthritis and arthrosis [3,4,5]. Some etiologies for osteonecrosis can be cited: steroid and/or alcohol use, prolonged corticosteroid therapy, the presence of sickle cell disease, systemic lupus erythematosus and/or rheumatological pathology, post-traumatic events, exposure to radioactive factors and idiopathic causes. Faced with the diagnostic challenge, a stage-based classification was created as follows: stage 1 presents a positive finding on MRI or bone scintigraphy with a normal humeral head on plain radiographs, stage 2 marked by sclerotic alteration without cephalic collapse, stage 3 designated subchondral bone collapse without or with mild loss of cephalic sphericity, stage 4 by advanced head collapse with intact glenoid and stage 5: AGH incongruence and arthritic alteration [8,9]. Osteonecrosis may not be identified initially because there is no significant weight load on the shoulder, since there is no significant load on the joint, and because the symptoms may be mild or slight. As far as pain is concerned, it is usually pulsatile and deep, radiating from the shoulder to the elbow, although it is difficult to pinpoint the exact location of the pain.With regard to mobility, active movement is generally limited, while passive movement is maintained [10].

Therapeutic management is the most important in controlling the patient's quality of life and reducing the risk of complications, so staging is taken into account and treatment options include: non-operative management, arthroscopic debridement, core decompression, osteochondral autograft or allograft transplantation, surface replacement, hemiarthroplasty (HA) and total shoulder arthroplasty (TSA). Thus, the therapeutic indication is based on the classification, such as the example of arthroplasty which is performed on patients with substantial pain with functional disability in late stage 3 to stage 5 and who do not respond to non-operative treatment [6,7].

Conservative treatment can be effective in stages I and II or early stage III (Robert G et al., 2008). When radiological and clinical progression is evident, approaches that preserve the joint are chosen, such as core decompression and arthroscopic debridement. Then, in cases of advanced stages (III, IV or V), with progression of ON too painful subchondral fractures, irregular articular surfaces and subsequent degenerative changes, prosthetic replacement is constantly advised. Hemiarthroplasty (HA) or humeral head replacement has been indicated for stage III and IV pathology, when the glenoid cartilage is intact, and total shoulder arthroplasty (TSA) for stage V [11,12,13]. Due to its high prevalence at a young age, it is essential to scientifically evaluate the results of therapeutic management. Arthroplasty provided limited results, and most studies had few cases or only short-term follow-up. Therefore, investigating the clinical outcomes, complications and reoperations of each arthroplasty is important, as is comparing the effectiveness of other surgical techniques [7]. The aim of this study was to investigate the results of the effectiveness of surgical treatment in patients with osteonecrosis of the humeral head (ONHH) by means of a systematic review of comparative analysis of hemiarthroplasty (HA) and total shoulder arthroplasty (TSA) with metanalysis.

The systematic review protocol was registered in the International prospective register of systematic reviews (PROSPERO) under ID CRD42024508197 in order to improve the quality and suitability of the results to the proposed objectives. This study followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline and is a retrospective study carried out using a systematic review with METANALYSIS.8 The search was carried out using the following search terms in the PubMed, EMBASE, Scopus, Ovid and Cochrane Library databases. A text search strategy was employed using the search sequence osteonecrosis of the humeral head OR avascular necrosis OR head replacement or hemiarthroplasty or total shoulder arthroplasty, excluding all non-shoulder articles. As this was a meta-analysis of published studies, there was no need for approval by the ethics committee or institutional scientific review board. The reference lists of included and previously published articles were searched for more relevant studies that met the eligibility criteria. The inclusion criteria for the selection of studies were constructs whose sample included patients with osteonecrosis of the humeral head treated by surgical procedures. Studies addressing osteonecrosis of another anatomical region were excluded from the search.

We sought to analyze the applicability of the surgical procedure in the approach to humeral osteonecrosis, eligibility criteria for the procedure, associated risks and benefits, correlation between the procedure and functionality after surgery or length of hospital stay after the procedure. Three independent reviewers applied the eligibility criteria and selected the studies for inclusion in the systematic review. The researchers were blind to each other's decisions. Information on study design, study methodology, participant demographics (age, gender), baseline characteristics (degree of osteonecrosis, previous surgeries) was extracted from the studies. The Systematic Review Data Repository-Plus software tool was used. It was also decided to use the Cochrane risk of bias tool for greater reliability of the results.9 The results were synthesized qualitatively, as well as by combining data from individual studies that addressed the associated benefits and/or risks.

Initially, 224 articles were selected in the data search process, excluding those that had been published for more than 15 years, leaving 52 studies. After evaluating the title and abstract, 30 articles were eliminated, leaving 22 for full reading. Of these, 17 were excluded because they did not meet the inclusion criteria and 5 were included in the study (Figure 1).

Figure 1: Selection of studies according to the PRISMA methodology.

Source: Authorship own (2024).

The 05 articles selected presented patients diagnosed with osteonecrosis of the humeral head who had undergone treatment by total arthroplasty or hemiarthroplasty. The degrees of elevation, external and internal rotation of the shoulder after the established procedures were analyzed, in addition to the VAS (pain analogue assessment) and ASES (American Shoulder and Elbow Surgeons) scales present in some of these studies. A total of 271 patients were included, of whom 144 underwent hemiarthroplasty and 127 total shoulder arthroplasty.  Table 1 shows the articles selected and their outcomes (Table 1) [13,14,15,16,17].

Table 1: Studies analyzed and their outcomes.

VAS - Pain Analog Scale; ASES - American Shoulder and Elbow Surgeons.

Tables 2 and 3 show the range of motion of patients undergoing hemiarthroplasty and total arthroplasty, respectively, for the treatment of osteonecrosis of the humeral head (Table 2 and Table 3) [10,11,12,13,14].

Table 2: Analysis of patients’ range of motion undergoing hemiarthroplasty for the treatment of osteonecrosis of the humeral head.

Table 3: Analysis of patients’ range of motion undergoing total arthroplasty for the treatment of osteonecrosis of the humeral head.

Figure 2 shows a comparative analysis of the range of motion of limb elevation in patients undergoing total arthroplasty and humeral head hemiarthroplasty (Figure 2). [13,14,15,16,17].

Figure 2: Analysis comparative range of motion when lifting the limb of patients undergoing total arthroplasty and hemiarthroplasty of the humeral head.

The study by Schoch et al10 showed that the main causes of humeral osteonecrosis were steroid use (56 percentage) and idiopathic causes (26 percentage). The VAS scores of patients treated with HA and TSA were similar (2.3 +- 1.5 versus 2.5 +- 1.3), while 85% and 77% of these patients, respectively, were very satisfied with their post-operative results. Feeley et al11 demonstrated a better range of movement in patients undergoing hemiarthroplasty (150º+-8.0 versus 90º +- 12.6), and the complication rate was significantly higher with TSA (22 percentage) than with HA (8 percentage). With regard to the ASES score, a tool for assessing shoulder function, the data showed 70.8 +- 7 and 52.3+-6.1, after HA and TSA, respectively. The study by Orfaly et al12 showed similar results between the range of motion of the limb of patients undergoing hemiarthroplasty and total arthroplasty (129º +- 5.0 versus 123º +- 5.6). With regard to the VAS scale for pain, the TSA and HA groups obtained values of 18 and 17, respectively. While the study by Mansat et al13, evaluating the degree of limb mobility after hemiarthroplasty and total arthroplasty, showed insignificant differences between the findings, with a degree of limb elevation after surgery for corticosteroid causes (main causes) of 110°, 28.5° for external rotation and internal rotation for L3. 

Finally, Rhee et al14 demonstrated a mean VAS pain score, UCLA score (functional shoulder assessment scale) and ASES score with significant improvement from 6.3, 11.6 and 35.0 before surgery to 2.2, 28.9 and 82.6 at the final follow-up assessment. A greater angle of elevation was observed in patients who underwent hemiarthroplasty compared to those who underwent total arthroplasty (123.1° versus 96.9°), and there were no differences in the other ranges of movement.

The therapeutic indication for the treatment of osteonecrosis of the humeral head takes into account the age range and staging of the disease. The studies showed that TSA was preferentially used in older patients, with HA being preferred in cases where the glenoid cartilage remains intact or almost intact. Both HA and TSA provided significant pain relief and limb function, with HA being related to a greater degree of limb mobility. However, the reoperation rate is more common in partial arthroplasties. Furthermore, core decompression is safe and effective for treating low-grade osteonecrosis of the humeral head, while hemiarthroplasty and total shoulder arthroplasty should be considered for high-grade osteonecrosis. However, there is still controversy over the choice of arthroplasty type according to stage. The most important factor in deciding the type of arthroplasty is the incongruity of the AGJ or arthritic change of the glenoid due to long-term problems, glenoid wear in patients with HA and glenoid loosening in patients with TSA can be problematic and necessitate revision surgery. In one scientific study, the choice between HA and TSA was based on intraoperative findings; when four-fifths or more of the glenoid area was covered by cartilage, HA was performed. However, TSA was performed in patients with severe glenoid wear or humeral head subluxation in the deficient glenoid area. In general, hemiarthroplasty is indicated when glenoid wear is minimal or non-existent, while total shoulder arthroplasty is usually indicated when both the humerus and glenoid are involved [18,19].

Hemiarthroplasty is widely accepted for patients with intact rotator cuffs and no or little glenoid wear, two circumstances present in osteonecrosis. However, in young individuals with wear of this joint, there is no agreement on the ideal type of arthroplasty [20,21]. In young, active patients, HA is effective, especially in those with minimal glenoid wear. However, the results are potentially less pain relief, less mobility, lower patient satisfaction and a more constant need for revision surgery. Possible advantages include less blood loss, shorter surgical time and a technically easier surgical procedure.22, 2:. A known complication of HA is the development of erosion of the glenoid cavity, which is the main cause of unfavorable evolution in the medium and long term, culminating in progressive loss of range of motion, decreased shoulder function, pain and, consequently, increased dissatisfaction rates for the operated individual. It is also the main reason for clinical deterioration and revision to total arthroplasty [24,25,26,27,28,28,30]. With regard to TSA, it causes a risk of loosening of the glenoid component, which becomes increasingly common over the years and correlates with increased pain and decreased functionality. In addition, changes in bone stock are common after the procedure and complicate revision surgery. Therefore, the selection of the best type of prosthesis should be based on reasons that influence the survival of the prosthesis and functional results [31,32]. Both total and partial arthroplasty have shown satisfactory clinical results in short, medium and long-term follow-up, with both HA and TSA considerably improving the clinical function of affected patients after 8.2 years. Revision surgeries due to glenoid wear were low, while complications following TSA were high, with 32 percentage presenting with glenoid loosening and 26 percentage of patients undergoing revision surgery [33,34,35]. 

Hemiarthroplasty revisions due to glenoid erosion are not routine, while total shoulder arthroplasty revisions are more common and require greater technique, as well as greater care with bone stock, due to the risk of intraoperative fracture. This is why HA should preferably be indicated in young patients without glenoid injury. TSA, on the other hand, is indicated for grade V or other stages in conjunction with a damaged glenoid. But both allow for good clinical results and therefore patient satisfaction.

This study has no conflicts of interest

This study received no financial support from public, commercial or non-profit sources.