Shoulder complications in sickle cell disease: Challenges, management strategies, and future directions

Abstract

Sickle cell disease (SCD) is a genetic disorder characterized by chronic hemolysis and vaso-occlusive crises (VOCs), leading to musculoskeletal complications that significantly affect quality of life. Among these, shoulder complications are a concern, with humeral head avascular necrosis (AVN) being the second most common site of involvement after the femoral head. Other shoulder pathologies, including osteomyelitis and septic arthritis, further contribute to morbidity. However, these conditions remain underdiagnosed and understudied, often due to overlapping symptoms with VOC-related bone infarctions. Imaging, particularly magnetic resonance imaging, is crucial for early diagnosis and accurate differentiation. Management strategies range from conservative pain control to surgical interventions, including core decompression for early-stage AVN and arthroplasty for advanced joint destruction. Surgical outcomes in SCD, however, remain inconsistent due to higher complication rates and a lack of standardized guidelines. Despite advancements in diagnosis and treatment, shoulder pathology in SCD remains an area of limited research. This review highlights the need for larger, long-term studies with a homogeneous etiology to support and refine current treatment strategies and improve patient outcomes.

Key Words: Core decompression; Osteomyelitis; Arthroplasty; Avascular necrosis; Shoulder; Sickle cell disease

Core Tip: This comprehensive narrative review highlights the challenges of managing shoulder complications in sickle cell disease (SCD) which discusses a critical and underexplored part of this disease. The review synthesizes current evidence on the three primary complications avascular necrosis, osteomyelitis, and septic arthritis. SCD leads to chronic hemolysis and vaso-occlousive crisis contribute to ischemic damage, with humeral head avascular necrosis often underdiagnosed due to overlapping symptoms with other conditions. Early diagnosis and differentiation rely on advanced imaging and laboratory tests. Management includes pain control, core decompression for early stages, and arthroplasty for advanced cases. Given the scarcity of research on shoulder pathology in SCD, larger, long-term studies are needed to refine treatment strategies and improve patient outcomes.

INTRODUCTION

Sickle cell disease (SCD) is a genetic blood disorder caused by a point mutation in the β-globin gene on chromosome 11 that results in the production of abnormal hemoglobin S (HbS)[1]. This mutation deforms red blood cells (RBC) into rigid and sickle shapes, resulting in vascular obstruction, chronic hemolytic anemia, and a cascade of complications, including vaso-occlusive crises (VOC), organ damage, inflammation, and endothelial dysfunction[1-3]. SCD significantly affects the musculoskeletal system, with shoulder pathologies such as avascular necrosis (AVN)/osteonecrosis (ON), osteomyelitis, and septic arthritis emerging as clinically important contributors to patient morbidity[2,4,5]. The humeral head is particularly susceptible to AVN due to blood supply obstruction caused by vaso-occlusion in SCD, often resulting in pain, restricted mobility, and joint destruction that severely impairs quality of life[3,6,7].

Sickle cell anemia (SCA), the homozygous (HbSS) form, is the most common and severe subtype of SCD, characterized by more frequent and severe clinical complications[8]. SCD is most prevalent in regions historically impacted by malaria, such as sub-Saharan Africa, the Mediterranean, the Middle East, and parts of India, where the sickle cell trait (SCT) typically have hemoglobin AS provides some protection against malaria[8,9].

Globally, SCD affects millions, with approximately 100000 individuals diagnosed in the United States[10]. Between 2000 and 2021, the number of births with SCD increased by 13.7% (95%CI: 11.1-16.5), reaching approximately 515000 cases globally, and the total number of people living with SCD rose by 41.4% (95%CI: 38.3-44.9) to 7.74 million[9]. The total mortality burden from SCD-related complications was nearly 11 times higher than cause-specific estimates, highlighting its significant global impact[9].

The management of shoulder complications in SCD typically begins with conservative measures, such as pain relief and physical therapy, while surgical options are reserved for advanced cases, with mixed outcomes depending on disease stage and severity[11].

Although musculoskeletal complications in SCD have been explored in broader orthopedic contexts, most studies have focused on the hips[12,13], with shoulder involvement receiving limited attention. This is primarily due to the complexity of shoulder manifestations, which are often underdiagnosed[14,15]. Therefore, this review aims to synthesize existing research on shoulder complications in SCD to provide a comprehensive understanding that can inform clinical practice and guide future investigations.

PATHOPHYSIOLOGY OF SCD

Genetic mutation and HbS

SCD is an autosomal recessive disorder that occurs when an individual inherits two mutated alleles in the β-globin gene, one from each parent[1,3]. The mutation involves a single nucleotide substitution at the sixth codon of the β-globin gene, where adenine (A) is replaced by thymine (T), resulting in the replacement of glutamic acid with valine and the production of abnormal HbS[1,3]. HbS has a propensity to polymerize when deoxygenated, causing RBC to deform into a sickle rigid shape. These abnormal cells obstruct blood flow, resulting in VOC and chronic hemolytic anemia, which are hallmark features of SCD[1,3]. Recurrent VOCs increase the risk of ischemic injury to the humeral head, predisposing it to subsequent AVN[6,16]. Individuals with two mutated alleles (homozygosity) develop the full clinical spectrum of SCD, characterized by episodes of pain, anemia, and multi-organ complications[1]. Those with one mutated allele (heterozygosity) have SCT and are typically asymptomatic but may experience mild symptoms under extreme conditions such as dehydration or high altitude[1].

Vaso-occlusion and ischemia

In SCD, vaso-occlusion may occur at various skeletal sites, including the humeral epiphysis, leading to AVN[16,17]. This process is driven by complex interactions between impaired blood rheology, chronic hemolysis, and inflammation[1]. Under hypoxic conditions, sickle hemoglobin polymerizes, reducing RBC deformability and increasing plasma viscosity, which further impairs capillary blood flow[1]. The humeral head is particularly susceptible to ischemia due to its relatively limited vascular supply[18,19]. Although the humeral head is perfused by both the anterior and posterior circumflex humeral arteries, this supply may be insufficient to maintain adequate perfusion under conditions of microvascular compromise, such as in SCD[18]. This anatomical vulnerability combined with repeated vaso-occlusive events, may explain why the humeral head is the second most common site of AVN in SCD after the femoral head[16-19].

Vaso-occlusion can also affect the medullary portion of the humeral metaphysis or diaphysis, which evolve into regions of reactive sclerosis and new bone formation[4,16,17]. This results from the increased adhesiveness of sickled erythrocytes, leukocytes, and platelets to the vascular endothelium, along with the activation of hemostatic mechanisms[20]. These cellular aggregates obstruct post-capillary venules, causing localized ischemia and initiating a cascade of inflammatory responses that exacerbate endothelial dysfunction and perpetuate the vaso-occlusive cycle[4,20,21]. Factors such as dehydration, infection, and acidosis can precipitate VOC by promoting further sickling of RBCs under low oxygen conditions, thereby increasing their rigidity and the propensity for microvascular occlusion[1].

Chronic hemolysis and inflammation

In SCD, chronic hemolysis plays a central role in driving inflammation and endothelial dysfunction[4]. Hemolysis releases hemoglobin into the bloodstream, where it reacts with nitric oxide to form reactive oxygen species (ROS), reducing its bioavailability and impairing vascular relaxation[1,4]. Oxidized hemoglobin generates cell-free heme, a damage-associated molecular pattern that activates toll-like receptor 4 and triggers inflammatory pathways, exacerbating tissue injury and chronic organ damage[1,22,23]. Endothelial activation results in upregulating adhesion molecules, including P-selectin, E-selectin, VCAM-1, and ICAM-1, facilitating the adhesion of leukocytes and platelets to the vascular endothelium, which is a critical event in the development of vaso-occlusion[1,24]. This interaction amplifies inflammation, with neutrophils and monocytes adhering to both endothelial cells and sickled erythrocytes[1,24]. Platelets further enhance the inflammatory response by releasing pro-inflammatory cytokines, including tumor necrosis factor-alpha, interleukin (IL)-1β, IL-6, and IL-8, which injure and activate the endothelium[1,23,24]. Leukocyte-platelet aggregates, mediated by P-selectin interactions, exacerbate microvascular obstruction. Neutrophils may also release neutrophil extracellular traps, contributing to vessel occlusion and propagating[1,24]. This interplay between hemolysis, inflammation, and endothelial dysfunction drives the chronic vascular complications and VOC in SCD patients, predisposing the humeral head to ischemia and AVN[1].

CLINICAL MANIFESTATIONS OF SCD IN SHOULDER

Previous reviews have provided an overview of the orthopedic complications associated with SCD[12,13,17,25]. In this review, our primary focus is on shoulder complications.

AVN

AVN or ON of the shoulder is a condition where the bone tissue of the humeral head deteriorates due to an insufficient blood supply[16,26]. As mentioned in the pathophysiology section, the repeated VOCs in SCD patients disrupts the vascular integrity of the humeral head, leading to ischemia, subchondral bone collapse, joint degeneration, and progressive pain[4,16-19]. Clinically, AVN presents with shoulder pain, reduced range of motion, and functional disability, which significantly impacts quality of life and daily activities[16,26].

AVN of the humeral head is the second most common site affected in patients with SCD after the femoral head[6,13,16]. Current literature suggests that approximately 5% of patients with SCD are associated with humeral head avascular necrosis (HHAVN)[16,27], while the overall risk of AVN at any site is as high as 50% by the age of 35[16]. However, the specific prevalence of AVN in shoulder might be underestimated as nearly half of the SCD patients with radiographic changes remain asymptomatic[15,16]. In a United Kingdom cohort of 138 patients with SCD reported that 28% had radiographic evidence of AVN, mostly bilateral, with 50% of those experiencing four or more VOCs per year developing AVN[6]. Of those with avascular shoulder changes, 63% reported functional impairment[6]. More recent study of 257 Saudi participants reported that the prevalence of ANV in shoulder among the study sample was 5.4%[27]. Similarly, a study in Qatar found that 36.7% (18/49) of SCD patients had AVN shoulder involvement[28]. At a specialist United Kingdom haemoglobinopathy center, 19% (11/59) of femoral head AVN cases also involved the humeral head[29], highlighting the frequent co-occurrence of AVN in the hip and shoulder joints[28-31]. According to Cruess classification[32], a variation of the Ficat and Arlet classification, AVN of the shoulder progresses through several stages[33] (Table 1).

Table 1 Cruess classification of shoulder avascular necrosis.

Stage	Radiographic features	Additional features/details
Stage 1	Radiographs appear normal	Magnetic resonance imaging detects early changes in the bone marrow signal, indicating the onset of the disease without structural alterations
Stage 2	Reparative process with sclerotic or mottled osteopenia	The sphericity of the humeral head is preserved
Stage 3	Appearance of the "crescent sign"	Subchondral radiolucent line signifies a subchondral fracture, with minor joint surface depressions due to localized subchondral collapse
Stage 4	Complete collapse of the articular surface	Destruction of the trabecular pattern and compromise of joint structural integrity
Stage 5	Articular changes in the glenoid	Joint incongruity with osteo-cartilaginous flaps detaching and becoming loose bodies within the joint

Osteomyelitis

Osteomyelitis is a serious bone infection and a frequent cause of hospitalization, commonly seen in long bones such as the femur and humerus[5,34]. In SCD, medullary bone infarction and necrosis create an environment conducive to bacterial growth, exacerbated by hyposplenism from repeated infarctions and fibrosis[34]. Clinically, osteomyelitis presents with localized bone pain, fever, elevated C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR) levels[5]. Although these features often overlap with VOC, distinguishing it from osteomyelitis is crucial for effective treatment. Five factors are associated with complex osteomyelitis, including severe SCD, lower limb involvement, bacteremia, positive magnetic resonance imaging (MRI) findings, and the need for surgical intervention[35].

The causative pathogen for osteomyelitis in SCD varies across different regions. For example, a 2006 meta-analysis found that Salmonella species were the most prevalent in the United States (70%) and Europe (64%), while Staphylococcus aureus was the leading pathogen in Saudi Arabia (62.1%) and Nigeria (38.5%)[36]. More recent studies highlight osteomyelitis in the upper limb. For example, a retrospective cohort in Oman found that 40.6% of osteomyelitis cases affected the upper limb[35]. The pathogen was identified in 9 cases, with Gram-negative bacteria (Salmonella, Klebsiella, Escherichia coli) responsible for 8, and MRSA in one case[35]. Similarly, a multicenter study in French Guiana found that 11% of infections involved the humerus, with Salmonella as the predominant pathogen[37]. In children, osteomyelitis commonly affects both the femur and humerus, with each site exhibiting a 22% prevalence[38].

Septic arthritis

Septic arthritis of the shoulder is an acute infection of the glenohumeral joint space that can lead to rapid joint destruction if not promptly treated[39]. In SCD patients, it is a rare but serious complication, typically presenting with severe pain, swelling, fever, leukocytosis (> 15000/mm³), and elevated CRP levels (> 20 mg/L)[39].

A retrospective study reported septic arthritis in 3% of SCD patients (59 out of 2000), with 95% occurring in patients with the hemoglobin SS genotype and an average age of 25[39]. Of these, only three cases involved the shoulder[39]. The study reported that the majority of septic arthritis cases (88%) were blood-borne, while only seven patients had osteomyelitis contiguous to septic arthritis[39]. Staphylococcus aureus was the most common pathogen identified in joint infections, followed by Salmonella[39]. Conversely, in the pediatric population, Salmonella was the most common pathogen isolated, followed by Staphylococcus aureus[38]. Rare pathogens have also been identified, including Clostridium difficile in an 11-year-old and Salmonella in a 28-year-old woman with shoulder septic arthritis[40,41].

Arthropathy

Arthropathy refers to pathological changes affecting the joints and surrounding bone structures[14]. In SCD patients, particularly SCA, osteoarticular complications are typically secondary to ON[14,30,42-44]. Research on shoulder arthropathy in SCD is limited, with most existing studies focusing on the hip joint[42,43].

In a cross-sectional study of 55 patients (mostly with SCA) reported that 61.8% experienced up to three attacks per year, 65.8% reported high-intensity pain, and 7% had osteoarticular changes in the humerus[14]. Chronic osteoarticular complications in SCD include ON, vertebral collapse, inflammatory arthritis, osteoarthritis, bone marrow hyperplasia, osteopenia, and osteoporosis[14]. These conditions often progress silently or after acute events, resulting in functional impairments and frequent underdiagnosis, particularly during the steady phase of the disease[14].

Muscle and soft tissue pathology

Muscle and soft tissue damage are less common in SCD than osseous complications but still require prompt recognition and treatment[17]. These complications often arise from vaso-occlusion and can lead to conditions such as myonecrosis, abscess formation (either direct or secondary to osteomyelitis), and myositis[17,25]. Ultrasound imaging has identified various abnormalities, including lymph node enlargement, hematomas, fat necrosis, muscle induration, subcutaneous tissue changes, and abscess collections[45].

A two-year prospective study of 23 SCD patients reported one case of thickened rotator cuff muscles with irregularities, another with a forearm abscess, and a third with suspected joint effusion; however, the remaining abnormalities were predominantly located in the hip and femur regions[45].

DIAGNOSTIC APPROACH

A comprehensive approach that combines clinical symptoms (previously mentioned), imaging, and laboratory results, is essential for management guidance.

Imaging studies

Imaging plays a crucial role in diagnosing musculoskeletal complications, including shoulder abnormalities in SCD, with the choice of modality depending on the clinical presentation and suspected pathology[5,46]. The following summarizes each type of imaging modality commonly used in these cases.

Plain X-rays are often the first step in assessing osteoarticular changes. In early stages of conditions like osteomyelitis or VOC, radiographs may appear normal or show subtle signs, such as soft-tissue swelling or periostitis[5]. Radiographic changes characteristic of osteomyelitis, including bone lysis, typically emerge about two weeks after infection onset, limiting the sensitivity of X-rays for early detection[5]. Consequently, plain radiography has limited sensitivity and specificity for early detection of these conditions[5,46]. However, in advanced stages, x-rays may reveal flattening, sclerosis, or subchondral collapse of affected bone areas[13].

Ultrasound is particularly effective in the acute phase for identifying extra-osseous pathologies, such as periosteal elevation in osteomyelitis[47]. When combined with elevated CRP or white blood cell levels, its diagnostic accuracy can reach up to 74%[47]. However, its specificity is limited, as subperiosteal fluid can also occur during VOC, causing diagnostic overlap[2]. Combining scintigraphy with ultrasound improves diagnostic accuracy by enabling guided aspiration[46].

Computed tomography (CT) scans are valuable for detecting abscesses and fluid collections, guiding joint aspirations, and evaluating recurrent joint effusion. However, they are less specific than magnetic resonance imaging for osteomyelitis diagnosis[39].

MRI is the gold standard tool for early detection of osteoarticular changes in SCD, revealing abnormalities before they appear on plain radiographs[46,48]. MRI is particularly useful for diagnosing HHAVN, bone infarction, muscle abnormalities such as infarctions or infections, and assessing bone marrow changes, including the transition from hematopoietic to fatty marrow[17,48]. Additionally, MRI differentiates medullary infarcts, which exhibit low T1 and high T2 signals, from osteomyelitis[17,48]. Gadolinium enhancement further aids in this differentiation, as osteomyelitis typically enhances with gadolinium, whereas infarction often does not[13,46].

Positron emission tomography-CT, particularly with 18F-fluorodeoxyglucose, is a valuable imaging tool for detecting inflammation or infection, especially in evaluating osteomyelitis during its acute phase[46]. It is particularly effective in distinguishing infarction, inflammation, and infection and is recommended when bone or leukocyte scintigraphy results are inconclusive[46].

Bone scans with 99mTc-diphosphonate are effective in differentiating osteomyelitis from bone infarction[49]. They provide early diagnostic accuracy for osteomyelitis by detecting hyperemia and isotope uptake[50]. In contrast, bone marrow scans show decreased uptake for infarction and normal or high uptake for osteomyelitis, making them more useful[50]. However, clinical correlation and microbiological studies are critical as increased isotope uptake may indicate either condition[51].

Laboratory tests

Laboratory tests are critical for diagnosing and managing osteoarticular complications in SCD[34,46]. Blood tests including white blood cell count, ESR, and CRP, help detect systemic inflammation or infections like osteomyelitis[46]. Microbiological tests such as blood cultures and bone biopsies are essential for identifying causative pathogens and guiding effective antimicrobial therapy[34,39,46].

MANAGEMENT AND TREATMENT OF SHOULDER PATHOLOGY

Pain management

SCD patients experience both acute and chronic shoulder pain due to the interaction between vaso-occlusion and hemolysis[52,53]. Effective pain management requires a multidisciplinary team approach tailored to individual patient needs[52]. The American Society of Hematology (ASH) recommends standardized protocols for acute pain management such as rapid administration of analgesia within one hour of presentation, reassessment every 30-60 minutes, intravenous hydration, and blood transfusions as indicated[54,55]. Individualized opioid dosing, based on prior responses, is emphasized, alongside nonsteroidal anti-inflammatory drugs (NSAIDs) for up to seven days if no contraindications exist[55]. For refractory pain, subanesthetic doses of ketamine or regional anesthesia may be considered[55].

Chronic pain in SCD is multifactorial, driven by neuropathic mechanisms, psychosocial factors, and coexisting comorbidities. ASH recommends that chronic pain without an identifiable cause may benefit from serotonin-norepinephrine reuptake inhibitors (SNRIs, e.g., duloxetine), tricyclic antidepressants (TCAs, e.g., amitriptyline), or gabapentinoids (e.g., pregabalin). When pain stems from specific complications, such as AVN, NSAIDs and SNRIs are recommended[55]. Chronic opioid therapy is reserved for cases unresponsive to other treatments, with careful monitoring to minimize dependency and side effects[52,54-56].

In general, shared decision-making is critical when initiating or continuing opioid use[54]. In addition, the ASH suggests using nonpharmacological therapies as adjuncts to pharmacological treatments[55]. Techniques such as massage, guided relaxation, yoga, and transcutaneous electrical nerve stimulation may help reduce pain intensity. Also, cognitive-behavioral therapy and coping skills training can improve pain coping and overall quality of life[55].

Blood transfusion

Blood transfusion plays an important role in managing SCD by decreasing the proportion of RBC containing HBS and enhancing oxygen-carrying capacity[57]. It is commonly used to prevent neurological issues and reduce perioperative risks among other issues[57]. However, due to limited evidence, routine transfusion solely for pain management in children and adults with SCD is not recommended[55]. Risks such as iron overload, which may necessitate chelation therapy, and careful planning when discontinuing transfusion therapy should also be considered to prevent the recurrence of complications.

Hydroxyureas use

Despite advancements in SCD therapy, hydroxyurea (HU) remains the cornerstone treatment due to its ability to increase fetal hemoglobin (HbF) levels and thereby alleviating symptoms and complications[58]. The proposed mechanisms include reductions in leukocyte and platelet counts, decreased endothelial adhesion molecule expression, and enhanced nitric oxide and cyclic nucleotide levels, all of which contribute to vascular dilation and HbF induction[58]. HU is recommended for children as young as nine months and adults with recurrent VOC, acute chest syndrome, or severe symptomatic anemia[58].

Evidence regarding the effect of HU on shoulder complications is limited. A single cross-sectional study of 55 HbSS patients reported that regular HU use for ≥ 1 year significantly reduced osteoarticular lesions, including humeral shoulder involvement[14]. Among regular HU users, only 8.6% reported five or more injuries compared to 30% of non-regular users (P = 0.0038), indicating a potential protective effect against radiographic lesions[14]. While these findings suggest that HU may lower the prevalence of chronic osteoarticular lesions, high-quality longitudinal studies with larger sample sizes are required to fully understand its role in reducing long-term skeletal complications.

Newer medications such as voxelotor and crizanlizumab have been introduced to reduce vaso-occlusive complications in SCD[58]. However, their clinical use remains limited due to regulatory restrictions, cost, and emerging concerns regarding long-term safety and efficacy[58]. To date, none of these medications have shown significant benefit in preventing chronic osteoarticular complications, leaving HU as the most established therapy.

Infections management

Osteomyelitis management involves both medical and, in some case, surgical interventions. Antibiotic therapy should be guided by culture results and infectious disease consultation[59]. A culture should be obtained before initiating antibiotics, except in cases of septic or hemodynamically unstable patients, where empiric therapy must be started immediately[5]. For common pathogens like Staphylococcus aureus, Salmonella spp., and other Gram-negative bacilli, third-generation cephalosporins are recommended as first-line treatments[5]. Alternatively, a combination vancomycin and ciprofloxacin may be used[60]. Tailoring treatment to the organism and clinical status is important to minimize complications.

Surgical intervention is necessary when antibiotics alone are insufficient to control infection, particularly in case of persistent infection, abscess formation, or necrotic bone requiring debridement[5,61]. The primary surgical approach for shoulder involvement is debridement, which removes necrotic and infected tissue to create a viable environment for healing[5]. During surgery, cultures are taken to confirm the pathogen responsible for the infection allowing for tailored postoperative antimicrobial therapy. In cases of large bone defects, techniques like the Masquelet method (cement spacers and delayed bone grafts) may be employed[5]. Additionally, soft-tissue coverage or flap techniques are used to protect exposed bone and facilitate healing[5].

A recent case study described a 23-year-old woman with chronic humeral osteomyelitis secondary to SCD who underwent debridement and osteotomy to remove affected bone tissue[61]. The bone defect was filled with calcium phosphate beads impregnated with antibiotics and a tissue graft was applied to enhance healing[61]. Postoperatively, the patient received oral levofloxacin for eight weeks. By four months, the injury had consolidated, and at six months, the patient reported complete symptom resolution and restored functionality[61]. At 22 months post-treatment, there was no recurrence, indicating a successful resolution of the osteomyelitis[61].

Surgical management of AVN

The surgical management of AVN in patients with SCD presents significant challenges, primarily due to limited high-quality evidence, heterogeneous patient populations with diverse AVN etiologies, and the absence of standardized treatment protocols[11]. Surgical options for HHAVN in SCD include core decompression, arthroscopic intervention, and various forms of shoulder arthroplasty [resurfacing, hemiarthroplasty (HA), total shoulder arthroplasty (TSA), and reverse shoulder arthroplasty (RSA)][11]. The choice of procedure depends on the disease stage, with joint-preserving techniques like core decompression preferred in pre-collapse stages, while arthroplasty is reserved for advanced joint damage[11]. Despite this staged approach, long-term outcome data remain limited. The following section will briefly discuss each surgical technique.

Core decompression

Core decompression is a joint-preserving surgical technique aims to reduce intraosseous pressure and restore blood flow to the humeral head[7]. It is typically used in younger patients with early stage AVN to alleviate pain, improve range of motion, and delay the need for arthroplasty[11].

Evidence from SCD-specific cohorts regarding the outcomes of core decompression has been inconsistent. For example, Harreld et al[62] studied 26 shoulders with ON of varying etiologies, including five cases associated with SCD. They reported significant clinical and functional improvements in 25 shoulders following smaller-diameter decompression, with no patients requiring arthroplasty over a mean follow-up period of 32 months_. In contrast, Kennon et al[7] studied 25 shoulders treated for AVN, including six SCD shoulders at Stage I or II and two related to chronic steroid use (CSI), all of which underwent core decompression. After two years, all SCD cases and one shoulder for CSI progressed to humeral head collapse. Other large studies such as those by Mont et al[63] and LaPorte et al[64] reported high success rates for core decompression in treating early-stage AVN, however, these studies included very few SCD patients, thereby limiting the generalizability of their findings to this group.

The variability in outcomes likely reflects pathophysiological features unique to SCD. Patients with SCD experience ongoing microvascular disease and repeated vaso-occlusive episodes that compromise revascularization of necrotic bone[4]. Bone quality is also reduced due to marrow hyperplasia, cortical thinning, and trabecular fragility, which impair the structural integrity needed for healing[4]. In addition, frequent sickle crises can disrupt postoperative recovery by causing recurrent ischemia in healing tissues[4]. Collectively these factors likely contribute to higher rates of humeral head collapse and the less predictable outcomes observed in SCD patients compared with other AVN populations.

Arthroscopic treatment

Arthroscopically assisted core decompression is a minimally invasive surgical technique, often performed under arthroscopic guidance[65]. This technique shows promising results for early-stage AVN of shoulder in SCD patients[65].

Clinical outcomes have shown mixed results. In one case study, an expandable reamer was inserted retrograde from humeral metaphysis toward the humeral head, followed by bone graft injection under arthroscopic guidance to avoid intra-articular penetration[66]. The patient reported significant pain relief and successful graft incorporation without joint collapse at eight months in one shoulder, while the contralateral side showed mild residual stiffness at four months[66]. In a larger cohort, Colegate-Stone et al[65] reviewed 45 patients with hematological-induced glenohumeral ON, including those with SCD. Among the 16 patients with stage 2 ON, 31% underwent arthroscopic core decompression, bursectomy, and subacromial decompression, resulting in significant pain relief and improved patient satisfaction. In contrast, of the four patients with stage 3 ON, half underwent arthroscopic debridement, capsular release, bursectomy, and subacromial decompression, but achieved only slight pain relief, with one eventually requiring arthroplasty_.

These findings suggest that arthroscopic interventions are more effective in early-stage ON/AVN, while advanced stages may necessitate joint replacement for better outcomes. However, the effectiveness of core decompression in SCD patients remains controversial. Given the conflicting evidence, the choice of surgical intervention for HHAVN in SCD patients should be tailored to individual patient factors and disease progression.

Arthroplasty

Arthroplasty is a surgical procedure involving replacement or resurfacing of the humeral head and, in some case, the glenoid surface[65,67]. In SCD patients, arthroplasty is primarily used for advanced AVN with osteoarthritic changes and functional impairment[65,67]. Common arthroplasty procedures involve HA, which replaces only the humeral head; TSA, which replaces both the humeral head and glenoid; and RSA, in which the ball and the socket configuration is revered[67]. The choice of procedure depends on disease stage, rotator cuff integrity, glenoid involvement, and patient-specific functional demands[68-71].

The clinical implications of these surgical decisions have been explored in a limited number of outcome studies involving SCD populations. The details of arthroplasty studies including complications are summarized in Table 2. In a matched cohort study, Marigi et al[67] studied 17 shoulders in SCD patients undergoing arthroplasty (9 HA, 7 TSA, 1 RSA) compared to 34 non-SCD shoulders over a mean follow-up of 5.9 years. Significant improvements postoperative were observed in pain, range of motion (ROM), strength, and functional scores (Table 2). However, SCD patients reported significantly higher preoperative pain and less effective postoperative pain relief compared to non-SCD patients. The overall complication rate was also higher in the SCD group with events including rotator cuff failure, glenoid loosening, humeral fractures, and postoperative hematoma. Furthermore, reoperation was required in 18% of SCD cases.

Table 2 Arthroplasty management in sickle cell disease studies summary.

Ref.	Population	Procedure (n)	Follow-up	Outcomes	Complications	Notes
Marigi et al[67]	17 shoulders (SCD) compared with 34 shoulders in a matched cohort of non-SCD patients	HA (9), TSA (7), RSA (1)	5.9 years	VAS: 9.1→3.8a; ASES: 48.6→73.5a; ROM: FE: 95° to 128°a, ER: 24° → 38°a, IR score:3.2 → 5.2a ASES 48.6 → 73.5a. significant improvement in strength (FE: 4.2 to 4.8a, ER: 4.1 → 4.7a, IR: 4.1 → 4.7a)	29% complication rate (glenoid loosening, RCT, hematoma, and fracture); 18% reoperation rate	Compared with non-SCD, SCD group has higher pre and post pain VAS 9.1 vs 7.4b, 3.8 vs 1.3b and higher complication rate 29% vs 12%
Colegate-Stone et al[65]	7 shoulders (Stage 4 HHAVN)	Glenoid-sparing HA (2), RSA (1)	NA	VAS: 9.5→4.1; satisfaction: 8.5/10	No significant complications noted in arthroplasty subgroup	Glenoid-sparing HA for younger patients with intact rotator cuffs and minimal glenoid changes, while RSA was indicated for older, lower-demand patients with rotator cuff deficiency
Kennon et al[7]	9 shoulders (7 SCD)	Resurfacing (7) for Stages II and III, TSA (2) for Stage IV	2-year	UCLA: 9.6→29b; ASES: 19.7→81.4a; Constant: 28→87a	3 resurfacing cases required revision (glenoid wear, stiffness, subscapular insufficiency)	Further studies needed to assess the long-term effectiveness of humeral head resurfacing
Lau et al[15]	8 shoulders (all SCD)	HA (7), TSA (1)	NA	Varied (excellent: 2, acceptable: 4, poor: 2)	Sickle cell crises, stiffness, glenoid wear	Excellent: 2 patients with high ASES scores with excellent pain relief, full functional recovery, and high satisfaction. Acceptable: 4 patients demonstrating improved function but little to no improvement in pain. Poor: 2 patients experienced decreased ASES scores, reduced activities of daily living (ADL) scores, and no pain relief
Ristow et al[70]	29 shoulders (8 SCD)	HA (NA), TSA (NA)	3.9 years	ASES: 27.3→84.2b; UCLA: 11.5→25b; Constant: 42.6→96.6b	Combined complication rate: 6.9%	HA for less glenoid wear while TSA for more arthritic glenoid. TSA tended toward better outcomes in functionality and pain relief
Feeley et al[71]	64 shoulders (4 SCD)	HA (2/37), TSA (2/27)	4.8 years	Improved ASES, L’Insalata, ROM, satisfaction (except one patient). No differences between HA and TSA in SCD subgroup	TSA: 22% (reoperation, loosening); HA: 5% (glenoid wear)	The SCD subgroup showed better scores compared to other etiologies, but these differences were not statistically significant, likely due to the small sample size

^aP < 0.001.

^bP < 0.05.

HHAVN: Humeral head avascular necrosis; FE: Forward elevation; ER: External rotation; IR: Internal rotation; SCD: Sickle cell disease; HA: Hemiarthroplasty; TSA: Total shoulder arthroplasty; RSA: Reverse shoulder arthroplasty; RCT: Rotator cuff tear; VAS: Visual analogue scale; ASES: American shoulder and elbow surgeon score; UCLA: University of California, Los Angeles shoulder score; ROM: Range of motion; VAS: Visual analogue scale.

In a cohort study, Colegate-Stone et al[65] examined glenoid-sparing and non-glenoid-sparing arthroplasty in 45 patients with hematological disease-related humeral head AVN, including those with SCD. Glenoid-sparing HA was preferred for younger patients with intact rotator cuffs and minimal glenoid changes, while RSA was indicated for older, lower-demand patients with rotator cuff deficiency. Among the seven patients with stage 4 HHAVN, two underwent HA and one RSA. Pain and satisfaction scores improved, and no significant complications were reported in the arthroplasty subgroup[65]. Additional insight was provided by Kennon et al[7] who examined humeral head resurfacing for early-stage disease and TSA for advanced HHAVN. Among SCD patients, seven at Stages II–III underwent resurfacing, either as a primary procedure or following failed core decompression. Two others underwent TSA (one after failed core decompression and the other for Stage IV). At the two-year follow-up, functional scores improved significantly. However, three resurfacing cases required revision surgery due to progression to AVN, glenoid wear with stiffness, and subscapularis muscle insufficiency; these were revised to HA, TSA, and RSA, respectively[7]. While resurfacing shows promising short-to intermediate-term results, long-term effectiveness requires further study, especially considering the increased likelihood of progression in SCD patients due to the underlying vascular and bone fragility issues associated with the disease.

Additionally, Lau et al[15] assessed eight SCD patients with HHAVN (seven HA, one TSA), reporting varied outcomes: Two excellent (high ASES scores, full recovery, no residual pain), four acceptable (improved function but persistent intermittent pain), and two poor (decreased ASES and ADL scores, no pain relief). Reported complications included two sickle cell crises, one case of postoperative stiffness requiring arthroscopic capsular release, and one revision surgery (HA to TSA due to glenoid wear). Lau et al[15] also compared their results to prior shoulder arthroplasty studies for AVN from other causes. They found that outcomes in SCD patients were generally less favourable. For instance, Mansat et al[68] reported pain relief in over 80% of patients with atraumatic AVN following shoulder arthroplasty. Similarly, Trail and Nuttall[69] observed a significant improvement in visual analog scale (VAS) scores (7.6 to 2.5) in patients with rheumatoid arthritis. These findings suggest that arthroplasty outcomes for SCD-related AVN may be inferior to those associated with other etiologies.

Further, Ristow et al[70] evaluated 29 cases of HHAVN, including eight with SCD over a mean follow-up of 3.9 years. All patient groups demonstrated significant functional improvements. Although the trauma group had the greatest improvements, differences between etiologies, including SCD, were not statistically significant. The authors also compared HA and TSA, observing that TSA showed a tendency toward better outcomes in functionality and pain relief, though the difference was not significant, likely due to the small sample size. TSA was used in cases with Outerbridge stage ≥ 2 glenoid involvement, while HA was reserved for less glenoid damage. The overall complication rate was 6.9%, which is lower than other studies[67,71]. These findings emphasize the effectiveness of arthroplasty in HHAVN, including SCD patients and highlight the importance of selecting implants based on glenoid condition[70].

Finally, Feeley et al[71] studied 64 HHAVN treated with either HA (n = 37) or TSA (n = 27), including four patients with SCD (two HA, two TSA) over a mean follow-up of 4.8 years. Functional scores and patient satisfaction improved across all patients, with all but one reporting fair satisfaction. The SCD subgroup showed higher scores than other etiologies, however, the difference was not significant.

In the same study, when comparing HA and TSA outcomes across all etiologies, no significant differences were found in ASES or L’Insalata scores. Within the SCD subgroup, no significant differences were observed between HA and TSA outcomes, likely due to the limited sample size. Notably, TSA was associated with a higher complication rate (22% vs 5% in HA), with six reoperations (four loose glenoid components, one loose humeral stem), while HA had two reoperations due to glenoid wear. Both HA and TSA improved function and pain, but TSA was linked to more complications, particularly in severe glenoid involvement[71].

Although the available evidence is limited by small and heterogeneous SCD cohorts, some clinical implications can be drawn. Arthroplasty improves pain, ROM, and function in SCD patients, but it is associated with higher complication rates and less effective pain relief compared to non-SCD patients. The choice between TSA and HA remains central, with TSA offering superior functional outcomes but at the cost of higher complications compared to HA. RSA is suggested for older, low-demand patients with rotator cuff arthropathy, while TSA and HA are recommended for younger patients with intact rotator cuffs. Furthermore, TSA is preferred in cases with significant glenoid wear, while HA is reserved for patients with minimal glenoid damage. These results underscore the importance of individualized treatment plans, long-term follow-up, and further research to refine surgical strategies for SCD patients.

PROMISING PREVENTIVE AND CURATIVE STRATEGY

Although preventing SCD ultimately requires ensuring that individuals with the SCT do not pass the condition to their offspring, this approach presents ethical, cultural, and practical challenges. Genetic counseling, education, and awareness should empower individuals and communities to make informed reproductive choices.

Gene therapy and hematopoietic stem cell transplantation (HSCT) have emerged as a potential curative approach for SCD[54,58]. HSCT, especially with HLA-matched sibling donors, provides survival rates exceeding 90%, particularly in younger patients. However, risks like graft-versus-host disease and transplant-related mortality limit its widespread use[54,58]. Gene therapy, using CRISPR-Cas9 and lentiviral vector-based methods, addresses the genetic defect by reactivating HbF or introducing functional beta-globin genes. Promising therapies like Exa-cel and lovotibeglogene autotemcel have reduced VOC frequency and improved quality of life. Yet, challenges such as high costs, myeloablative conditioning, and accessibility in resource-limited settings remain significant barriers[58].

CONCLUSION

This comprehensive review highlights the challenges of managing shoulder complications in patients with SCD, including AVN, osteomyelitis, and septic arthritis. Chronic hemolysis and recurrent VOC contribute to ischemic damage, with HHAVN often underdiagnosed due to overlapping clinical symptoms with other conditions. Early diagnosis relies on advanced imaging and laboratory tests to differentiate between mimicking conditions. Management strategies vary by disease stage, ranging from pain control and antibiotics to core decompression and arthroplasty for advanced joint destruction.

The current literature is limited by small, heterogeneous cohorts, retrospective designs, and short follow-up durations, which restrict the generalizability of findings specific to SCD population. Future research should prioritize larger, multicenter studies with extended follow-up, such as prospective registries specifically monitoring shoulder complications. Studies should also evaluate the diagnostic accuracy of advanced imaging modalities for detecting early-stage HHAVN and differentiating infarction from infection. Additionally, prospective studies evaluating different surgical approaches are needed. These may include, but are not limited to, randomized controlled trials comparing core decompression techniques (e.g., standard vs arthroscopic-assisted) in early-stage SCD-related HHAVN to clarify long-term outcomes, compare surgical strategies, and refine recovery protocols for this unique population.