1 INTRODUCTION

Haemophilia is an X-linked, inherited bleeding disorder caused by complete or partial deficiency of coagulation factor VIII (haemophilia A) or factor IX (haemophilia B).¹ Spontaneous and traumatic bleeding episodes are the hallmark of haemophilia, with large synovial joints including the elbows, knees, and ankles being the most affected.² Haemophilic arthropathy results from repeated haemarthroses and remains a significant cause of morbidity for persons with haemophilia (PwH).^3-5 Even a single episode of haemarthrosis, clinically overt or unrecognised subclinical bleeding, can start the vicious cycle of chronic synovial inflammation and thickening, neoangiogenesis, rebleeding, remodeling and eventual osteochondral surface destruction and joint space narrowing.⁶ Patients with haemophilic arthropathy suffer from joint instability, deformity, stiffness and chronic pain ultimately resulting in inactivity, osteoporosis, muscle atrophy and significantly reduced quality of life.^{7, 8}

2 DISCUSSION

The last decade has witnessed a breakthrough in the treatment of PwH with the advent of factor, non-factor, and gene therapies. Treatment goals shifted from preventing life threatening bleeding and increasing life expectancy, to protecting against joint bleeding, preserving joint health, and allowing for a more active lifestyle.⁸ The treatment landscape has never been more complex and individualising treatment regimens is not an easy task with the large number of available options; management decisions are best shared with patients and their caregivers.⁹

3 JOINT ASSESSMENT

Joint health evaluation is multifaceted and includes clinical joint assessment tools, imaging scanning protocols and scoring systems, and biomarkers of cartilage and bone turnover, angiogenesis, and inflammation. Biomarkers have gained attention lately as a potential tool to guide treatment decisions and monitor joint outcomes.^{10, 11} However, biomarker research in haemophilic arthropathy remains in an explorative stage and the data is heterogenous and not yet broadly translated into clinical practice.¹⁰

Clinical Assessment: includes examination by an experienced haemophilia provider and/or physiotherapist and clinical joint assessment tools¹² including the World Federation of Haemophilia Orthopaedic Joint score (Gilbert Score), Colorado Physical Examination Scale, Petrini Joint Score, and the Haemophilia Joint Health Score (HJHS version 2.1). HJHS measurement properties were most extensively studied and it is validated for use in adults with haemophilia.¹³

Imaging Assessment Tools: include X-rays, magnetic resonance imaging (MRI) and joint ultrasound (US), full or point-of-care.⁸ Computed tomography scans are used less frequently considering the risks of ionising radiation.

Every imaging modality has its own advantages and limitations, with different applications based on the clinical question being asked (see Table 1). The most widely used are Pettersson score by X-ray,¹⁴ Haemophilia Early Arthropathy Detection with US (HEAD-US),¹⁵ the Joint Activity and Damage Exam (JADE),¹⁶ and the recently revised International Prophylaxis Study Group Haemophilic Arthropathy Score by MRI.¹⁷

X-rays are relatively insensitive to early joint changes such as synovial proliferation and are more suited for evaluation of advanced or late osteochondral changes including joint profile irregularities, bone cysts, erosions and narrowing of joint space.⁸

When X-rays are normal and joint disease is early or even subclinical, US and joint MRI provide information complementary to clinical assessment. With excellent spatial resolution, multi-tissue imaging, and high soft tissue contrast MRI remains the gold standard for joint assessment, allowing for evaluation of early and advanced arthropathic changes.⁸ MRI is costly however and not readily available, it may also require sedation in children and is not feasible for simultaneous surveillance of multiple normal or near-normal joints.

US, point-of-care musculoskeletal ultrasound (POC-MSKUS) in particular, has emerged as a promising tool for the temporal evaluation of joint health in PwH and as an adjunct to physical examination. It is particularly helpful in the evaluation of synovial proliferation, a surrogate of disease activity, and in more accurate evaluation of painful musculoskeletal episodes.^{18, 19}

Compared to MRI, US is more sensitive in distinguishing bloody from non-bloody effusions²⁰ and has high to excellent correlation with MRI for detecting pathological soft tissue.²¹ US can also provide a general idea of the osteochondral surface, revealing marginal bone erosions, superficial subchondral cysts, and peripheral articular cartilage defect.²²

Compared to clinical assessment, US has good correlation with HJHS 2.1 scores suggesting that they both provide complementary data on joint disease in adults and children with haemophilia; US detected more early joint changes (synovitis and osteochondral damage), while HJHS 2.1 showed the added value of detecting relevant physical and functional changes.^23-27Another recent systematic review and meta-analysis included fifteen studies reporting on swelling at physical examination and synovial proliferation on US in 2890 joints of 627 patients concluded that joint swelling had low sensitivity for presence of US-detected synovial proliferation in haemophilia, suggesting underestimation of synovial proliferation by physical examination alone.²⁸ Prevalence of subclinical synovial proliferation in the reviewed studies ranged from 0% to 55%.

Another recent cross-sectional study by van Bergen et al.,²⁹ which consecutively included patients with severe haemophilia A, aged 16 years and older, born after 1969, who had access to prophylaxis and were treated at the VanCreveldkliniek (UMC Utrecht, the Netherlands), reported the prevalence of subclinical synovial proliferation on US in the studied patient cohort to be 43%. The authors also proposed two subtypes of subclinical synovial proliferation in this study: active and inactive (fibrotic) synovial proliferation. Active synovial proliferation was defined as a HEAD-US synovium score > 0 and the presence of at least one of the following: synovial hyperaemia on US (JADE score > 0); presence of joint swelling or warmth on physical examination; newly detected synovial proliferation (no history of synovial proliferation based on US assessments and medical records for the last 3 years); current episode treated as synovitis (intensified prophylaxis combined with celecoxib according to the local protocol). Synovial changes not meeting the criteria above were classified as inactive (fibrotic) subclinical synovial proliferation.²⁹ The annualised joint bleeding rate over 5 years was 0.6 (0.2–1.1). Active (22%) and inactive subclinical synovial proliferation (22%) were both prevalent in this cohort. Almost 60% of the patients with synovial proliferation on US did not show warmth or swelling during physical examination. Lastly, synovial, and osteochondral biochemical markers including osteopontin, sVCAM, Coll2-1, COMP, CS846, uCTX-II and TIMP were unable to identify and did not correlate with study patients with active subclinical synovial proliferation. It is important to highlight that the proposed definitions are yet to be validated, but the observations are of interest especially with growing interest in detecting synovial proliferation, which is associated with increased bleeding and progression of haemophilic arthropathy,^{3, 4, 7} at an early or subclinical phase when it is presumably reversible.²⁸

US does have limitations, however. US does not allow for soft tissue discrimination (which includes blood clots, synovium, and fat pads)²¹ or detailed and accurate assessment of the central aspect of the cartilage and subchondral bone.^{22, 30} MRI is recommended when a more detailed evaluation of joint recess contents and articular surfaces is needed.^{8, 20} Furthermore, there is no consensus about whether hemosiderin can be reliably assessed with US or not,³⁰ and US is operator dependent. The importance of training and unique expertise is highlighted when scanning and interpreting paediatric joint scans. In immature articulation of paediatric patients, chronic synovitis may result in hypertrophy of the epiphyseal growth plates, premature closure of growth plates, limb length discrepancy, and angular deformities.³¹ Interpretation of paediatric joint scans requires specialised training, awareness of imaging pitfalls,³⁰ and a sophisticated understanding of the immature skeleton and appearance of growth plates and secondary ossification centres. Furthermore, paediatric atlases with normative values for expected epiphyseal cartilage thickness and vascularization, and normal ossification patterns of the maturing healthy children joints are not available and are urgently needed.³²

Taken collectively, MRI remains the gold standard for the evaluation of haemophilic arthropathy and joint health assessment in PwH. X-rays and clinical exam are less sensitive to early changes. POC-MSKUS is a safe, cost-effective, readily available, and sensitive tool that allows for accurate diagnosis of acute haemarthrosis^{19, 26, 33, 34} and serial evaluation of synovial proliferation (active vs. inactive)^{18, 28, 29} and the overall condition of osteochondral surfaces of index joints in PwH.^{22, 26} It is crucial to point out that while early diagnosis and management of haemophilic arthropathy is becoming more relevant for clinical management and individualised patient regimens²⁹ evidence to conclude that US-detectable findings are sensitive to or could inform changes or intensification of prophylactic therapy in PwH remains insufficient.²³ Significant gaps in the evidence for all assessment tools remain, and these tools complement one another in painting an accurate picture of the joint status.

Lastly, assessing the impact of bleeding on the musculoskeletal system is best completed according to The World Health Organization's (WHO) International Classification of Functioning, Disability and Health (ICF) model with its domain-related outcome assessment instruments. This ensures that disciplines are connected by the same model with the focus on the impact of the disease on body structure and function as well as activities and participation.² For activities and participation, the World Federation of Haemophilia and WHO recommend The Haemophilia Activities List (HAL), the PaediatricHAL, Functional Independence Score in Haemophilia (FISH) to name a few. It also recommends quality of life assessment tools including Patient-Reported Outcomes, Burdens, and Experiences (PROBES), and Haemophilia-Specific Quality-of-Life Questionnaire for Adults (HAMEO-QoL-A).²

4 PREVENTION AND MANAGEMENT OF HAEMOPHILIC ARTHROPATHY

Early institution of primary prophylaxis is the gold standard for long-term preservation of joint health and range of motion in PwH.⁴ Primary prophylaxis started before any joint disease is documented, usually in children aged 2 years and younger, can be achieved using haemostatic factor replacement therapy: plasma derived, standard or extended half-life products, and bypassing agents for inhibitor patients.^{4, 5, 9} These therapies are costly however and patients in developing countries typically receive low-dose prophylactic regimens (10–15 IU/kg, 2–3 times a week)² only when available given limited access to safe therapies, specialised centres that provide comprehensive care, and limited budget allowance for factor concentrates.^{35, 36}

More recently, non-factor replacement therapies³⁷ including emicizumab, a bispecific monoclonal antibody, led to greater patient satisfaction and reduced prophylactic treatment burden given its infrequent, subcutaneous administration in patients with factor VIII deficiency with or without an inhibitor.³⁸

Despite the institution of individualised, high-dose, and early prophylactic regimens, haemophilic arthropathy continues to occur.^{3, 5} MRI data indicate that joint damage can still be identified in about 20% of clinically asymptomatic patients even while receiving prophylaxis,^{4, 5} and maintaining factor levels over a certain threshold and essentially converting severe haemophilia patients to mild or moderate ones does not fully prevent subclinical haemarthrosis and joint disease.³⁹ Joint abnormalities on imaging and clinical outcome measures have been reported even in mild, moderate PwH and in haemophilia carriers as well.^40-42 This further supports the value of early detection of subclinical bleeding, synovial proliferation, and early osteochondral damage in “normal” and “healthy” joints by exam complemented by imaging.

Treating patients with haemarthrosis and haemophilic arthropathy requires a multidisciplinary approach involving haematology, physiotherapy, and orthopaedic surgery with the patient at the centre of the care model.

Acute Haemarthrosis: Treatment goals include effective and rapid haemostatic management, rebleed prevention, pain management and rest (with or without splinting).² Non-weight-bearing on lower limb joints is important. After resolution of the acute injury, subsequent restoration of joint function, range of motion and stability through a well-paced, supervised, physiotherapy program, which includes gentle muscle stretching and isometric exercises is necessary.⁸

Early joint aspiration under haemostatic coverage and in strictly aseptic conditions is a subject of ongoing debate, but preliminary data is promising with immediate relief, accelerated resolution, functional and clinical recovery, and resumption of normal activity.^43-45 It appears to be safe, effective with benefits outweighing the risks for patients with a first episode of joint bleeding in a previously healthy joint. Intra-articular injection of corticosteroids,⁴⁶ viscosupplement (hyaluronic acid), and/or platelet-rich plasma may also be considered but randomised clinical trial data is not yet available.⁴⁷ POC-MSKUS is useful in evaluating the intra-articular blood phase (liquid versus coagulated) and in guiding the needle path to minimize chance of injury during arthrocentesis and joint injection. Cyclo-oxygenase −2 inhibitors are increasingly used because of their anti-inflammatory, antiangiogenic and analgesic properties,⁴⁸ as are short oral corticosteroid courses.

Haemophilic Synovitis and Advanced Arthropathy: When well-tailored, individualised prophylaxis and physiotherapy regimens fail to abort the bleed-synovitis-rebleed cycle, strategies targeting chronic hypertrophic haemophilic synovitis should be attempted to halt the long-term damage of blood in the joint. Synovectomy (synoviorthesis) should be limited to patients who failed conservative measures and aims to destroy and inactivate the inflamed, hypertrophied synovium and stop the bleeding. Synovectomy^{49, 50} can be non-surgical including chemical and radioactive, and surgical including arthroscopic and open.

Non-surgical synovectomy: Chemical synovectomy⁵¹ is safe, inexpensive, and Rifampicin and oxytetracycline continue to be used in developing countries as a good alternative to radioactive synovectomy (RS), which uses a beta-emitting radioisotope Yttrium-90 (⁹⁰Y) for the knees, and Rhenium (¹⁸⁵Re) for elbows and ankles coupled with a carrier particle. Patients should not undergo more than three RS within a 6-month interval.^{52, 53} Non-surgical synovectomy is effective in decreasing the number of joint bleeds, controlling pain, and improving range of motion, but RS appears to be more effective than chemical synovectomy in larger joints with a mid- and long-term efficacy of 75%–80%.⁵² A rehabilitation program should be put in place after the first 48 hours to regain strength and mobility. Patients should be reassessed closely every 3–6 months documenting haemarthrosis, synovial thickness and inflammation, and functional joint status. The shortest interval for repeating RS is at least 6 months.^{49, 52-54}

Five decades of data show that RS is cost-effective, minimally invasive, safe, and well-tolerated. It should be considered early and first when conservative measures fail at preventing progression of haemophilic arthropathy in PwH with or without inhibitors.^{50, 52, 53, 55} This is important to prevent irreversible osteochondral damage. A review of the American Thrombosis and Haemostasis Network dataset shows no increase in prevalence of malignancy in PwH who had undergone no more than three early consecutive RS within a 6-month interval.⁵⁶ When RS fails to halt synovitis, a surgical synovectomy should be then considered.⁴⁹ Unfortunately, limited access to and availability of radioisotope has increased chemical synovectomies in developing countries especially post COVID-19 pandemic.

Another non-surgical approach is successful, selective, angiographic embolization of joint arteries. This has been reported for recurrent haemarthrosis of the elbow and the knee,⁵⁷ and more recently, preoperative genicular artery embolization for patients undergoing total knee replacement reduced moderate to severe synovial hyperaemia in patients with haemophilia, potentially protecting the joint from bleeding during rehabilitation.⁵⁸

Surgical Synovectomy and Minimally Invasive Orthopaedic Procedures: Major surgical procedures such as total joint replacement, arthrodesis and open synovectomies were standard treatment options for end-stage haemophilic arthropathy. Minimally invasive orthopaedic procedures have become increasingly popular however and have now become the first choice when considering invasive surgical treatment. These procedures reduce injury, allow for minimal but crucial repair, which interrupt the vicious cycle and enhance recovery.

Surgical synovectomy, open or arthroscopic, is indicated for patients with subacute or chronic synovitis with no response to medical treatment for at least 6 months and with moderate radiological scores.^{49, 50, 59} Relative contraindications include high neutralising inhibitor titer and advanced arthropathy. Although > 80% effective in decreasing subsequent haemarthrosis, open synovectomy is expensive, requires a prolonged hospital stay and haemostatic factor replacement, and comes with several possible serious complications including loss of movement, infection, stiffness, fracture and postoperative haemarthrosis.^{49, 60} On the other hand, arthroscopic synovectomy is less invasive, more cost-effective, requiring a shorter hospital stay and less haemostatic factor therapy.^{49, 50} It is effective at removing the pathologic inflamed synovium and in advanced cases debridement of loose bodies, microfracture and repair of limited cartilage defects allows for delaying or even aborting total joint replacement all together. Recovery appears to be rapid and range of motion improve especially when done in knees, elbows, and ankles and coupled with early continuous passive motion.⁶¹

Excision of the radial head is another consistently reliable, minimally invasive procedure that appears to prolong the functional life of the elbow joint. With proper patient selection it dramatically reduces the rate of joint bleeds, improves forearm function and rotation up to 60%, and decreases pain without compromising elbow and wrist joint stability.⁶² Lastly, platelet-rich plasma (PRP) therapy is emerging as another minimally invasive orthopaedic procedure which relies on the chondroregenerative properties of growth factors in PRP for patients with haemophilic arthropathy and chronic synovitis. Data is yet to be validated, but preliminary reports show that PRP injections effectively reduce inflammation, pain, bleeding episodes and improves joint range of motion.⁶³

Minimally invasive orthopaedic procedures can truly grant PwH with haemophilic arthropathy years without pain. Success of minimally invasive procedures including arthroscopic synovectomy is dependent on careful patient screening and selection (young, timely intervention, early arthropathy), and a multidisciplinary approach with experienced haematologists, orthopaedic surgeons, and physiotherapists. Patient motivation and support is crucial.⁴⁹

Orthopaedic Surgery in Advanced Arthropathy: When all else fails, total joint replacement (knee/hip) is successful at diminishing pain, reducing bleeding episodes and improving function, with decades of experience and publications to support it. Minimally invasive options should be carefully evaluated and considered first, and the timing of orthopedic surgeries including total joint replacements should take the patients age, functional limitations, quality of life and bone quality. Limited access to safe clotting factor concentrates delays surgical interventions in developing countries.

5 AGING WITH HAEMOPHILIC ARTHROPATHY

Left untreated, haemophilic arthropathy is devastating to PwH. As PwH age, they continue to face multiple musculoskeletal problems and their bone and joint health remains compromised.³ During peak growth and bone building years of childhood several older patients did not have access to safe factor products and individualised prophylactic regimens. This left them in self-perpetuating, chronic, bleed-chronic synovitis-rebleed cycles. When injured, patients were held out of school and asked to sit out childhood activities. Acute management of musculoskeletal bleeding episodes included splinting, immobilisation to a bed or a wheelchair, and limiting weight bearing. This resulted in decreased weight bearing time and intensity, which at any age leads to loss of bone density. Repeated haemarthrosis in this population resulted in osteochondral derangement and synovial thickening. The smoothness and congruency of the joint surfaces was lost and uneven joint surfaces ground on each other and pinched the synovial lining during movement. Simple weight bearing and/or continued movement of the joint with these internal articular surface derangements progressed joint arthropathy further. It is thus not surprising that the more severe haemophilic arthropathy was, the less often and with less load these patients bore weight. Decreased weight bearing load over time is one of the probable causes of low bone mineral density (BMD) in PwH compared to healthy individuals. immobilisation, human immunodeficiency virus and/or hepatitis C virus, and vitamin D deficiency are known risk factors but a clear pathogenesis is yet to be clarified. Thrombin stimulates differentiation and activity of osteoblasts and it is hypothesised that low factor levels could impair bone remodeling and mineralization, and that deficiency of FVIII-VWF complex augment osteoclastogenesis.^{64, 65} A recent multicentre study by Klintman et al. showed that BMD differed in persons with severe haemophilia on prophylaxis as compared to those treated on-demand, with patients on prophylaxis displaying high degree of normal BMD comparable to mild haemophilia age-matched patients. The exact type of prophylactic regimen did not reflect on BMD.⁶⁴ Furthermore, bone health of haemophilia A patients using emicizumab does not appear to be declined compared to factor prophylaxis, and biomarkers of bone resorption/formation, cartilage degradation/synthesis, and inflammation did not change significantly during emicizumab therapy.⁶⁶ The search for sensitive, specific, and reliable bone turnover markers in PwH continues. Prevention of osteoporosis is of paramount importance and is ideally done with effective joint bleed prevention with prophylactic therapy, limiting duration of immobilisation and weight bearing restrictions, regular exercise program that includes strength training, weight management, calcium and vitamin D rich diet, and counseling against smoking and alcohol use.⁶⁵ Patients with reduced BMD are much more susceptible to fractures and injuries after a fall and risk prevention and mitigation is key. Partnering with an experienced physiotherapist is important to initiate balance training, discuss the dangers of injury and fractures, and evaluate home and work environments. Balance training decreases the amount of risk that the patient will face from falling and incorporates strengthening and stretching into a patients daily routine of balance exercise.^67-69

Accommodations for patients with arthropathic joints, painful movement and/or limited function can range from a very minor heel lift to oral anti-inflammatory agent, anti-inflammatory intra-articular injection, arthroscopic joint debridement, surgical joint replacement, and all the way up and including thorough structural changes made to a patients home. Accommodations to patients homes can keep them living independently and in their home for a much longer time.^{70, 71} This can be a ramp to replace porch stairs, a motor operated stair chair for a two plus story home, modifying countertop heights, and positioning of toilets, sinks and showers.⁷²

6 CONCLUSION

In conclusion, with development of novel factor replacement therapies, nonfactor treatments and gene therapies, severe haemophilia phenotype is being shifted to mild and moderate, phenotypes that have been historically neglected and continue to have unmet needs. Zero and low joint bleeding rates are expected with highly personalised and effective prophylactic regimens, but considerable evidence exists that subclinical bleeding with synovial proliferation continues. We are yet to fully understand and effectively prevent joint and bone disease in haemophilia in both developing and developed countries.

With proper training and practice, US can facilitate serial joint health monitoring complementing clinical outcome scores and joint MRI, allowing for early diagnosis and intervention at a stage where joint disease is hypothesised to be reversible, and provide objective insights into the efficacy of treatment regimens and gene therapy trials. More high-grade evidence is needed, however, to substantiate its role. Handheld probes are being optimised with artificial intelligence algorithms for provider and patient use. A deeper understanding of the unique pathophysiology of haemophilic arthropathy is key, and the search for biomarkers for cartilage degradation/synthesis, bone resorption/formation, angiogenesis and synovial inflammation should continue, only then will we be able to propose early, directed targets for management of haemophilic arthropathy.^{8, 29, 59} Advances in haemophilia therapy and imaging are exciting, but one should not forget the importance of preserving bone marrow density, quality of life, and supporting our PwH as they navigate through aging. As haemophilia treatment providers we should strive to allow our patients to live independently, in their homes, surrounded by their loved ones for as long as they can, and only a multidisciplinary team can achieve that.

CONFLICT OF INTEREST STATEMENT

The authors have no competing interests.

ETHICS STATEMENT

Ethical approval(s) or informed consents were not needed for this review manuscript.