Guidelines for the Management of Acute Joint Bleeds and Chronic Synovitis in Haemophilia a United Kingdom Haemophilia Centre Doctors’ Organisation (UKHCDO) Guideline

Home , Arthropathy, Hemarthrosis, Joint effusion

Haemophilia (2017), 1–10 DOI: 10.1111/hae.13201

REVIEW ARTICLE Guidelines for the management of acute joint bleeds and chronic synovitis in haemophilia A United Kingdom Haemophilia Centre Doctors’ Organisation (UKHCDO) guideline

J. HANLEY,* A. MCKERNAN,† M. D. CREAGH,‡ S. CLASSEY,§ P. MCLAUGHLIN,¶ N. GODDARD,¶ P. J. BRIGGS,* S. FROSTICK,** P. GIANGRANDE,†† J. WILDE,‡‡ J. THACHIL§§ and P. CHOWDARY¶ ON BEHALF OF THE MUSCULOSKELETAL WORKING PARTY OF THE UKHCDO *Haemophilia Centre, Royal Victoria Inﬁrmary, Newcastle upon Tyne; †Department of Haematology, Derby Hospitals NHS Foundation Trust, Derby; ‡Haemophilia Centre, Royal Cornwall Hospitals NHS Trust, Truro; §Haemophilia Centre, Guys and St. Thomas’ NHS Foundation Trust, London; ¶Katharine Dormandy Haemophilia and Thrombosis Centre, Royal Free Hospital, London; **Institute of Translational Medicine, University of Liverpool, Liverpool; ††Haemophilia Centre, Churchill Hospital, Oxford; ‡‡Haemophilia Centre, Queen Elizabeth Hospital Birmingham, Birmingham; and §§Haemophilia Centre, Manchester Royal Inﬁrmary, Manchester, UK

Keywords: arthropathy, haemarthrosis, haemophilia, radioactive synovectomy, synovitis

Introduction Although the exact pathogenesis of haemophilic arthropathy has not been elucidated, several in vitro ani- Prior to the availability of clotting factor concentrates, mal and human experiments have demonstrated blood- the natural history of arthropathy in severe haemophilia induced changes in all the ‘components’ of large synovial was well described, with recurrent bleeds into joints joints including the synovium itself, cartilage and the leading to progressive joint damage and ultimate subchondral bone. The two major changes in the syn- destruction with associated functional problems. The ovium are hypertrophy and hypervascularity. The key prophylactic use of factor concentrates has dramatically stimulant for these changes is iron released into the syn- modified the natural history of haemophilic arthropa- ovial fluid, which is both pro-inflammatory and proan- thy. The early use of prophylaxis can prevent joint giogenic [2–4]. The consequence of neovascularisation bleeding and avoid the cycle of damage associated with in the synovium is predisposition to more bleeding since recurrent haemarthrosis. Children who are treated on these new vessels are friable. This leads to a vicious circle prophylaxis schedules often reach skeletal maturity of bleeding, iron accumulation, synovial hypertrophy with well-preserved joint function. However, in addi- and hypervascularization leading to further bleeding tion to the clinically obvious bleeds, recurrent subclini- and ultimately progressive joint damage. Interventions cal episodes may also contribute to joint damage [1]. aimed at preventing or breaking this cycle are key strate- Once joint arthropathy is established, prophylactic gies for the preservation of joint function in people with treatment may not prevent further joint deterioration. haemophilia. Bleeding is particularly problematic in the diarthrodial-hinged joints such as the knee, elbow and ankle. It is therefore important that acute bleeds in Correspondence: John Hanley, Haemophilia Centre, Royal Vic- patients on either ‘on demand’ or ‘prophylaxis’ regimens toria Infirmary, Queen Victoria Road, Newcastle upon Tyne are treated optimally with the aim of minimizing the NE1 4LP, UK. synovial proliferation secondary to bleeding which is Tel.: 0191 282 4159; fax: 0191 282 0814; central to the pathogenesis of arthropathy. e-mail: [email protected] A recent survey has shown significant variation in Accepted after revision 24 January 2017 practice in the UK [5], which suggests that there is a

© 2017 John Wiley & Sons Ltd 1 2 J. HANLEY et al. need for guidelines as a framework for best practice. The objective of initial factor replacement therapy is Evidence-based guidelines were developed summariz- to improve haemostasis in order to arrest bleeding. ing best practice for the assessment and management Subsequent treatment aims to prevent recurrent bleed- of acute joint bleeds and chronic synovitis in persons ing and the onset and progression of joint damage. with haemophilia. This guideline does not include sur- Prompt treatment of bleeding is important and as it gical procedures such as surgical synovectomy, has the advantage of potentially preventing permanent arthrodesis and arthroplasty. damage and this is best achieved with home therapy and patients have been traditionally advised to treat Materials and methods any early symptoms and signs of bleeding. The advan- tages of home therapy include significant decreases in The information contained in this guideline was days lost from work and school, days hospitalized and gathered from an appropriate and pertinent literature hospital visits [8]. In those with established arthropa- search in relation to UK practice. The writing group thy, it may be difficult to distinguish early bleeds from produced the draft guideline, which was subsequently flares of arthritic pain [9,10]. revised by consensus by members of the United Kingdom Haemophilia Centre Doctors’ Organisation Haemostatic management of patients with (UKHCDO) Advisory Board. The ‘GRADE’ system Haemophilia A and B was used to quote levels and grades of evidence, details of which can be found at: http://www. Dose of initial therapy. Early bleeds in patients who gradeworkinggroup.org. Strong recommendations are on prophylaxis are usually adequately treated by a (grade 1, ‘recommended’) are made when there is single dose with resumption of usual prophylaxis there- confidence that the benefits either do or do not out- after. Early studies attempted to investigate the opti- weigh the harm and costs of treatment. Where the mum treatment for bleed resolution. The interpretation magnitude of benefit or not is less certain, a weaker of these studies is complicated by the variable severity grade 2 recommendation (‘suggested’) is made. Grade of the bleeds included and the lack of standardized 1 recommendations can be applied uniformly to assessment criteria. Prior to the availability of factor most patients, whereas grade 2 recommendations concentrates, it was demonstrated that doses of cryo- require a more individualized application. The qual- precipitate achieving FVIII plasma levels of 28%, 35% ity of evidence is graded as high (A), based on high and 53% correlated with successful treatment in 90%, quality randomized clinical trials, moderate (B), low 95% and 99% of bleeding episodes respectively [11]. (C) or very low (D). Many early studies demonstrated that prompt, low doses of factor were equally effective in resolution of Management of acute haemarthrosis bleeds [12–14] but more rapid improvements were demonstrated with higher factor levels [15]. Prophylaxis is the current standard of care, although There are few randomized trials which have com- individual patients continue with on-demand treat- pared different initial doses of factor. One study found ment [6]. In patients receiving on-demand treatment, À that an initial treatment with factor VIII 28 IU kg 1 spontaneous and traumatic bleeds are common [7]. À was no more effective than 14 IU kg 1 in terms of For those receiving prophylaxis, trauma is often the time to symptom resolution and requirement for fur- major precipitant of bleeding but inadequate trough ther treatment [16]. However, the beneficial effect of levels, missed doses, underlying synovitis or estab- a higher initial dose has been demonstrated where lished joint damage may be contributory factors. The patients with levels of 40% compared to 20% had signs and symptoms associated with joint bleeds are shorter time to full functional recovery [17]. Pivotal shown in Table 1. A re-bleed is defined as worsening studies of recombinant factor VIII and Factor IX con- of the condition either on treatment or within 72 h centrates, including studies of extended half-life after stopping treatment [7].

Table 1. Symptoms and signs of joint bleeds.

Bleed type Symptoms Signs References Early bleed A sensation of fullness, stiffness, Normal motion [7,73] discomfort, pain or tingling at the end of range of movement in the absence of trauma. Often described as aura by patients and at times can be difﬁcult to distinguish from arthritic pain Moderate Pain Some swelling. Restriction of movement Severe Severe pain Marked swelling. Almost complete restriction [11] of movement ‘Joint immobilizing bleeds’ [16]

Haemophilia (2017), 1--10 © 2017 John Wiley & Sons Ltd JOINT BLEEDS AND SYNOVITIS IN HAEMOPHILIA 3 products, have shown variation in the doses used, and there is inadequate response in the first 36 to 48 h the median dose used for bleed resolution was around (1C). À 30 IU Kg 1 in most studies, with 90 to or -95% of the bleeds requiring one or two infusions for complete Haemostatic management of patients with resolution [12,18–21]. In a retrospective study factor, À inhibitors to Factor VIII and IX doses of 25 to –40 IU kg 1 per bleeding episode were inferior only to prophylaxis in terms of reducing the Patients with inhibitors do not necessarily have an progression of arthropathy [22]. increased frequency of spontaneous joint and soft tissue bleeds compared to non-inhibitor patients but do have Subsequent factor treatment. Subsequent factor ther- a higher tendency to develop target joints, which can apy depends on the severity of the bleed. Serial ultra- be difficult to manage. Many inhibitor patients are now sounds of joint bleeds show a persistent effusion for a well established on home therapy for the treatment of few days and confirms previous clinical observations bleeds. However, to enable the optimal management of that time to complete resolution is longer than the bleeding episodes, there should be close liaison between time to loss of pain [23]. Early minor bleeds are gen- the patient and centre staff. As for bleeds in patients erally adequately treated by a single dose, resuming without inhibitors, treatment should be initiated as usual prophylaxis thereafter. For more significant early as possible to restrict the extent of the bleed. bleeds or in those who are not on prophylaxis, treat- Early treatment optimizes the chance of a single treatment should continue for a couple of days after symp- ment settling the bleed, thereby minimizing patient tomatic relief. This is usually achieved by a daily incapacity and factor consumption. Troublesome dosing schedule. For severe bleeds, 12-hourly dosing bleeds should be reviewed early at the haemophilia cen- may be required initially. tre and there should be a low threshold for admitting the patient for optimization of bleed management. Non - Inhibitor patients - Recommendations Patients with inhibitors should be treated with either the activated prothrombin complex concentrate aPCC â 1. All patients with severe haemophilia A and B and (FEIBA , Factor Eight Inhibitor Bypassing Activity) â other patients at risk of joint bleeding should be or recombinant factor VIIa (Novoseven , rFVIIa). offered home treatment (1B). Through personal experience of bleed responsiveness, 2. All patients must have an individual treatment patients tend to prefer one product over the other. Some protocol that explains the management of joint patients, however, favour the use of rFVIIa for the and other bleeds with instructions on initial treatment of certain types of bleed and FEIBA for dosage, frequency and when to contact the haemo- others. Furthermore, some patients find rFVIIa effective philia centre for advice (1C). in arresting bleeds that can be treated early but favour 3. The initial treatment of early and moderate bleeds aPCC for bleeds which have been established for longer should aim for a peak factor VIII/IX of between 50 to before they are treated. It is therefore becoming increas- À À 60 IU dL 1. This is equivalent to 25 to 30 IU kg 1 ingly common for patients to have both products sup- for severe haemophilia A for standard and extended plied for home treatment needs. aPCC is usually dosed À À half-life products and 40 to 60 IU kg 1 for severe between 50 and 100 IU kg 1 depending on the severity haemophilia B with extended half-life factor IX being of the bleed and further doses administered between dosed at the lower end of the recommended range. 12- and 24 h-intervals until the bleed is fully controlled. Early bleeds often do not require a second infusion, The total dose of aPCC administered over a 24-h period À and moderate bleeds often respond to a single infu- should not exceed 200 IU kg 1. rFVIIa is now usually À sion but may require up to two infusions (1B). given as a single dose of 270 lgkg 1 which is as effica- 4. Children may require more frequent or higher cious and more convenient than the previous conven- À doses as they have a shorter factor half-life com- tional two to three doses of 90 lgkg 1 given every two pared to adults (1B). hours [24,25]. If a bleed is not responding to one agent, 5. For joint immobilizing bleeds, higher initial doses then the other can be used in the hope that this will be are recommended which aim to raise the peak fac- effective. If both products are not effective, use of À tor VIII/IX level to 60 to 80 IU dL 1. Doses sequential alternating therapy can be considered in the should be administered every 24 h until complete hope of achieving a synergistic effect [26]. Use of resolution of pain. For severe bleeds, more fre- tranexamic acid should be considered to optimize treat- quent administration may be required in the initial ment of bleeding episodes. The concomitant use of 48 h with standard factor VIII or IX(1B). tranexamic acid and rFVIIa has not been associated 6. Patient education on the identification and man- with increased thrombotic risk [27]. Historically, how- agement of bleeds should be ongoing (1C). ever, due to a perceived risk of thrombosis, tranexamic 7. Patients on home therapy should be encouraged acid has been contraindicated in combination with to contact the haemophilia centre for review if aPCC. It is now considered that this risk is likely to

© 2017 John Wiley & Sons Ltd Haemophilia (2017), 1--10 4 J. HANLEY et al. have been exaggerated and it is becoming increasingly Paracetamol (acetaminophen) is the accepted basis of common practice to use tranexamic acid alongside acute pain relief, although there are no randomized trials aPCC [28]. in haemophilia. Non-cyclo-oxygenase-2 (COX-2) selective non-steroidal analgesics (NSAID’s) such as ibupro- Inhibitor patients - Recommendations fen carry an excess incidence of gastrointestinal bleeding 1. Inhibitor patients should be encouraged to be on a and perforation and so are not advisable. By comparison, home treatment programme and bleeds should be studies in haemophilia have shown COX-2 selective treated as early as possible (1A). NSAID’s to have a lower risk for gastrointestinal side 2. There should be close liaison with haemophilia effects [30] and offer relief of pain and chronic synovitis centre staff members to agree upon appropriate [31,32]. However, COX-2 NSAID’s do have a greater management of difﬁcult bleeds (1C). À À risk for cardiovascular disease, particularly in those with 3. aPCC 50–100 lgkg 1 or rFVIIa 270 lgkg 1 as a À established cardiovascular disease, cerebrovascular dis- single dose (or 90 lgkg 1 2–3 hourly) are equally ease, peripheral arterial disease, and mild to severe heart acceptable treatments for joint or soft tissue bleeds failure, when they are contraindicated [33]. Both selec- with repeated doses as necessary. The frequency of tive and non-selective NSAID’s may impair renal func- infusion and duration of treatment should be tion. Therefore, COX-2 selective NSAID’s, for example, determined by the clinical response (1B). celecoxib and etoricoxib, which are effective and safe in 4. The total daily dose of aPCC should not exceed À haemophilia joint bleeds can be used with due caution 200 IU kg 1 (1B). for the shortest duration possible [34]. 5. Tranexamic acid can be considered as adjunctive Opioid analgesics range in strength from weaker therapy to aPCC and rFVIIa (2C). opioids such as codeine, dihydrocodeine and tra- 6. Sequential alternating treatment of aPCC and madol, to the stronger morphine, oxycodone and fen- rFVIIa can be considered for the management of tanyl preparations. Data to support the use of opioid limb/life threatening bleeds, and this is associated analgesic in haemophila are limited to anecdotal expe- increased risk of thrombosis (2B). rience and derived from non-cancer analgesic guidance. For acute analgesia, a short acting preparation Non-haemostatic management may be advisable. Joint aspiration. Small, non-randomized studies have Pain relief - recommendations assessed the utility of joint aspiration in acute haemar- 1. Ice cooling as part of the PRICE process may alle- throses and some have reported a good long-term out- viate pain (1C). come [29]. However, there is no deﬁnite evidence that 2. Analgesia should be prescribed by a stepwise pro- joint aspiration helps to accelerate the recovery from a cess of progression; of which paracetamol is gener- haemarthrosis or improve outcome. There is the ally the most appropriate initial treatment (1C). potential concern of introducing infection into the 3. COX-2 selective NSAID’s are effective and safe in joint. Aspiration may be helpful when there is severe haemophilia joint bleeds (1B). bleeding and pain with a tense haemarthrosis [23]. If 4. Opioid analgesia is appropriate in patients with aspiration is performed, factor cover and an aseptic moderate to severe or refractory pain (1C). technique are required. Joint aspiration - recommendations Physiotherapy 1. Joint aspiration is not routinely recommended unless there is concern about potential septic Although consensus guidelines recommend physiother- arthritis (1C). apy following acute haemarthrosis [12], there is a very 2. Joint aspiration may be useful for pain relief in limited objective evidence base in relation to the opti- tense haemarthrosis under appropriate haemostatic mal timing and types of rehabilitation strategies fol- therapy (2B). lowing resolution of a joint bleed. Clinical physiotherapy intervention is aimed at symptom con- Pain relief. In acute haemorrhage, ice cooling as part trol, prevention of bleed recurrence, prevention of of the PRICE protocol may alleviate pain (see Physio- joint damage and restoration of full function and therapy section below). Using the principles of an activity. Early management strategies are often encap- analgesic ‘ladder’, common to pain relief guidance, sulated within the mnemonic PRICE - Protection, medication should be prescribed with a stepwise pro- Rest, Ice, Compression and Elevation. gression from ‘simple’ paracetamol through to ‘strong’ opioid analgesics. This process matches the ceiling Acute phase effect of analgesic drugs to the degree of pain present, such that if pain is severe or analgesia ineffective, then Protection and joint rest—The aim in the early stages an ascent of the ladder is recommended. is to relieve the acute pain and decrease the risk of re-

Haemophilia (2017), 1--10 © 2017 John Wiley & Sons Ltd JOINT BLEEDS AND SYNOVITIS IN HAEMOPHILIA 5 bleeding. Unloading the affected joint protects injured [49]. No advantage has been established for the effec- tissue from excessive mechanical stress thereby mini- tiveness of any individual protocol (method of compres- mizing ongoing bleeding and aiding healing [35–37]. sion/compression pressure). Guidelines suggest that if Animal models have demonstrated that weight bearing compression is to be used, then it should configure as in the presence of a haemarthrosis may have a delete- best as possible to the limb/joint shape, provide a grad- rious effect on joint cartilage [38]. However, pro- uated compressive force and be comfortable for the longed rest can negatively affect joint function individual [37,50]. through reduction in muscle strength affecting tissue biomechanics and dynamic joint control [39,40]. Sub-acute phase and rehabilitation. Once adequate Where bleed symptoms persist for more than 24 h, haemostasis is assured, and both subjective and objec- a period of non-weight-bearing/partial weight-bearing tive signs of improvement are forthcoming, the reha- for a lower limb joint bleeds and immobilization for bilitation programme should commence. At this stage upper limb bleeds has been recommended. World Fed- it is also important to consider other aspects of joint eration of Haemophilia (WFH) guidelines suggest health such as the chronicity of joint swelling, the risk immobilization until pain resolves [7] with empirical of target joint formation as well as already established reports suggesting initial early bed rest for 1 day and arthropathic or biomechanical changes in the affected avoidance of weight bearing for up to 4 days [41]. joint and the wider musculoskeletal system. However, in practice, relative rest/protection should Chronic effusions and long-term changes to a joint continue until all clinical symptoms of haemarthrosis capsule as a result of recurrent effusions, coupled with (acute swelling, acute palpable localised tissue inflammatory mediators, are likely to decrease propri- warmth, acutely reduced range of joint motion and oceptive ability in a joint [51,52]. Following injury to acute joint pain) have resolved and potentially a joint, the barrage of nociceptive input from the joint beyond, taking account of the severity and site of the pressure and injury site decrease afferent activity in bleed. Depending upon age, pre-existing disability, site the surrounding muscles and disrupt joint mechanore- and severity of the bleed, the patient may require hos- ceptors [53,54], further affecting proprioception. Joint pital admission and bed rest or may be managed as an damage associated with existing arthropathy has also outpatient with the appropriate aids such as a sling, been shown to decrease afferent activity in the sur- brace, cast, walking aid or wheelchair. rounding muscles [55]. If both scenarios were to occur together, such as in an acute haemarthrosis, a large Ice/cryotherapy—Cryotherapy continues to be recom- reduction in activation may prevent the threshold for mended as a method of acute pain relief in haemarthro- stimulation of muscle hypertrophy from being sis [7,12]. The mechanism by which pain relief is reached, potentially impeding rehabilitation [56]. achieved is thought to be through a combination of Early intervention to reduce nociceptive input may decreasing local nerve conduction velocity [42], as well reduce potential motor control dysfunction. Therefore, as central mechanisms mediated at spinal cord level the introduction of exercises aimed at maintaining [43,44]. The clinical effect of ice on bleeding is subject to motor control patterns may be of considerable impor- some debate, with a suggestion that it may have a delete- tance [53]. Motor control of everyday actions is task rious effect on coagulation [45]. A systematic review specific; therefore, functionally orientated exercise suggests that cold application does not have a strong should be incorporated as early as possible in the man- clinical evidence base in the management of joint swel- agement [57]. Exercise programmes should encompass ling [37]. However, subjectively, haemophilia patients local joint control and global conditioning, such as have reported that the benefits of cryotherapy include weight-bearing activity to improve joint position sense both reduced pain and swelling with no reported bleed- [55] and strengthening activity [23]. This will also pro- ing issues. [46–48]. Following a systematic review of the mote neuromuscular feed-forward changes – important literature, crushed ice is probably the safest method of for the overall functional outcome and to minimise the application, but other modalities including gel packs risk of future recurrence of injury [57]. and cyrocuffs can be used. Application should not exceed 20 minutes at two hourly intervals and always be Physiotherapy – recommendations guided by levels of pain and discomfort [37]. 1. Review by a physiotherapist is recommended as soon as a patient presents with or reports a joint Compression and elevation. Both compression and ele- bleed – advising and educating on the anticipated vation are potentially beneficial treatment adjuncts timeline for rehabilitation (1B). when swelling and pain compromise function and 2. Initial assessment should be comprehensive – attempts to restore the pressure gradients within including history of trauma/injury as well as base- affected tissue are perceived to be beneficial [37]. These line joint health/function (1C). changes are usually short-lived, with recurrence of swel- 3. Rest/immobilization – should be monitored closely ling once the limb is returned to a dependent position and kept to a clinical minimum (1C).

4. Cryotherapy to be used with the limb in elevation, attractive option for patients with inhibitors or those preferably with crushed ice, and for no more than unsuitable for surgical synovectomy [61]. Radioactiv- 20 mins per application. Gel packs and cyrocuffs ity emitted by the appropriate isotope causes atrophy can be used, but ice cubes or cooling blocks that do and fibrosis of the synovium, with a consequent reduc- not mould to joint contour should be avoided (1B). tion in the bleeding tendency. Patients suitable for 5. Compression/elevation – to be used according to consideration for radioactive synovectomy are those patient comfort and monitored closely (2B). with demonstrable synovitis, not just evident on clini- 6. Rehabilitation should focus on regaining dynamic cal examination but also proven by soft tissue imaging joint control and towards improvement in baseline such as ultrasound and/or MRI. function. Techniques used in routine clinical prac- A number of isotopes have been used for this proce- tice can be used in persons with haemophilia. This dure, including phosphorus P32, yttrium Y90, erbium should be preceded by a discussion with the Er169, gold Au198 and rhenium Re186 [62]. Proper- haematologist to ensure appropriate clotting factor ties to be taken into consideration when selecting a support if indicated (1C). suitable isotope include particle size, type of emission, range of soft tissue penetration and half-life. Re186 Management of chronic synovitis and target and Y90 have been increasingly adopted in recent years and they have relatively short half-life of 2.7 to joints 3.7 days and a soft tissue penetration range of 1 to 5 mm. The precise dose used will depend on both the Target joint definition joint and the size of the patient. This procedure can Bleeding into joints leads to the development of syn- only be performed in hospitals which have the neces- ovitis, which is characterized by a painless chronic sary equipment and protocols in place to handle swelling of the affected joint. Such joints are more radioactive isotopes [63]. It may be cost-effective to prone to recurrent bleeds and progressive joint dam- arrange for several patients to receive treatment on the age. The term ‘target joint’ is defined as a joint with same day. An infusion of clotting factor concentrate is recurrent bleeds, and the most recent International required prior to the procedure. The isotope is intro- Society of Thrombosis and Hemostasis (ISTH) defini- duced into the joint using a 16 or 18 gauge needle, tion suggests three or more spontaneous bleeds into after aspiration of any intra-articular effusion or blood. single joint within a consecutive 6-month period. The Although not obligatory, fluoroscopic monitoring is joint ceases to be target joint when there have been advised to ensure correct needle placement to avoid less than two bleeds into the joint within 12 consecu- extra-articular leak of the radioisotope and to ensure tive months [58]. The PEDNET registry has a more that no blood vessels are inadvertently punctured. narrow definition of three or more bleeds in a 6- Imaging may also help to document uniform distribu- month period, and for the joint to cease being a target tion of the isotope, as loculation due to fibrosis in the joint, there must be no bleeds over a 12-month period joint can hinder this. [59]. In order to prevent progressive joint damage, Inflammatory reactions, including fever and pain, further intervention may be needed. have very occasionally been reported after the proce- Such definitions are appropriate for registries and dure. It is important to flush the needle with sterile clinical trials, but in patients on prophylaxis any joint water for injection before withdrawing the needle from which has more than two bleeds over a 6-month period the joint carefully as radioactive burns may occur along should be considered an ‘at risk joint’ requiring thor- the needle track. Injecting an anti-inflammatory agent ough examination and investigations. This includes such as hydrocortisone acetate into the joint before ensuring an adequate trough level for those receiving removal also effectively flushes isotope out, reducing prophylaxis, and imaging to assess for synovitis and to both the risk of leakage as well as a subsequent inflam- exclude a mechanical cause of bleeding. Ultrasound matory reaction and radiation skin burn. assessment in patients presenting with acute and High dose radiation has been shown to damage chronic pain in joints commonly shows inflammatory articular cartilage in experimental studies, but at the soft tissue change [60]. An approach to the manage- therapeutic doses used in radioactive synovectomy ment of target joints is outlined in Fig. 1. this is unlikely [64]. A recent longitudinal study of 2412 patients who had undergone the procedure for synovitis of varying aetiology (mainly rheumatoid Radioactive synovectomy (synoviorthesis) arthritis) showed no excess incidence of neoplasia Radioactive synovectomy offers a conservative alterna- [65]. There have been no confirmed reports of malig- tive to surgical synovectomy in patients with a target nancy as a consequence of radioactive synovectomy joint associated with synovitis and recurrent bleeding including use in children [66], and the recommended which has proved refractory to intensive treatment radiation dose is below the recommended annual with clotting factor concentrates. It is a particularly exposure limit [67].

Target Joint Identified ˃2 breakthrough bleeds in 6 months

Check adequate trough level and inhibitor status Consider half-life study Commence or intensify prophylaxis if needed Optimise physiotherapy/activity modification Consider Ultrasound/MRI joint imaging

Good response Poor response

MRI Imaging

Haemophilia MDT Clinic assesses need for radioactive synovectomy

Radioactive synovectomy required? No

Revert to standard Yes prophylaxis

Arrange radioactive synovectomy

MDT/Orthopaedic review for further orthopaedic intervention

Good response Review by Haemophilia MDT

Poor response

Fig. 1. Management of Target Joints in PWH – suggested pathway.

A reduction in the bleeding tendency is usually evi- Intra-articular steroid injection dent within 2 to 3 weeks of the procedure. A positive Steroid injections are used in a wide range of inflam- response in terms of reduced bleeding frequency and matory arthritic conditions. There are no randomized reduction in pain has been documented in 75 to 85% or large non-randomized studies to inform an evidence of patients but no significant improvement in range of base in relation to the use of intra-articular steroids in joint motion should be anticipated [62,63]. Estab- haemophilia with associated arthropathy. There are lished joint damage is not a contraindication to some reports of a temporary improvement in symp- radioactive synovectomy where it is equally effective toms in patients with chronic synovitis [72] but there in reducing joint bleed frequency and pain [68]. is no clear role for intra-articular steroids to control Selective angiographic embolization bleeding. There may be a role for intra-articular steroids as a pain control measure prior to surgical inter- Successful control of recurrent bleeding into target vention. joints and improvement of chronic synovitis has been – reported using selective angiographic embolization Intra-articular steroid injections recommendations [69–71]. There are no large studies but the approach 1. No firm recommendation for the use of steroid may be useful in selected patients. injections can be made. Individually selected patients may derive some short-term symptomatic Chronic synovitis and target joints – recommendations benefit (2D). 1. Recognition, imaging and intervention to minimize synovitis in any target joints should be based on an individualized approach which includes opti- Author contributions mization of factor replacement therapy and evalu- JP, AM, DC, SC, PM, NG, PB, SF, PG, JW, JT and PC contributed to the ation for chronic synovitis or potential mechanical literature review and interpretation, and drafting and revision of the cause due to arthropathy (1B). manuscript. All authors had access to the final data, participated in the interpretation, and vouched for the completeness and interpretation of 2. A radioactive synovectomy should always be con- the literature. sidered when there is evidence of chronic synovitis which has not responded to more conservative interventions (1B). Disclosures 3. Selective angiographic embolization may be useful The authors stated that they had no interests which might be perceived as in some patients (2C). posing a conflict or bias.

References Haematology: guideline on the use of pro- 14 Honig GR, Forman EN, Johnston CA, Seeler phylactic factor VIII concentrate in children RA, Abildgaard CF, Schulman I. Administra- 1 Madhok R, Bennett D, Sturrock RD, and adults with severe haemophilia A. Br J tion of single doses of AHF (factor VIII) con- – Forbes CD. Mechanisms of joint damage in Haematol 2010; 149: 498 507. centrates in the treatment of hemophilic – an experimental model of hemophilic 7 Srivastava A, Brewer AK, Mauser-Bunscho- hemarthroses. Pediatrics 1969; 43:26 33. arthritis. Arthritis Rheum 1988; 31: 1148– ten EP et al. Guidelines for the manage- 15 Brown DL, Hardisty RM, Kosoy MH, 55. ment of hemophilia. Haemophilia 2013; Bracken C. Antihaemophilic globulin: – 2 Roosendaal G, Jansen NWD, Schutgens R, 19:e1 47. preparation by an improved cryoprecipita- Lafeber FPJG. Haemophilic arthropathy: 8 Abildgaard CF. Current concepts in the tion method and clinical use. Br Med J – the importance of the earliest haemar- management of hemophilia. Semin Hematol 1967; 2:79 85. – throses and consequences for treatment. 1975; 12: 223 32. 16 Aronstam A, Choudhury D, Wassef M, Haemophilia 2008; 14:4–10. 9 Abildgaard CF, Britton M. Treatment of Turk P, McLellan D. Double-blind con- 3 Wen F-Q, Jabbar AA, Chen Y-X, Kazarian hemophilic patients with factor VIII inhibi- trolled trial of three dosage regimens in T, Patel DA, Valentino LA. c-myc proto- tor. Anna N Y Acad Sci 1975; 240: 400–4. treatment of haemarthroses in haemophilia – oncogene expression in hemophilic synovi- 10 Lusher JM, Shapiro SS, Palascak JE, Rao A. Lancet 1980; 315: 169 71. tis: in vitro studies of the effects of iron AV, Levine PH, Blatt PM. Efﬁcacy of pro- 17 Aronstam A, Wassef M, Hamad Z, Car- and ceramide. Blood 2002; 100: 912–6. thrombin-complex concentrates in tlidge J, McLellan D. A double-blind con- 4 Acharya SS, Kaplan RN, Macdonald D, hemophiliacs with antibodies to factor VIII: trolled trial of two dose levels of factor Fabiyi OT, DiMichele D, Lyden D. Neoan- a multicenter therapeutic trial. N Engl J VIII in the treatment of high risk haemar- giogenesis contributes to the development Med 1980; 303: 421–5. throses in haemophilia A. Clin Lab Haema- of hemophilic synovitis. Blood 2011; 117: 11 Allain JP. Dose requirement for replace- tol 1983; 5: 157–63. 2484–93. ment therapy in hemophilia A. Thromb 18 Mahlangu J, Powell JS, Ragni MV et al. 5 McKernan A, Hanley J, Chowdary P, Clas- Haemost 1979; 42: 825–31. Phase 3 study of recombinant factor VIII sey S, Frostick S. A survey of the muscu- 12 Hermans C, De Moerloose P, Fischer K et al. Fc fusion protein in severe hemophilia A. loskeletal management of haemophilia in Management of acute haemarthrosis in hae- Blood 2014; 123: 317–25. the UK. BSHT/UKHCDO Joint Annual mophilia A without inhibitors: literature 19 Tarantino MD, Collins PW, Hay CR et al. Meeting, Edinburgh. Edinburgh, Surgeon’s review, European survey and recommenda- Clinical evaluation of an advanced category Hall, EH8 9DW, 2014. tions. Haemophilia 2011; 17:383–92. antihaemophilic factor prepared using a 6 Richards M, Williams M, Chalmers E et al. 13 Srivastava A. Dose and response in plasma/albumin-free method: pharmacoki- A United Kingdom Haemophilia Centre haemophilia–optimization of factor replace- netics, efﬁcacy, and safety in previously Doctors’ Organization guideline approved ment therapy. Br J Haematol 2004; 127: treated patients with haemophilia A. Hae- by the British Committee for Standards in 12–25. mophilia 2004; 10: 428–37.

20 Konkle BA, Stasyshyn O, Chowdary P 33 Britain BMJGatRPSoG. British National management of joint bleeds and painful et al. Pegylated, full-length, recombinant Formulary. London and Chicago: BMJ joint arthropathy in patients with factor VIII for prophylactic and on-demand Pub. Group and Pharmaceutical Press, haemophilia? WFH Musculoskeletal Con- treatment of severe hemophilia A. Blood 2013. gress 2015 Abstracts. Haemophilia 2015; 2015; 126: 1078–85. 34 Arachchillage DR, Makris M. Choosing 21: e526–54. 21 Collins P, Chalmers E, Chowdary P et al. and using non-steroidal anti-inflammatory 49 Mayrovitz HN, Delgado M, Smith J. Com- The use of enhanced half-life coagulation drugs in haemophilia. Haemophilia 2016; pression bandaging effects on lower factor concentrates in routine clinical prac- 22: 179–187. extremity peripheral and sub-bandage skin tice: guidance from UKHCDO. Haemophil- 35 Buckwalter JA. Activity vs. rest in the treat- blood perfusion. Ostomy Wound Manage ia 2016; 22: 487–98. ment of bone, soft tissue and joint injuries. 1998; 44:56–60, 2, 4 passim. 22 Aledort LM, Haschmeyer RH, Pettersson Iowa Orthop J 1995; 15:29–42. 50 Tsang KKW, Hertel J, Denegar CR. Vol- H. A longitudinal study of orthopaedic out- 36 Broughton G II, Janis JE, Attinger CE. The ume decreases after elevation and intermit- comes for severe factor-VIII-deficient basic science of wound healing. Plastic tent compression of postacute ankle sprains haemophiliacs. The Orthopaedic Outcome Reconstruct Surg 2006; 117: 12s–34s. are negated by gravity-dependent position- Study Group. J Int Med 1994; 236: 391–9. 37 Bleakley CM, Glasgow PD, Phillips N et al. ing. J Athl Train 2003; 38: 320–4. 23 Rodriguez-Merchan EC, Jimenez-Yuste V, Management of Acute Soft Tissue Injury 51 McNair PJ, Marshall RN, Maguire K, Aznar JA et al. Joint protection in Using Protection Rest Ice Compression and Brown C. Knee joint effusion and proprio- haemophilia. Haemophilia 2011; 17(Suppl. Elevation. Recommendations from the: ception. Arch Phys Med Rehabil 1995; 76: 2): 1–23. Association of Chartered Physiotherapists 566–8. 24 Santagostino E, Mancuso M, Rocino A, in Sports and Exercise Medicine (ACPSM), 52 Guido J Jr, Voight ML, Blackburn TA, Mancuso G, Scaraggi F, Mannucci P. A 2011. Kidder JD, Nord S. The effects of chronic prospective randomized trial of high and 38 Hooiveld MJ, Roosendaal G, Jacobs KM effusion on knee joint proprioception: a standard dosages of recombinant factor et al. Initiation of degenerative joint dam- case study. J Orthop Sports Phys Ther VIIa for treatment of hemarthroses in age by experimental bleeding combined 1997; 25: 208–12. hemophiliacs with inhibitors. J Thromb with loading of the joint: a possible mecha- 53 Sterling M, Jull G, Wright A. The effect of Haemost 2006; 4: 367–71. nism of hemophilic arthropathy. Arthritis musculoskeletal pain on motor activity and 25 Kavakli K, Makris M, Zulfikar B, Erhardt- Rheum 2004; 50: 2024–31. control. J Pain 2001; 2: 135–45. sen E, Abrams ZS. Home treatment of hae- 39 Hortobagyi T, Dempsey L, Fraser D et al. 54 Riemann BL, Lephart SM. The sensorimo- marthroses using a single dose regimen of Changes in muscle strength, muscle fibre tor system, part II: the role of propriocep- recombinant activated factor VII in patients size and myofibrillar gene expression after tion in motor control and functional joint with haemophilia and inhibitors. A multi- immobilization and retraining in humans. J stability. J Athl Train 2002; 37:80–4. centre, randomised, double-blind, cross- Physiol 2000; 524(Pt 1): 293–304. 55 Knoop J, Steultjens MP, van der Leeden M over trial. Thromb Haemost 2006; 95: 40 Berg HE, Dudley GA, Haggmark T, Ohlsen et al. Proprioception in knee osteoarthritis: 600–5. H, Tesch PA. Effects of lower limb unload- a narrative review. Osteoarthritis Cartilage 26 Gringeri A, Fischer K, Karafoulidou A, ing on skeletal muscle mass and function in 2011; 19: 381–8. Klamroth R, Lopez-Fern andez M, Mancuso humans. J Appl Physiol 1991; 70: 1882–5. 56 Hurley MV. The effects of joint damage on E. Sequential combined bypassing therapy 41 Rodriguez-Merchan EC, Hedner U, Heijnen muscle function, proprioception and reha- is safe and effective in the treatment of L et al. Prevention of haemophilic bilitation. Man Ther 1997; 2:11–7. unresponsive bleeding in adults and chil- arthropathy during childhood. May com- 57 van Vliet PM, Heneghan NR. Motor con- dren with haemophilia and inhibitors. Hae- mon orthopaedic management be extrapo- trol and the management of musculoskele- mophilia 2011; 17: 630–5. lated from patients without inhibitors to tal dysfunction. Man Ther 2006; 11: 208– 27 Valentino LA, Holme PA. Should anti-inhi- patients with inhibitors? Haemophilia 13. bitor coagulant complex and tranexamic 2008; 14(Suppl 6): 68–81. 58 Blanchette VS, Key NS, Ljung LR, Manco- acid be used concomitantly? Haemophilia 42 Algafly AA, George KP. The effect of Johnson MJ, van den Berg HM, Srivastava 2015; 21: 709–14. cryotherapy on nerve conduction velocity, A. Definitions in hemophilia: communica- 28 Holmstrom€ M, Tran H, Holme P. pain threshold and pain tolerance. Br J tion from the SSC of the ISTH. J Thromb Combined treatment with APCC (FEIBAâ) Sports Med 2007; 41: 365–9; discussion 9. Haemost 2014; 12: 1935–9. and tranexamic acid in patients with 43 Watanabe S, Kakigi R, Hoshiyama M, 59 Donadel-Claeyssens S. Current co-ordi- haemophilia A with inhibitors and in Kitamura Y, Koyama S, Shimojo M. Effects nated activities of the PEDNET (European patients with acquired haemophilia A–a of noxious cooling of the skin on pain per- Paediatric Network for Haemophilia Man- two-centre experience. Haemophilia 2012; ception in man. J Neurol Sci 1996; 135: agement). Haemophilia 2006; 12: 124–7. 18: 544–9. 68–73. 60 Kidder W, Nguyen S, Larios J, Bergstrom 29 Manners PJ, Price P, Buurman D, Lewin B, 44 Ernst E, Fialka V. Ice freezes pain? A J, Ceponis A, von Drygalski A. Point-of- Smith B, Cole CH. Joint aspiration for review of the clinical effectiveness of anal- care musculoskeletal ultrasound is critical acute hemarthrosis in children receiving gesic cold therapy. J Pain Symp Manage for the diagnosis of hemarthroses, inflam- factor VIII prophylaxis for severe hemophi- 1994; 9:56–9. mation and soft tissue abnormalities in lia: 11-year safety data. J Rheumatol 2015; 45 Forsyth AL, Rivard GE, Valentino LA et al. adult patients with painful haemophilic 42: 885–90. Consequences of intra-articular bleeding in arthropathy. Haemophilia 2015; 21: 530– 30 Eyster ME, Asaad SM, Gold BD, Cohn SE, haemophilia: science to clinical practice 7. Goedert JJ. Upper gastrointestinal bleeding and beyond. Haemophilia 2012; 18(Suppl. 61 Pasta G, Forsyth A, Merchan CR et al. in haemophiliacs: incidence and relation to 4): 112–9. Orthopaedic management of orthopaedic use of non-steroidal anti-inflammatory 46 d’Young AI. Domiciliary application of arthropathy of the ankle. Haemophilia drugs. Haemophilia 2007; 13: 279–86. CryoCuff in severe haemophilia: qualitative 2008; 14: 170–6. 31 Rattray B, Nugent DJ, Young G. Celecoxib questionnaire and clinical audit. Hae- 62 Dunn AL, Busch MT, Wyly JB, Abshire in the treatment of haemophilic synovitis, mophilia 2008; 14: 823–7. TC. Radionuclide synovectomy for hemo- target joints, and pain in adults and chil- 47 Rajamanickam M, Michael R, Sampath V, philic arthropathy: a comprehensive review dren with haemophilia. Haemophilia 2006; John JA, Viswabandya A, Srivastava A. of safety and efficacy and recommendation 12: 514–7. Should ice be used in the treatment of acute for a standardized treatment protocol. 32 Tsoukas C, Eyster ME, Shingo S et al. haemarthrosis in haemophilia? Haemophil- Thromb Haemost 2002; 87: 383–93. Evaluation of the efficacy and safety of ia 2013; 19: e267–8. 63 Rodriguez-Merchan E, Quintana M, De La etoricoxib in the treatment of hemophilic 48 Sayers FCJ, Benson G. Is there a role for Corte-Rodriguez H, Coya J. Radioactive arthropathy. Blood 2006; 107:1785–90. domiciliary cold therapy in the synoviorthesis for the treatment of

haemophilic synovitis. Haemophilia 2007; 67 Rodriguez-Merchan EC, Valentino LA. therapy of hemarthrosis using endovascular 13:32–7. Safety of radiation exposure after embolization of knee and elbow arteries. 64 Silva M, Luck JV, Leissinger C. Opinions radiosynovectomy in paediatric patients Cardiovasc Intervent Radiol 2013; 36: on radiosynovectomy for chronic haemo- with haemophilia. Haemophilia 2015; 21: 964–9. philic synovitis: point/counterpoint. Hae- 411–8. 71 Klamroth R, Gottstein S, Essers E, Land- mophilia 2012; 18: 836–42. 68 Rodriguez-Merchan EC, De la Corte-Rodri- graf H, Wilaschek M, Oldenburg J. Suc- 65 Infante-Rivard C, Rivard G-E, Derome F guez H, Jimenez-Yuste V. Radiosynovec- cessful angiographic embolization of et al. A retrospective cohort study of cancer tomy in haemophilia: long-term results of recurrent elbow and knee joint bleeds in incidence among patients treated with 500 procedures performed in a 38-year per- seven patients with severe haemophilia. radiosynoviorthesis. Haemophilia 2012; 18: iod. Thromb Res 2014; 134: 985–90. Haemophilia 2009; 15: 247–52. 805–9. 69 Obaji S, Jones C, Yates A et al. Selective 72 Fernandez-Palazzi F. Treatment of acute 66 De Podesta Haje D, Ono F, De Oliveira angiographic embolization for recurrent and chronic synovitis by non-surgical GB et al. Orthopaedic evaluation in chil- elbow and knee haemarthroses in haemo- means. Haemophilia 1998; 4: 518–23. dren with severe haemophilia A or B sub- philia: a retrospective case series. Hae- 73 Britton M, Harrison J, Abildgaard CF. mitted to primary prophylaxis therapy in a mophilia 2015; 21: e226–8. Early treatment of hemophilic hemarthroses coagulopathy treatment centre. Haemophil- 70 Galli E, Baques A, Moretti N, Candela M, with minimal dose of new factor 8 concen- ia 2011; 17: 228–32. Caviglia H. Hemophilic chronic synovitis: trate. J Pediatr 1974; 85: 245–7.