The combination of the complexity of the coagulopathy in haemophilia with the relative low frequency of occurrence of the condition poses a formidable challenge to respond to scientific questions. Consequently, the gold standard of care has arisen from tradition and become, by virtue of habit, into paradigms. Under these constrains, when the paradigm is challenged by fragments of data, in the absence of a randomized controlled trial, a negative emotional response is typically generated that may hinder clinical progress. In this study, we will address four subjects where fragmented evidence from basic science studies or advances in achieving reliable coagulation allow challenge of the paradigm. Hemarthrosis is the most common clinical manifestation of haemophilia, and its repetitive occurrence leads to irreversible joint damage. It is generally accepted that early factor replacement therapy should be started when initial symptoms of joint leakage are detected, to avoid evident swelling of the joint and synovitis and to favour early and complete recovery. If infusion of factor is attained early following the initiation of the bleed, the perceptible clinical relevance of the hemarthrosis is diminished, and rehabilitation of the joint can start early and clinical recovery is attained [1]. However, experimental evidence suggests that there is more than that meets the eye. Exposure of cartilage tissue in vitro to whole blood leads to disturbance of cartilage matrix turnover, diminishing the synthesis of aggrecan, which in turn results in a decrease of the glycosaminoglycan content of the cartilage matrix [2]. Additionally, induction of hemarthrosis in haemophilic mice produced an increase in several pro-inflammatory cytokines, establishing the existence of a synovial inflammatory component in haemophilic synovitis [3]. The testing of these findings in larger animal models highlights several dimensions of the question, which are probably related to the long-term clinical outcome of joint deterioration in humans. Some of these are: the velocity of clearance of blood from the joint [4], the length of time that synovial activation remains and resulting inhibition of the cartilage matrix turnover [3], the tolerance of hyaline cartilage to the biochemical aggression resulting from the exposure to blood, pro-inflammatory cytokines and the resulting deleterious enzymes [5] and the reversibility of histological injury [6]. The experimental design used to characterize the biochemical response to repetitive bleeding mimics the circumstances of limited or no access to factor concentrate. We have believed for years that lowering the bleeding magnitude and frequency to marginal or imperceptible levels would be enough to prevent arthropathy. However, Manco-Johnson and colleagues demonstrated that even subclinical bleeding in patients with high compliance prophylaxis led to joint deterioration [7]. Is it time to redefine the clinical determinants of joint aspiration after acute bleeds? Arthrodesis of the ankle has long been the standard of care for painful grade IV haemophilic ankle arthropathy [8]. Tsailas and Wiedel recently reviewed the results of 20 ankle fusions in 13 patients, eight of which had a subtalar fusion as well. With a mean operation age of 39 years and a mean follow-up of 9 years, there was no recurrent bleeding or deep infection. The procedure was successful in all but one patient that required a revision for tibiotalar non-union. There was a high degree of satisfaction for the patients with the fusion achieved primarily with the use of two cross screw fixation [9]. However, the pressure resulting from the patient’s concern with loss of ankle motion related to joint fusion, and recent improvement in the design of ankle implants has brought to centre stage the need to re-consider the standard of care. With current designs, survivorship data compare favourably with first and second generation implants that were abandoned by most surgeons due to unacceptably high complication and failure rates [10]. However, they do not yet meet the standards of success of hip and knee arthroplasties [11]. A survivorship analysis on the use of total ankle replacements in 257 non-haemophilic patients in the Norwegian population over a 12-year period produced an overall 5-year and 10-year survival of 89% and 76%, respectively [11]. A similar study from the New Zealand National Registry of 202 total ankle replacements in 183 non-haemophiliac patients found an overall cumulative 5-year failure-free rate of 86%. An unfavourable patient score at 6 months after the initial procedure turned out to be a good predictor of subsequent failure. The cumulative 5-year failure-free rate was 65% at 5 years for patients with an unfavourable score, and 95% for those who had a favourable patient score [12]. Increasingly, case reports and case series reporting on ankle arthroplasty for the treatment of haemophilic arthropathy have become available [13-16]. They report a high satisfaction rate from the patients in terms of pain relief and return of range of motion and a low complication rate. An additional angle to this paradigm challenge arises from the possibility of converting arthrodesed ankles to ankle arthroplasties. In a recent study, in non-haemophiliacs, 29 ankles in 27 patients with painful fused ankles were converted to a total ankle replacement. Their American Orthopedic Foot and Ankle Society hindfoot Score increased from 34.1 preoperatively to 70.6 at the time of the latest follow-up. Twenty-four patients (82.7%) were satisfied with the results. While five ankles were completely pain-free, twenty-one ankles were moderately painful, and three remained painful. The average clinically measured range of motion of 24.3 amounted to 55.1% of that of the contralateral, unaffected ankles [17]. Is it time to re-consider ankle fusion as the treatment of choice for advanced arthropathy of the ankle? Muscle haematomas are the second most common manifestation of haemophilia. While most bleeds do not represent a therapeutic challenge, those located within selected muscle groups can produce significant injury. Haemorrhages located within the calf and anterior portion of the forearm represent the highest risk for the development of increased compartment pressures and permanent muscle injury, however, any muscle group with a well-defined fascia can develop a compartment syndrome [18]. The treatment of choice for patients who are not haemophiliacs when faced with a compartment syndrome is, unambiguously, a fasciotomy. It is universally considered an emergency, and trauma teams are trained to perform it, even when in doubt of the diagnosis, to prevent the establishment of motor and sensory loss, contracture and severe extremity impairment. Due to the fear of uncontrolled bleeding after a faciotomy which may risk loss of life, the role of this procedure in patients with haemophilia has been downplayed and care focuses on haemostatic manoeuvres [1]. While appropriate haemostasis will stop the bleeding within the muscular compartment, the latency between the correction of the coagulopathy and the decrease in compartment pressures that will allow perfusion of the muscle may be long enough to lead to muscular death by isquemia and future contracture [19]. To challenge the paradigm, Caviglia et al. have developed an algorithm that requires optimal correction of the coagulopathy and compartment pressure monitoring such that pressures above the 45 mmHg threshold would require mandatory fasciotomy [18]. Implementation of this algorithm may help prevent the severe neuromuscular lesions leading to contractures and loss of function that are often seen in haemophilia centres around the world [20]. Due to the fear of articular and muscular haemorrhages, there is a strong tendency among parents of boys with haemophilia to prevent them from engaging in exercise programmes. Visible signs of sedentarism are muscle atrophy, instability and restriction of motion [21]. These are more present in adults than in the younger patients [22]. First subclinical symptoms like tender ligaments are found even in the clinically healthy young group [23]. This leads to a lack of physical activity and exercise that results in a poor physical condition with diminished muscle strength, aerobic/anaerobic power, proprioception and flexibility [24]. Interestingly, increasing amount of evidence indicates that biological changes induced by physical activity produce a transient hypercoagulability state. This is mostly due to increased thrombin generation, platelet hyperactivity and increased activity of several coagulation factors, especially factor VIII and von Willebrand′s [25, 26]. These findings strengthen the foundations for the recommendation of exercise in persons with haemophilia. Regular and controlled exercise that significantly improves the physical condition are: training with light [21], medium or in special cases also heavy weights [27], dynamic [28], isokinetic [29] and isometric [30] or electrically stimulated strength exercise [31]. Proprioception is the second important exercise field, performed alone or in combination with strength training [27]. The fitness of 255 children and young adults (8–25 years) was tested for proprioception, strength, flexibility, endurance and body fat. In a comparison between the very active and the non-active group, proprioception and the total fitness were significantly better in the active group [32]. Thus, exercise should be carried out regularly, 30–120 min for 2–4 times a week, as recommended in the literature. All these interventions from 2 weeks up to 2 years showed first and last positive results. As a direct effect of exercise, we may expect a lower frequency of hemarthrosis [21], an improvement of postural balance [33] and of overall proprioception [27], improvement of ROM [34], force and isometric strength [27]. All authors agree on an improvement of quality of life. Although there are many studies with a wide range of measurement techniques, and only few with control groups and most of them with a few number of patients, the results are congruent and therefore reliable. Exercise should start as early as possible to prevent symptomatic structural, because in motion analysis, we see early functional changes in gait and squat, and before structural changes are manifested [35].