In 1992, Professor Inga Marie Nilsson et al.  published the results of long-term factor prophylaxis in a cohort of 60 patients with severe haemophilia A followed in the haemophilia treatment centre in Malmö, Sweden. This pivotal communication, together with a follow-up report in 1997, established the significant musculoskeletal benefits of factor prophylaxis if begun at an early age in boys with severe haemophilia with the goal of maintaining a circulating factor level consistently above 1% of normal . The rationale for the Swedish approach was based on the observation, reported by Ahlberg  in 1965, that patients with a moderate form of haemophilia, i.e. those with a circulating factor level of 1–5% of normal, rarely develop chronic arthropathy. The Swedish physicians reasoned that, if the plasma concentration of factor VIII/IX could be maintained at a level of at least 1% of normal by regular replacement therapy (prophylaxis), it might be possible to convert the severe haemophilic phenotype to a milder one. In Sweden, prophylactic treatment on a small number of boys with severe haemophilia A was begun in 1958 and of those with severe haemophilia B in 1972. The results in compliant patients were impressive and led to the adoption of primary prophylaxis as the ideal care for young boys with severe haemophilia by a number of organizations and national haemophilia foundations [4,5].
Supported by the results of a number of observational studies and the recent reports of two prospective clinical trials of factor prophylaxis in young boys with severe haemophilia A [6–10], the benefits of primary prophylaxis over on-demand therapy are now beyond dispute. However, many challenges regarding optimizing factor prophylaxis and management in the haemophilia population remain:
- 1For the first decade of life: when should primary prophylaxis be ideally started in young boys with severe haemophilia and how to deal with the problems of frequent venous access?
- 2For the second decade: how to maintain prophylaxis during the transition towards independence?
- 3For the third decade of life: when, if ever and in whom, should prophylaxis be discontinued?
Independent of age, it is a challenge to determine the role of factor prophylaxis in special patient groups, e.g. persons with haemophilia and circulating inhibitors. The clinical challenges of providing prophylaxis to patients with severe haemophilia are discussed here.
Challenges during the first decade of life: initiating early prophylactic treatment
Several studies have demonstrated that prophylaxis should be started early, before the onset of arthropathy (primary prophylaxis), to preserve joint function in adulthood. Data from Sweden and the Netherlands provide support for starting prophylaxis early in life (generally by age 2–3 years) and after no more than one to two haemarthroses [11,12]. Some groups advocate starting prophylaxis before the first joint haemorrhage ; the disadvantage of this approach is that a subset of boys who may have a less severe phenotype will be started on full-dose prophylaxis with the attendant difficulties of securing reliable venous access. As the time of the first joint bleed varies considerably in boys with severe haemophilia , and because it is not possible to predict with accuracy when the first joint bleed will occur, it appears reasonable to allow the individual bleeding phenotype of boys with severe haemophilia to dictate the timing of starting factor prophylaxis. The one caveat to this recommendation is the recent report that early prophylaxis may be protective against inhibitor formation in boys with severe haemophilia . If confirmed, this observation may result in starting prophylaxis regardless of the age of onset of joint bleeding. An additional benefit, yet unproven, from such an approach might be a decreased risk of intracranial haemorrhage (ICH) in young boys with severe haemophilia . The choice of the prophylactic regimen is largely influenced by cost and the issue of venous access. Again, a Swedish report provided important data showing that the frequency of infusions at start of prophylaxis was less important in determining outcome than the age prophylaxis was started . Consequently, step-up regimes starting with once weekly infusions are used more often and may help avoid some of the problems with venous access [9,17]. Another advantage of such escalating regimens is that dose and frequency of infusions are adjusted according to individual bleeding patterns.
In addition, the latest Canadian and American studies have introduced enhanced episodic therapy for breakthrough haemarthrosis: breakthrough joint haemorrhages were treated with a minimum of three infusions of factor VIII (FVIII) spread over a 4-day period [9,10]. The role of enhanced episodic therapy during prophylactic treatment, however, is difficult to establish. Preventing haemorrhage is still the primary treatment goal.
Overall, the principles of optimal treatment in the first decade of life have been well established: prophylaxis should be started early, preferably not later than after the second joint haemorrhage, and the dose and frequency should be rapidly increased to prevent haemarthrosis but allow full activity of the child.
Challenges in prophylaxis during the second decade of life: independence and adherence
In the second decade, most patients on prophylaxis learn self-infusion and thus obtain more independence. If the patient is increasingly managing his own treatment, it is important to confirm that his knowledge of the disease is adequate. Patient education is an important aspect in the preparation for the transition to the adult haematology department. A recent survey among 307 members of the UK Haemophilia Society showed that there was a need for information on medical management, life-style management, psychosocial aspects as well as social and employment issues . One of the striking findings was that 57% of patients wanted to be informed about current research projects. Unfortunately, the age of the respondents was 18–86 years, and responses were not specified according to age. One of the very effective strategies in educating teenagers is peer contact. Many physicians have reported favourable effects of camps and peer groups for both education and self-esteem. However, formal research in this area is lacking.
The second decade of life is associated with major physical and psychological changes. In chronic diseases, denial and reduced adherence are often observed.
Adherence can be defined as ‘the extent to which a person’s behaviour coincides with medical or health advice’ . Adherence in adolescence has been studied in several chronic conditions; in a large Finnish study of 1061 patients with chronic conditions ranging from asthma to insulin-dependent diabetes, only one-fifth of the patients studied felt they had complied fully with health regimens, whereas 60% reported satisfactory adherence and 17% reported poor adherence . Many factors interfering with adherence were identified: developmental issues, medical and demographic factors, cognitive-emotional and motivational factors, family support, peer support and the quality of interaction with health care providers.
In the case of haemophilia, adherence is very high in childhood, but decreases with increasing age . Hacker et al.  found an adherence rate of 59% in patients between 1 and 18 years old. A more recent study by Lindvall et al.  found that the average age at which patients take primary responsibility for their disease and treatment is 14.1 years. Although 99% of the 108 respondents were aware of the reason for prophylactic treatment, only 40% (between 13 and 25 years old) reported that they might not follow the prescribed prophylactic treatment. Adherence is clearly reduced, but its consequences are currently unknown and more work in this field is needed.
The challenge is to promote adherence and continue to prevent bleeding during this important period of rapid growth. Management options may include education and support from parents, nurses and physicians as well as peer contacts.
Challenges in prophylaxis during the third decade of life: tailoring according to bleeding pattern
In Northern Europe, the third decade of life often represents a change in lifestyle. Patients may get a job and physical activity may be more confined to well-defined periods during the week. In the Netherlands, where primary prophylactic treatment is standard for all patients with severe haemophilia, about two thirds of patients experiment with discontinuing their long-term prophylaxis . In their early twenties (median age 21.4 years), 35% of a cohort of 80 Dutch and Danish patients switched to on-demand treatment for more than 1 year . Having received early prophylaxis, these patients only experienced a median of three joint bleeds per year without prophylaxis. Median clinical scores were similar in patients who discontinued prophylaxis and those who continued, as were median Pettersson scores (median 13 points). In the subgroup of 35 patients who where protected by prophylaxis during the period of rapid growth (started prophylaxis before age of 6 years and continued for at least 10 years), outcome according to treatment discontinuation vs. continuous prophylaxis was also similar: joint bleeding frequency in the last year was 0.0 in patients who had stopped prophylaxis permanently and 1.2 in patients still on prophylaxis; median orthopaedic joint scores were similar at 2.0 points and 3.0 points, respectively, as were median Pettersson scores at 7.0 points in patients who had discontinued prophylaxis vs. 7.5 points in patients who continued prophylaxis. However, longer evaluation is needed to assess the safety of discontinuing prophylaxis in these young adults.
It is important to note that patients who successfully discontinue prophylaxis are those with a milder bleeding pattern . Currently, these patients may be best defined by a later onset of joint bleeding and a lower treatment requirement while on prophylaxis. Tapering or discontinuing prophylaxis while maintaining a low bleeding frequency may be possible only for patients with milder bleeding patterns. The challenge is to optimize efficiency by individualizing prophylactic dose and frequency according to lifestyle and bleeding pattern, as well as to identify the candidates for a less intensive regimen.
Secondary prophylaxis in patients with inhibitors
The regular, scheduled administration of bypassing agents is increasingly being used for patients with inhibitors to FVIII because of the high risk for joint disease and severe haemorrhage. Up to 30% of patients with severe haemophilia A develop inhibitors , and their need for secondary prophylaxis may be even more compelling than that of patients without inhibitors. Bleeding associated with high-titre, high-responding inhibitors [>5 Bethesda Units (BU)]  is often more difficult to control once it starts because the haemostatic effect of bypassing therapy is unpredictable [27,28]. Furthermore, no laboratory assays are available that correlate with dosing or efficacy . As a result, compared with patients without inhibitors, individuals with inhibitors are at increased risk for severe joint disease and significant musculoskeletal limitations [30,31]; this may develop after only a few bleeding episodes, most likely because of protracted haemarthrosis. Moreover, surgical interventions to address joint disease are less likely to be utilized in inhibitor patients because of the greater potential for bleeding complications . Consequently, inhibitor patients often experience significantly impaired mobility and a lower quality of life (QoL) .
Prophylaxis is considered optimal care for individuals with severe haemophilia A without inhibitors, and may be especially beneficial for patients with FVIII inhibitors because they are at increased risk for serious haemorrhage and arthropathy. Although the largest experience with bypassing agent prophylaxis is in the perioperative setting, bypassing therapy, in particular activated prothrombin complex concentrates, e.g. FEIBA, are increasingly used to prevent bleeding during immune tolerance induction (ITI) therapy and in patients who have failed ITI or who are not candidates for or who refuse ITI .
As is true for FEIBA, the largest experience with the prophylactic use of NovoSeven is perioperatively, where its efficacy is at least 80% . Until recently, the literature contained only two reports describing the successful use of recombinant FVIIa (rFVIIa) prophylaxis in three inhibitor patients [36,37], one of whom was undergoing concomitant ITI .
In September 2007, Konkle et al.  published the results of the first clinical trial of the use of rFVIIa for short-term prophylaxis. Thirty-eight patients enrolled in the study. Of these, 22 patients satisfied the entry requirement for high baseline bleeding frequency (mean ≥ 4 bleeding episodes per month for 3 months) and were randomized to receive rFVIIa at a daily dose of 90 or 270 μg kg−1 for 3 months followed by a 3-month postprophylaxis period. Bleeding frequency was reduced by 45% during prophylaxis with the daily 90 μg kg−1 dose and by 59% with the daily 270 μg kg−1 dose (P < 0.001; no significant difference between the doses), and the majority of the reduction was sustained during the postprophylaxis period. The prophylactic effect was most pronounced for spontaneous haemarthrosis, and patients reported significantly fewer hospital admissions and days absent from school or work during prophylaxis as compared with the preprophylaxis period. The findings suggest that rFVIIa may offer an alternative to FEIBA for prophylaxis in inhibitor patients.
The paucity of controlled trial data to assess the efficacy and safety of bypassing agent prophylaxis and determine optimal dosing makes it difficult to draw solid conclusions about the benefits.
Possible uses of prophylaxis in inhibitor patients include initiation of primary prophylaxis or those who do not have arthropathy. No clinical trial data exist in this population. Prophylaxis with rFVIIa may be implemented to prevent bleeding, while the start of ITI is deferred until the inhibitor has declined below 10 BU mL−1  or with rFVIIa or FEIBA to reduce bleeding during immune tolerance induction [40,41] may be useful. Because of the difficulty of controlling bleeding episodes and the heightened risk for severe joint disease , many inhibitor patients begin to exhibit some degree of arthropathy very soon after the development of the inhibitor. A clinical trial of early prophylaxis in combination with ITI therapy after the diagnosis of a high-titre inhibitor may be warranted. Once joint bleeding has started, prophylaxis should not be expected to halt the progression of existing joint disease but may reduce joint bleeding frequency and improve QoL . Prophylaxis following ICH is frequently prescribed for patients with inhibitors. ICH has the highest mortality of any type of bleeding event in patients with haemophilia, and ICH survivors often experience serious long-term sequelae, including seizures and neurological impairment . Recurrent central nervous system (CNS) haemorrhage is common, with Antunes et al.  reporting 10 episodes of recurrent ICH in six haemophilia patients. The mean time between ICH and a subsequent recurrence of CNS haemorrhage was 1 year (range, 0.2–2.6 years), with 80% occurring within 1 year. In patients without inhibitors, prophylaxis has been shown to reduce recurrent CNS bleeding significantly .
In conclusion, long-term primary prophylaxis for patients with severe haemophilia is challenging; in patients without inhibitors, each decade of life poses its typical management issues. Initiating regular prophylaxis before the development of arthropathy and choosing a regimen while considering bleeding pattern and venous access are the first challenge. The second decade is dominated by issues of the patient taking over treatment from his parents, and the challenge of maintaining adherence to prophylaxis. In the third decade, treatment may be adapted to activities and bleeding pattern, including discontinuation in some patients with a milder bleeding phenotype.
In patients with inhibitors, there is a growing awareness of the need to prevent haemorrhages concomitant with immune tolerance therapy. Increasing numbers of reports on the use of prophylaxis with both FEIBA and rFVIIa suggest that it is possible to reduce the number of bleeds and improve QoL, both in children and in older patients.