An actuarial GPS for hemophilic longevity

HIV as a transfusion-transmitted pathogen drastically reduced the lifespan of people with hemophilia beginning in the 1980s. The question of what would have been the expected lifespan in this chronic disease, had HIV not happened, has been the subject of previous epidemiologic reports.

Data from the large, well-characterized United Kingdom cohort of hemophilia patients described by Darby and colleagues in this issue of Blood provides the most complete picture to date of how long HIV-uninfected individuals live and what disease or injury states are most frequently associated with their demise (see figure). The vital status of 6018 HIV-uninfected hemophilia A and B individuals of all clinical severities was captured comprehensively, and causes of death for virtually all decedents were ascertained. Without the confounding of HIV for more than 2 decades (1977-1999), these findings provide confirmation of previously defined mortality risk factors for hemophilia, and also elucidate how the impact of certain of these risks have evolved over time.

For example, the investigators confirm hepatitis as a mortality risk factor, but its weighted impact is not as striking as previous reports have implied in patients with severe hemophilia A or B. Conversely, a protective effect of hemophilia against death from coronary heart disease/myocardial infarction is demonstrated, confirming previous observations.¹  Not surprisingly, intracranial hemorrhage (ICH) persists as a hemorrhagic mortality risk factor, particularly in children younger than 4 years of age. Over the 23 years assessed, however, the age of those infants dying from ICH became progressively younger, particularly among those with severe disease. As the authors suggest, this effect may result from the introduction of primary prophylaxis in the United Kingdom in the early 1990s, which typically begins at approximately 1 year of age. Whether or not this represents the entire explanation, it illustrates how a comprehensively maintained longitudinal mortality dataset may not only inform the impact of previous interventions, but also suggest future therapeutic strategies. If it is now mostly the very young who are still highly susceptible to mortal ICH, we may need to explore strategies of earlier recognition of ICH in boys without prior family history of hemophilia or even, perhaps, re-examine management of parturition in known hemophilic carriers of male fetuses.

In an analogous way, citing previous examinations of this identical United Kingdom dataset for trends in the effect of inhibitors on overall mortality,²  the authors have made the following case: during the last third of the 23-year observation, the combination of more universal application of immune tolerance induction techniques coupled with more aggressive use of established Factor VIII inhibitor bypassing activity (FEIBA) and new (recombinant factor VIIa) bypassing therapies has reduced inhibitor-related mortality to that of severe hemophilia A/B without inhibitor (an absolute reduction in mortality in the inhibitor cohort from 8.6 per 1,000 years in 1977-1984 to 5.6 per 1,000 years from 1993-1999). Although no cohort of any size can substitute for prospective clinical trials, observations such as these can engender critical hypothesis testing of important clinical interventions.

Accordingly, these cogent data illustrate how a carefully studied cohort followed judiciously for longevity and cause of death can provide insight into both the natural history of chronic disease and the impact of therapy. Target subgroups may be identified for particular changes in therapeutic approach. This may help ensure that every stone has been overturned before we accept a plateau in survival. For hemophilia, the United Kingdom cohort continues to offer such illumination, and helps to justify efforts to expand and refine similar national datasets such as the Universal Data Collection (UCD) in the United States.