Objective: To determine causes of death and mortality rates in patients with hemophilia over a period of 20 years, to assess changes in mortality, and to distinguish between hemophilia-related death and recent death induced by viral infections.
Design: Cohort study of 919 patients followed from January 1986 to June 1992. Results were compared with outcomes of previous follow-up from 1973 to 1986.
Setting: Consecutive national questionnaire surveys on hemophilia, using patient registries of the Netherlands Hemophilia Society and Dutch hemophilia centers.
Patients: 919 males with hemophilia A or B who participated in a national questionnaire survey on hemophilia in 1985. Median duration of follow-up was 6.4 years, which yielded 5753 person-years of follow-up. The mean age at study entry was 30 years (range, 1 to 85 years).
Measurements: Standardized mortality ratios, causes of death, median life expectancy, age-adjusted relative risks associated with the type or severity of hemophilia, presence of inhibitors, prophylaxis, and human immunodeficiency virus infection.
Results: 45 patients (5%) died between January 1986 and June 1992; 22.6 patients had been expected to die. Thus, the overall standardized mortality ratio was 2.0. The overall median life expectancy was 66 years for the cohort studied from 1973 to 1986 and 68 years for the cohort studied from 1986 to 1992. When deaths related to viral infection were excluded, the life expectancy almost equaled that of the general male population. Between 1986 and 1992, 1 patient died of ischemic heart disease compared with the 5.2 who were expected to die of this disease. Infection with HIV was the strongest independent predictor of death (relative risk, 27.5 [95% CI, 5.7 to 132.8]). After adjustment for HIV infection, no other hemophilia-related risk factors were associated with the risk for death.
Conclusions: The acquired immunodeficiency syndrome and hepatitis strongly influence mortality in patients with hemophilia. In the absence of viral infections, the life expectancy of patients with hemophilia would almost equal that of the general male population.