When corrected for age, this difference was borderline significant (P-value 0.05). The cumulative incidences of malignancies were similar in both groups (9% vs. 10%). Diabetes mellitus, on the other hand, this website occurred significantly more often in HIV-positive than in HIV-negative patients (12% vs. 7%, adjusted

P-value 0.006). All but one of the HIV-positive patients were on HAART when their diabetes was diagnosed. The prevalence of chronic hepatitis C infection was not associated with HIV status. Body mass indexes (BMI) could be calculated for 42 HIV-positive (72%) and 134 HIV-negative haemophilia patients (88%). Mean BMI was significantly lower in the HIV-positive patients (22.1 vs. 25.7 kg m−2, adjusted P-value < 0.001), and the prevalences of overweight (BMI 25.1–30.0 kg m−2) and obesity (BMI >30.0 kg m−2) were also much lower in these patients (10% and 2% vs. 45% and 10% respectively). Thirty-one HIV-positive patients (52%) were deceased at the end of follow-up. Causes of death are shown in Table 3. Death was reported to be solely AIDS related in 19 patients (61%) and caused by a combination Raf inhibitor of HIV and hepatitis C in three patients (10%). Mean age at death was 36.9 years (range: 14–65 years). All but two AIDS-related deaths occurred in patients who were not on HAART. Only the two lymphoma patients were on HAART at time of diagnosis, but the second patient had

started this treatment only a few months earlier. In one other patient on HAART, death was reported to be caused by a combination of HIV and hepatitis C. Median interval between HIV seroconversion and death was 11 years (range: 4–26 years). No fatal non-virus related malignancies occurred in our cohort, nor were there any fatal ischaemic cardiovascular

events. Interestingly, seven of nine HIV-infected haemophilia B patients (78%) were deceased, but only 24 of 51 HIV-infected check details haemophilia A patients (47%). Death was solely or partially AIDS related in five haemophilia B patients (71%) and in 17 haemophilia A patients (71%). Median interval between HIV seroconversion and death was similar across haemophilia types (10 years in haemophilia B and 11 years in haemophilia A, P-value 0.21). In comparison, 28 of the 152 HIV-negative severe controls (18%) were deceased at the end of follow-up. Main causes of death in these patients were intracranial bleeding, malignancies, hepatitis C, other bleedings and infections. Compared with the HIV-negative patients, the age-adjusted odds ratio for dying was 4.1 in HIV-positive patients (95% CI: 1.9–8.7, P-value < 0.001). The cumulative survival since 1980 for both the HIV-positive and HIV-negative patients with severe haemophilia is shown in Fig. 2. Fifty patients (83%) ever received antiretroviral treatment, 32 of whom were treated with HAART. Median month of start of HAART was January 1997 (range: January 1996–April 2008). Of the 27 patients who were still alive and treated at our centre in 2010, 25 (93%) were on HAART.