OBJECTIVES:

To assess the mortality in a cohort of HIV-infected patients starting highly active antiretroviral therapy (HAART) compared to the mortality of the general population, focusing on the influence of the CD4 cell count at the time of starting HAART.

METHODS:

Patients in the HIV Cohort Study in Western Denmark starting HAART before 1 January 2002 were identified. For each patient, 100 population controls matched on age and gender were extracted from the Danish Civil Registration System. Mortality rates were compared between the two cohorts overall, and in four groups defined by baseline CD4 cell counts.

RESULTS:

A total of 647 HIV-infected patients and 64 700 population controls were included, accounting for 53 and 815 deaths during follow-up. In the HIV group, mortality rates were 70.0 per 1000 person-years at risk in the lowest CD4 cell group (< 50 x 10 cells/l), and 3.2 in the highest (> or = 200 x 10 cells/l). Compared with population controls, mortality rate ratios declined with increasing CD4 cell counts, being 15.3 [95% confidence interval (CI), 9.8-23.8], 8.6 (95% CI, 4.3-16.8), 5.9 (95% CI, 3.0-11.4), and 3.6 (95% CI, 2.0-6.5) in the groups with CD4 cell count < 50, 50-99, 100-199, and > or = 200 x 10 cells/l.

CONCLUSION:

In comparison with the general population, HIV-infected patients starting HAART with a CD4 cell count above 200 x 10 cells/l had low mortality rates that were comparable with the rates found in other chronic medical diseases. The mortality rates increased considerably when treatment was started at lower baseline CD4 cell counts.