Ten-year trends in CD4 cell counts at HIV and AIDS diagnosis in a London HIV clinic.

OBJECTIVE: To examine temporal trends (1986-1996) in the CD4 cell count at first HIV-1 positive test and initial AIDS diagnosis, and the influence of selected patient characteristics and treatment factors on these trends. DESIGN: A retrospective clinic-based study. SETTING: Three hospital-based clinics in West London. PATIENTS: A group of 5921 adult HIV-1-seropositive persons and 2835 reported patients with AIDS over a 10-year period from 1 January 1986 to 1 October 1996. METHODS: The CD4 cell count at HIV diagnosis (CD4HIV) was defined as the nearest CD4 cell count to within 2 months of HIV diagnosis; and the CD4 cell count at AIDS diagnosis (CD4AIDS) as the last CD4 cell count in the two months prior to the development of AIDS. Simple and multiple linear regression analysis were used to examine the influence of selected covariates on CD4HIV and CD4AIDS. RESULTS: The percentage of patients with an available CD4HIV and CD4AIDS increased from less than 5% in 1987 to 53% and 40%, respectively, in 1990, and 79% and 48%, respectively, in 1996. Patients with a missing CD4HIV or CD4AIDS were younger and less likely to have received antiretroviral therapy or prophylaxis for Pneumocystis carinii pneumonia (PCP). There was no significant change in CD4HIV over a 10-year period (median 334 x 10(6) cells/l), but a lower CD4HIV was associated with older age at presentation and injecting drug use. There was a delay in the onset of clinical AIDS, with a fall in the median CD4AIDS value from 99 x 10(6) cells/l prior to 1987, to 58 x 10(6) cells/l in 1990, 68 x 10(6) cells/l in 1994 and 60 x 10(6) cells/l in 1996; this decline in onset was seen for PCP as well as for cytomegalovirus and atypical mycobacterial infections. At all time periods, a lower CD4AIDS was associated with combined use of antiretroviral therapy and PCP prophylaxis. After adjustment for use of antiretroviral therapy and PCP prophylaxis prior to AIDS diagnosis, year of diagnosis was no longer associated with CD4AIDS. There was a significant trend towards an improved survival following AIDS diagnosis from 20.1 months prior to 1988, to 20.3 months (1989-1990), 21.0 months (1991-1992) and 22.1 (1993-1994) (P < 0.0005). CONCLUSIONS: The observed decline in CD4AIDS value was related to the introduction of antiretroviral therapy in 1988, and PCP prophylaxis in 1989. Temporal changes in the CD4 cell count at HIV and AIDS diagnosis among different demographic groups can provide insights into the changing natural history of the HIV epidemic and access to medical care. We recommend monitoring of the CD4 cell count at new HIV and AIDS diagnosis and at initiation of antiretroviral therapy as additional measures in national HIV/AIDS surveillance.