Logarithmic relationship of the CD4 count to survival in patients with human immunodeficiency virus infection.

BACKGROUND: Survival, and the incidence of events that define the acquired immunodeficiency syndrome (AIDS), are known to be inversely related to the CD4 count in patients with human immunodeficiency virus infection. We wished to quantify this relationship more precisely, particularly for patients with CD4 counts of less than 50/mm3. METHODS: Prospective surveillance for survival and for all AIDS-defining events was performed on all 2682 patients with human immunodeficiency virus infection who had at least one CD4 count performed at a large urban public hospital during a 3-year period. Product-limit survival and incidence of AIDS-defining events were calculated as a function of baseline CD4 count. RESULTS: The 1-year product-limit survival was 17% +/- 6% for patients after a baseline CD4 count of 1 to 4/mm3; 44% +/- 6% after a count of 5 to 9/mm3; 48% +/- 5% after a count of 10 to 19/mm3; 51% +/- 4% after a count of 20 to 39/mm3; 62% +/- 5% after a count of 40 to 59/mm3; 71% +/- 4% after a count of 60 to 99/mm3; 79% +/- 3% after a count of 100 to 199/mm3; and 92% +/- 2% after a count of 200 to 499/mm3. One-year survival and baseline CD4 count were related by the following formula: percent 1-year survival = 10 + 32(log10 CD4 count) (R2 = .97; P < .001). The 1-year incidence of a first AIDS-defining event and baseline CD4 count were related by the following formula: percent developing AIDS in 1 year = 104-36(log10 CD4 count) (R2 = .89; P < .001). Similar relationships were calculated between the logarithm of the baseline CD4 count and the 1-year incidence of most AIDS-defining events. These relationships were linear over the CD4 range of 1 to 499/mm3 and over follow-up periods of 6 months to 2 years. CONCLUSIONS: The relationship of the CD4 count to survival, and to the incidence of AIDS-defining events, is logarithmic. This relationship helps explain the substantial differences in 1-year survival associated with baseline CD4 counts in the range below 50/mm3.

Incidence of Mycobacterium avium-intracellulare complex bacteremia in human immunodeficiency virus-positive patients.

The product-limit incidence of Mycobacterium avium-intracellulare complex (MAC) bacteremia in 1006 human immunodeficiency virus (HIV)-positive patients followed at one institution over a 3-year period from the day of AIDS diagnosis with monthly lysis-centrifugation blood cultures was 21% +/- 2% SE at 1 year and 43% +/- 3% at 2 years. The product-limit incidence of MAC bacteremia at 1 year after the patients' first CD4 cell count was related to both the CD4 cell count and to whether they had an AIDS diagnosis (both P less than .0001) but not to age, sex, or race. This incidence was 39% +/- 6% for CD4 cell counts of less than 10/mm3, 30% +/- 5% for 10-19/mm3, 20% +/- 4% for 20-39/mm3, 15% +/- 4% for 40-59/mm3, 8% +/- 3% for 60-99/mm3, and 3% +/- 1% for 100-199/mm3. MAC may eventually infect most if not all HIV-positive patients who do not die from another HIV-related event.

Primary prophylaxis with fluconazole against systemic fungal infections in HIV-positive patients.

OBJECTIVE: To investigate the efficacy of fluconazole prophylaxis against systemic fungal infections in HIV-positive patients. DESIGN: Open label treatment compared with historical controls. SETTING: Patients were seen at the Parkland Memorial Hospital HIV Clinic, Dallas, Texas, USA between 1 March 1990 and 28 February 1991. PATIENTS, PARTICIPANTS: Three hundred and thirty-seven historical controls were followed for 157 patient-years, and 329 fluconazole-treated patients for 145 patient-years. INTERVENTIONS: Fluconazole (100 mg daily) was administered to all patients with CD4 lymphocyte counts less than 68 x 10(6)/l seen at our HIV clinic after 1 March 1990. MAIN OUTCOME MEASURES: Lysis-centrifugation blood cultures were recorded monthly for all patients during both study periods. RESULTS: Twenty infections (16 cryptococcosis, four histoplasmosis) occurred in 337 historical reference control patients (product-limit 1-year incidence, 7.5 +/- 2.0/year). Four infections (one cryptococcosis, three histoplasmosis) occurred in the treated patient group (product-limit 1-year incidence, 1.8 +/- 0.9/year). CONCLUSIONS: Fluconazole warrants further evaluation for prophylaxis against systemic fungal infections in HIV-positive patients.