Hospital service interventions and improving survival of AIDS patients St Mary's Hospital, London, 1982-1991.

Abstract

The relationship between changes in hospital service interventions at St Mary's Hospital, London, reduced case fatality for patients with their first episode of Pneumocystis carinii pneumonia (PCP) and improved survival from diagnosis of AIDS was investigated for the period 1982-1991. Multivariate logistic regression models identified factors independently associated with episode survival; for those patients who survived their first episode of PCP, survival from time of diagnosis of AIDS was analysed using multivariate Cox's proportional hazards models. The case-fatality rate after 1987 was significantly lower for the 159 subjects. Median survival from diagnosis of AIDS increased significantly from 142 days to 554 days (P=0.01). Improved survival of first episode of PCP was associated with it being the index diagnosis and having a haemoglobin at diagnosis of PCP greater than 12 g/dl. The presence of a concurrent AIDS-defining condition in patients who presented with an A-a gradient equal to or greater than 40 mmHg was associated with reduced episode survival, especially before 1987. For the 126 individuals who survived their first episode of PCP, death rates were lowest in patients treated with primary or secondary PCP prophylaxis and those who received zidovudine since their first episode of PCP. Survival in patients with HIV disease is better in patients who receive appropriate antiretroviral treatment of HIV infection and timely treatment of opportunistic illnesses. Early diagnosis of HIV-1 infection with early diagnosis and treatment of first episode of PCP was associated with improved episode survival. Subsequent medical follow up combined with PCP prophylaxis and zidovudine were significantly associated with long-term survival.