|
Impact of antiretroviral therapy in survival of HIV-infected patients admitted to intensive care unit
J. Croda1, M. Croda2, A. Neves3, S.D.S. Santos3
1Universidade de São Paulo - Faculdade de Medicina, Departamento de Doenças Infecciosas e Parasitárias e Departamento de Patologia, São Paulo, Brazil, 2Instituto de Infectologia Emílio Ribas, São Paulo, Brazil, 3Universidade de São Paulo - Faculdade de Medicina, Departamento de Doenças Infecciosas e Parasitárias, São Paulo, Brazil
Background: Although the reasons for admission and the prognostic markers of HIV-infected patients admitted to ICU have been extensively reported, there is urgent need of consistent data that support the ART utilization in critical ill patients.
Methods: Retrospective cohort study conducted in patients admitted to ICU in a tertiary-care teaching hospital of Universidade São Paulo from 1996 through 2006. The follow-up period extended for 6 months after ICU admission. A total of 278 HIV-infected patients admissions to ICU were selected. We excluded ICU re-admissions (37), ICU stay admissions of less than 24 hours (44) and patients with unavailable medical charts (36).
Results: The in-ICU and 6-month mortality rates were 55.4% and 69.2%, respectively. In multivariate logistic regression analysis and Cox-proportional hazards models the variables associated with in-ICU and 6-month mortality were sepsis as cause of admission (OR=3.469|95%CI=1.410-8.533; HR=1.724;95%CI=1.135-2.619), APACHE-II score >19 (OR=2.786|95%CI=1.207-6.428; HR=1.810|95%CI=1.173-2.793) and mechanical ventilation in the first 24 hours (OR=3.614|95%CI=1.606-8.133; HR=2.203|95%CI=1.409-3.443). LDH higher than 1000 mg/dl was only associated with ICU mortality (OR=3.871|95%CI=1.592-9.413). Use of ART in ICU was negatively predictive of 6-months mortality in the Cox model (HR=0.627|95%CI=0.393-0.999), mainly if this therapy was introduced during the first 4 days (HR=0.584|95%CI=0.410-0.832) of admission in ICU. When compared only the group of patients who had opportunity to beginning ART, the HIV-infected patients that staring ART during ICU admission have better prognostic adjusted by time and potential confounding (HR 0.497| 95%CI=0.250-0.989).
Conclusions: The ICU outcome of HIV-infected patients appears to be dependent not only of acute illness severity but also of administration of antiretroviral treatment.
|