Predictors of Short- and Long-term Survival in HIV-Infected Patients Admitted to the ICU: Discussion
Previously reported 1-year survival rates after ICU discharge of non-HIV-infected patients have varied widely in the few available studies, from about 30% in patients with severe sepsis and 49% in elderly subjects (with a mean survival time of 18 ±10 months) to <5% in patients who required cardiopulmonary resuscitation and in bone marrow transplant recipients. Our results in HIV-infected patients compare favorably with those previously reported for other high-risk patients or patients with ultimately fatal illnesses. Treatment of Bronchial
It is difficult to compare our long-term outcomes results with those obtained for HIV-infected patients who did not require ICU treatment. Despite the heterogeneity of our study population, most of our patients had severe HIV disease, with CD4+ counts below 0.100X109/L in 75% of cases and AIDS in 80%. In the present study, median survival time for patients with a first AIDS-defining condition was approximately 11 months, which is only half the median survival time reported following AIDS onset. However, this has also been observed in non-HIV-infected patients. In a selected population discharged from the ICU, the survival time was significantly lower when compared with the expected survival in the age- and sex-matched population.
To our knowledge, functional status has not been evaluated previously as a prognostic marker in the ICU but is currently used in HIV-infected patients. Our results provide evidence that this easy-to-calcu-late parameter is of value for predicting short- and long-term survivals. Importantly, we found that the functional status in patients still alive at the study end was similar to their functional status before ICU admission. This finding in patients with an ultimately fatal illness is extremely encouraging. Weight loss may be closely associated with the outcome in HIV-infected patients. The substantial impact of nutritional status on the short-term outcome in non-HIV-infected patients admitted to the ICU has been demonstrated convincingly Moreover, a recent study found evidence that this parameter also influenced the short-term outcome in HIV-infected patients admitted to the ICU. Our results indicate that weight loss prior to ICU admission is an important short- and long-term prognostic marker. The health status prior to ICU admission, as evaluated based on functional status and weight loss, is probably the most powerful predictor of both short- and long-term outcomes.