Background: Recent treatments in hematological malignancies have substantially improved. Unfortunately, once a patient with a hematological malignancy has complications, the prognosis is poor and the in hospital and ICU mortality rates are high. Debates concerning the reluctance to admit patients into ICUs with poor prognoses often emerge. The aim of the present study is to identify the patients who are more likely to benefit from ICU admissions.
Objectives: To assess the outcomes and to identify early mortality risk factors in patients with lymphoma and acute myeloid leukemia admitted to the Intensive Care Unit (ICU) at Songklanagarind Hospital in the south of Thailand.
Material and Method: This is a retrospective study of patients diagnosed with lymphoma and acute myeloid leukemia admitted to the ICU during the period of January 2004 through May 2008. Demographic factors, acute physiology, Acute Physiology and Chronic Health Evaluation (APACHE) II scores and variables noted in the first 24-hours were collected. The risk factors for deaths in the ICU were studied by univariate and multivariate analysis. The risk factors taken from the best multivariate analysis model were calculated to predict the probability of ICU mortality.
Results: A total of 145 patients were studied. The ICU mortality rate was 55.2%. The major cause of death was septic shock. Using univariate analysis, the significant mortality risk factors were neutropenia, mechanical ventilation, the use of vasopressors, abnormal serum creatinine (Cr) and APACHE II scores (p<0.05). Using multivariate analysis, ICU mortality was best predicted on admission by mechanical ventilation, the use of vasopressors and the APACHE II scores. The presence of neutropenia, mechanical ventilation, vasopressors and an APACHE II score of greater than 27 predicts 80% sensitivity and a 75% specificity for an 82% ICU mortality.
Conclusion: Patients with lymphoma and acute myeloid leukemia admitted into the ICU referral center in the south of Thailand who had mechanical ventilation, use of vasopressors and APACHE II scores greater than 27 were associated with a higher ICU mortality rate. The authors suggest that early identification of the subgroup of patients whose probability of survival is so low that advanced ICU support should not be continued would be a more reasonable goal. This will allow more efficient care to potential survivors not in this group.

Keywords: Mortality, Acute leukemia, Lymphoma, Intensive care unit