Early Admission to the Intensive Care Unit In High Risk Acute Myeloid Leukemia Patients

Abstract

4364 INTRODUCTION: Acute Myeloblastic Leukemia (AML) is considered as an oncology emergency as a proportion of patients experience life threatening complications within the first hours or days after diagnosis. Early death had been demonstrated to be statistically related to high white blood cell (WBC) and monoblastic leukemia, with leukostasis and lysis syndrome as the most deadful events. OBJECTIVES: To evaluate the relationship between timing of admission to the Intensive Care Unit (ICU) and outcomes in high risk AML patients at the earliest phase of the malignancy (before any chemotherapy) METHODS: Retrospective study in a tertiary care center teaching hospital between 1998 and 2008. Adult patients with newly diagnosed AML were included. Patients admitted for an immediate life sustaining therapy (ventilation, vasopressors or renal replacement therapy) were excluded. 42 patients admitted directly to the ICU (Early admission) were matched for age, WBC and FAB subtype with 42 patients primarily admitted in hematology ward. Medical charts were reviewed and datasets extracted. RESULTS: Overall 84 patients were included in the study (42 patients early admitted to the ICU and 42 patients admitted first to the wards). Median follow up was 10,3 months. Median age was 46,5 years (36-57). FAB M4 or M5 was retrieved in 58% of the patients. According to MRC, karyotype was favorable for 30% and poor for 19%. Median WBC was 103×109.L-1. No statistical difference was seen for demographic and hematological parameters between early admitted patients and matched controls. Among the 42 patients admitted first to the wards (controls), 20 were subsequently admitted to the ICU (Lately admitted) and 22 remained in ward during the entire treatment course (Never admitted). The median time between diagnostic and ICU admission of this last group was 4 (1-9) days. Strikingly, patients lately admitted had more frequently dyspnea, oxygen requirement, high respiratory rate, low diastolic arterial pressure and lower first 24h urine output (p<0,05 for each). Lately admitted patients were less likely to receive the complete dose of induction chemotherapy than early admitted patients (68% vs. 88%,p<0,05) Furthermore, Late admission resulted in increased use of invasive mechanical ventilation (60 vs. 33%) and vaso-active drugs (60 vs. 16%,p<0,05) These differences resulted in longer stay in ICU and decreased survival. CONCLUSIONS: Patients at the earliest phase of High risk AML who are lately admitted to the ICU experience worse outcomes, with increased use of life-sustaining therapies and higher mortality, compared to patients early admitted to the ICU. Physiologic parameters at the time of AML diagnosis such as respiratory rate, diastolic blood pressure, SpO2, or oxygen need are likely to help clinicians distinguish those patients at risk of late ICU admission and subsequent adverse outcomes. Studies are needed to assess the right place for newly diagnosed AML with physiological abnormalities but no organ dysfunction. Disclosures: No relevant conflicts of interest to declare.