Outcome and Prognostic Factors in Patients with Hematological Malignancy That Are Admitted to the Intensive Care Unit.

When patients with hematological malignancies develop a life-threatening complication there may be reluctance to admit them in intensive care units (ICU) because of their supposed poor prognosis. The objective of this study was to evaluate the mortality during the ICU admission, the long-term survival, and the prognostic factors that contribute to the survival of patients with hematological malignancies who were transferred to ICU due to a life-threatening complication. From January 2000 to May 2004, the variables at admission and during stay at the ICU, and the follow-up were reviwed in 58 consecutive critically-ill patients with a hematological malignancy from a single institution. The median age (range) was 55 (15–75) years and the male/female ratio was: 38/20. The hematological underlying diseases were: NHL (18 patients), AML (10), ALL (9), MM (6), chronic lymphoproliferative disorder (5), chronic myeloproliferative disorder (4), myelodysplastic syndrome (3), aplastic anemia (2) and Hodgkin’s lymphoma (1). Seven patients had received a hematopoietic stem cell transplant prior to the ICU admission. The main life-threatening acute illness precipitating the ICU transfer were: septic shock (26 patients, 45%), respiratory failure (21, 36%), non-septic hemodynamic instability (5, 9%), respiratory arrest related to a neurological event (2, 3%), post-surgical status (2, 3%), cardiac infarction (1, 2%) and polytrauma (1, 2%). Twenty-one patients (36%) could be discharged alive from the ICU. The median overall survival (range) for ICU discharged patients was 23 (0–54) months, with a median follow-up of 8 months. The actuarial probability of discharged patients to be alive was 56% (CI 95%: 31–75) at 6 months, and a 48% (CI 95%: 13–70) at 12 months. The mean Acute Physiology and Chronic Health Evaluation II (APACHE) score at admission, neutropenia, need for mechanical ventilation, maximum FIO2 requirements at 24 hours from admission, presence of septic shock, renal impairment or liver damage, were associated with a poor outcome in the univariate analysis. A documented infection was not associated with a higher mortality rate except for fungal infection. The APACHE II score at 48 and 72 hours of ICU admission decreased both in surviving and non-surviving patients due to therapeutic manoeuvres and was not predictive of the outcome. The type of the hematological malignancy, its prognosis and the presence of active disease at ICU admission did not predict patients outcome in our series. The number of failing organs also predicted a poorer survival for patients with more than two failing organs (p=0.038). In a multivariate logistical regression model, only the cardiovascular failure requiring vasoactive and the need of mechanical ventilation predicted outcome in the ICU admitted patients diagnosed with a hematological malignancy. A high proportion of admitted patients with a life-threatening complication and a hematological malignancy could be discharged from ICU. Although the mortality rate immediately after ICU discharge was high, those patients that survived the first week outside ICU had an expected survival only conditioned by their hematological malignancy.