Akut Lösemi Hastalarında Mekanik Ventilasyon Ihtiyacı: Sıklık, Risk Faktörleri ve Prognozu Etkileyen Faktörlerin Belirlenmesi

Abstract

Acute leukemia is a group of heterogeneous diseases characterized by proliferation of blastic cells in bone marrow. High mortality and morbidity are seen in acute leukemia patients. The prognosis has been relatively improved during last few decades due to better supportive care. There are a few studies related to respiratory support risk factors and prognosis in acute leukemia. In this study, we aimed to determine rate of respiratory support, prognostic factors and risk factors for acute leukemia. All acute leukemia patients followed in our center between 2003 and 2014 were included in this study. Demographic characteristics of patients, respiratory support rate and etiologies, risk factors for respiratory failure and prognostic factors were determined and overall survival of patients who needed respiratory support was determined. The rate of respiratory support was 30.6% (139/460) and pneumonia was the most common cause of respiratory failure. Other etiologies were ATRA syndrome, leukostasis, pulmonary haemorrhage, pulmonary infiltration of unknown reason and extra-pulmonary causes (neurological causes, hemodynamic instability and cardiopulmonary arrest) which were excluded from the statistical calculations for risk analysis and prognosis. 604 febrile neutropenia episodes which were identified in detail were used for determination of risk factors and prognosis. In multivariate analysis, age, ECOG performance status, febrile neutropenia prophylaxis, pulmonary infiltration during neutropenia episode and induction or reinduction chemoterapies were determined as independent risk factors for respiratory failure in acute leukemia. Using regression coefficients of statistically significant independent risk factors, a statistical model for risk prediction was developed. 60-days' overall survival of patients who needed mechanical ventilation was 5.5%. In multivariate analysis, pH>7.35 was determined to be the only statistically significant independent variable for prediction of improvement. Consequently, we found that the need for respiratory support during acute leukemia treatment was quite frequent. These patients have a high mortality risk. As determination of high risk patients may provide important clinical benefits, validation of our risk scoring system by appropriately designed studies has utmost importance.