Abstract
Introduction: Previous studies have indicated that the prognosis of patients with haematological malignancies who are admitted to intensive care unit (ICU) is poor. In particular, it has been suggested that the mortality for allogenic BMT patients requiring ICU admission is particularly high. The recent increased usage of reduced intensity conditioning has allowed allogenic transplantation of older patients who would previously be unsuitable for BMT. It is however unclear as to whether these patients may have a better ICU outcome.
Methods: A retrospective review was performed of all haemato-oncology admissions to Kings College Hospital from May 2000 to Apr 2004 who were subsequently admitted to ICU. Information was collected from all patients for demographic factors, haematological status, APACHE score, organ dysfunction, microbiological data, and supportive organ therapy at point of admission to and during ICU stay. All variables were evaluated for prognostic relevance by univariate and multivariate analyses. Post-ICU survival was examined at day 30 and 1 year.
Results: There were a total of 1249 admissions during the study period, of which 330(26.4%) were BMT patients. 57 ICU admissions (55 patients) were documented, 31 non-BMT (3.3%) vs 26 BMT (8.5%). The diagnoses were AML/MDS 26 (47.3%), ALL 6 (10.9%), NHL/HD 14 (25.5%), myeloma 5 (9.1%), others 4 (7.2%). Amongst post BMT patients, type of conditioning received was: reduced intensity 50%(13), standard myeloablative 34.6%(9), autologous 15.3%(4). 14 patients were early admissions within 6 months of BMT. The main cause of admission to ICU was due to chest sepsis with acute hypoxaemia. Conventional mechanical ventilation (MV) was used in 43(72.9%) of patients, and non-invasive MV in 16(27.1%). 30(50.8%) of patients received inotropic support during their ICU admission. Main cause of death was due to acute respiratory distress syndrome. There was no significant difference in age, duration of ICU admission and mechanical ventilation between non-BMT and BMT patients. However, the BMT group had higher numbers of myeloid malignancies, neutropenia, and intropic support. Overall ICU survival for the entire group, non-BMT, allogenic BMT (myeloablative + RIC) patients was 29.8%, 32.3% and 27.3% respectively. Kaplan-Meier estimation of longer term survival for these three groups at 30 day and 12 months was 23.7% and 14.6%, 20.1% and 10.9%, 24.3% and 19.5% respectively. The overall survival between these patient groups was not significant (p-value 0.757). Sub-analysis of RIC BMT data for 30 day and 1 year outcome was 35.8% and 29.3%, none of the 9 myeloablative patients survived beyond day 30. Univariate analysis identified intropic support, renal failure (creatinine >150), thrombocytopenia (platelet < 50) as significant variables for increased mortality (p-values 0.005, 0.012, 0.007 respectively). Results of multivariate analysis showed that inotropic support, was the only independant factor associated with increased ICU mortality. Estimated 30 day and 1 year survival for patients receiving vs not receiving inotropic support was 8% vs 39% and 6% vs 24%.
Conclusion: Our data demonstrates that the admission of haemato-oncology patients to ICU can be associated with a favourable outcome. Significantly, in our cohort the overall survival of allogenic BMT patients was comparable with non-BMT patients. In addition, RIC patients appear to have a good ICU outcome and longer term survival.
Candidemia in an intensive care unit
