Prognostic factors for invasive aspergillosis in patients with haematological malignancies

Publisher Website
Google Scholar
Abstract
Reliable data defining risk factors for invasive aspergillosis (IA) in haematological patients are limited. Analysis of factors influencing IA-associated death is especially valuable in the light of recent progress in managing IA. Between 1997 and 2005 we evaluated factors influencing IA-attributable mortality. For univariate analyses we used Wilcoxon and log-rank test, and for multivariate analysis a Cox model of logistic backward regression was applied. Attributable mortality was 41% after 1 y. 50% of attributable deaths occurred within 6 weeks after IA diagnosis. Various parameters significantly correlated with death after IA: 1) uncontrolled malignancy (p =0.007); 2) extrapulmonary disease (p =0.0003); 3) stable disease, mixed response or progressive disease at first radiological evaluation (p =0.004); 4) proven IA (p =0.02); 5) IA > 110 d after PBSCT (p =0.0112). Prolonged duration of neutropenia was associated with increased mortality (p =0.0001). We observed a trend towards improved survival of IA during recent y (2003–2005). In a multivariate analysis, factors independently associated with attributable mortality included y of first diagnosis (p =0.0492), extrapulmonary IA (p =0.0353) and duration of neutropenia (p =0.0088). In conclusion, the identified risk factors may serve for the definition of high-risk situations. In these settings, increased efforts of prevention, early diagnosis and aggressive treatment of IA are warranted in order to improve survival.

This publication has 20 references indexed in Scilit: