Skip to main content
. 2018 Dec 13;18:656. doi: 10.1186/s12879-018-3584-9

Table 4.

Predictors of mortality within 42 days of initiation of IA therapy in patients with hematologic malignancy and aspergillosis

Variables Univariate Multivariate analysis
Dead
(n = 132)
N (%)
Alive
(n = 208)
N (%)
p-value HR (95% CI) p-value
Age (years), median (range) 50 (16–80) 56 (14–86) 0.07
Gender, male 84 (64) 119 (57) 0.24
Diagnosis of invasive aspergillosis 0.001 0.023
 Definite 54 (41) 50/207 (24) 1.5 (1.1, 2.2)
 Probable 78 (59) 157/207 (76) Referent
Type of IA infection a 0.16 0.007
 Invasive pulmonary infection 111 (84) 174 (84) 3.3 (1.5, 7.2) (0.003)
 Disseminated infection 13 (10) 12 (6) 4.0 (1.6, 10.3) (0.003)
 Localized or sinus infection 8 (6) 22 (11) Referent
Type of cancer 0.05
 Leukemia 106/131 (81) 148/205 (72)
 Lymphoma 23/131 (18) 43/205 (21)
 Myeloma 2/131 (2) 14/205 (7)
Transplantation within 1 year prior to infection 41/131 (31) 91 (44) 0.022 0.60 (0.41, 0.88) 0.008
Type of transplantation within prior year 0.06
 Allogeneic transplant 39/41 (95) 74/91 (81)
 Autologous transplant 2/41 (8) 17/91 (19)
Graft vs Host Disease (GVHD) 29/39 (74) 53/74 (72) 0.76
Neutropenia (< 500 ANC) at onset of IA 79/130 (61) 76/199 (38) < .0001
Persistent neutropenia 59/130 (45) 63/179 (35) 0.07
Received immunotherapy 92 (70) 108/205 (53) 0.002
Received WBC transfusion 31 (23) 24/206 (12) 0.004
Year of IA diagnosis/treatment < .0001
 1993–2004 99 (75) 87 (42)
 2005–2016 33 (25) 121 (58)
Prophylactic antifungal treatment prior to infection 36 (27) 76/207 (37) 0.07 0.61 (0.41, 0.90) 0.012
Breakthrough 29/36 (81) 69/75 (92) 0.11
Primary therapy strategy < .0001 < .0001
 Diagnostic-driven therapy with voriconazole 2 (2) 42 (20) Referent
 Empiric antifungal therapy without voriconazole 123 (93) 97 (47) 18.0 (4.4, 73.2) (< .0001)
 Empiric antifungal therapy with voriconazole 7 (5) 69 (33) 2.2 (0.5, 10.6) (0.33)

IA = Invasive aspergillosis; WBC = White blood cell; HR = Hazard ratio; 95% CI = 95% Confidence interval

aAll patients with more than one types of IA infections in this study had invasive pulmonary infection and were included in invasive pulmonary infection category in this analysis