Introduction- Acinetobacter baumannii is a Gram-negative strictly aerobic non-fermentative coccobacillus, widely distributed in nature. Multidrug resistant strains, including carbapenem antibiotic resistant, have been associated with fulminant severe sepsis and high overall mortality rate, ranging between 25 and 54%. This is true mainly for intensive care unit (ICU) and immunocompromised patients. Although CRAB bacteremia has previously been reported, studies involving hemato-oncological patients have rarely been reported.

Aim - To investigate factors associated with 7 day mortality, defined as death within 7 days of documented bacteremia among hemato-oncological patients with CRAB bacteremia in a tertiary medical center in Israel.

Methods - A retrospective analysis of all patients with CRAB bacteremia treated in a large tertiary care hospital January 2008 to June 2018 was performed. Among them, 46 hemato-oncological patients were identified and analyzed.

Baseline patient and infection characteristics were collected from electronic medical records. In addition, antibiotic management and outcomes were documented. Univariable and multivariable analysis of risk factors for 7 day mortality were performed. Variables statistically significant in the univariable analysis were introduced into the regression model.

Results - A total of 46 hemato-oncological patients with CARB bacteremia were included in this study. At the time that bacteremia occurred - 18 (39.5%) patients were hospitalized in internal medicine departments, 4 (8.7%) patients in ICU, 13 (28.3%) patients in the hemato-oncology ward and 11 (23.9%) patients in the stem cell transplantation ward.

Mean age of patients was 62.5±16.8 yr., with 20 (43.5%) males. The mean length of stay before bacteremia occurrence was 29 ± 42.8 days. The presumed sources of bacteremia were respiratory tract (n = 25, 54.5%), vascular catheter (n = 14, 30.5%), urinary tract (n = 2, 4%), and other/unknown sources (n = 5, 11%). The susceptibility of CRAB to colistin was 100%, 7 isolates were susceptible to ampicillin/sulbactam and 4 of 38 tested were susceptible to tigecycline.

The overall 7 day mortality in patients with CARB bacteremia was 71.7% (33/46), with an overall 30 day mortality of 93.4% (43/46).

Univariable analysis showed that the significant risk factors for 7 day mortality in hemato-oncological patients with CARB bacteremia was higher Sepsis-related Organ Failure Assessment (SOFA) score (Survivors group - median score - 7, IQR 4-10.5, non-survivors group - median score- 13, IQR 10.5-16, p<0.001). Protective factors included: Foley catheter insertion prior to bacteremia (11/33, 84.6% vs. 16/33, 48.5%, p=0.044), appropriate antibiotic therapy within 48h of bacteremia (11/13, 84.6% vs. 11/33, 33.3%, p=0.003) and central venous catheter removal (7/9, 77.8% vs. 4/22, 18.2%, p=0.003) all of which were associated with decreased risk for early mortality (Table 1).

In multivariable analysis, factors remaining statistically significant were high SOFA score - associated with increased mortality (OR 2.021, 95% CI 1.246-3.277, p=0.004) and Foley catheter insertion prior to bacteremia as well as appropriate antibiotic therapy within 48h of bacteremia (OR 0.024, 0.001-0.593, p=0.023; OR 0.034, 95% CI 0.002-0.700, p=0.0028, respectively) associated with a decreased risk for 7 day mortality (Table 2).

Discussion - Early administration of appropriate antibiotics, mainly colistin, resulted in decreased early mortality in CRAB bacteremia. Patients with Foley catheter had better survival, probably reflecting an undiagnosed urinary cause (which is usually associated with a better outcome). A more severe clinical presentation, represented by higher SOFA score, was associated with higher mortality, underlying the need for urgent treatment in this group of patients. Overall 30 day mortality is extremely high in this group of patients with a death toll of nearly 100%.

Conclusions - Bacteremia with CRAB is often fulminant and fatal, particularly among hemato-oncological patients. Early implementation of appropriate antibiotics might improve outcome. Due to the severity of the infection and the high mortality rate, prevention of this catastrophic complication in hemato-oncological patients is crucial.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution