Abstract
Abstract 2077
Neutropenic fever is an important cause of morbidity and mortality in patients (pts) with acute leukemia who receive cytotoxic chemotherapy. Fluoroquinolone prophylaxis has been shown in randomized controlled trials to decrease the rate of neutropenic fever, with a trend towards decreased mortality in meta-analyses. The merit of fluoroquinolone prophylaxis has been debated based on the concerns of emerging antibiotic resistance, lack of a clear survival benefit and fluoroquinolone therapy related complications such as C. difficile colitis. This retrospective single-institution review examined the pattern of infections with the use of antimicrobial prophylaxis in pts with acute leukemia undergoing induction chemotherapy.
Pts >/=18y who completed induction chemotherapy for AML or ALL between Jan 2006 and Dec 2010 were eligible for the study if they received or could have received antibiotic prophylaxis for at least 1 day post-induction. Pts were excluded if they presented with fever or developed fever during induction and were continued on systemic antibiotics. Pts were stratified based on receipt of antibiotic prophylaxis versus no prophylaxis. The primary study endpoint was neutropenic fever episodes, with secondary endpoints of documented bacteremia, fungemia, C. difficile colitis, as well as time to onset of fever, length of hospital stay and mortality. Fever was defined as single oral temperature of 38.3 C or greater, or a temperature of 38.0 C for a period of an hour or longer, and neutropenia was defined as an ANC <500 cells/mm3 or an ANC that was expected to decrease to <500 cells/mm3 over the subsequent 48 hours. Odds ratios with 95% confidence intervals were calculated for neutropenic fever episodes, bacteremia, fungemia, C. difficile colitis and mortality. Time to onset of fever and length of stay were analyzed using the Wilcoxon rank-sum test.
Eighty-three consecutive pts with AML or ALL meeting the eligibility criteria were included in the study. Thirty-six pts received antibiotic prophylaxis while 47 did not. A significant decrease was noted in the rates of neutropenic fever and time to onset of fever for pts receiving prophylaxis (Table). There was no difference between prophylaxis versus no prophylaxis for other study endpoints.
| . | Patients receiving prophylaxis1 . | Patients not receiving prophylaxis . | Odds Ratio for Prophylaxis (95% confidence interval), (p-value) . |
|---|---|---|---|
| Number of pts | 36 | 47 | |
| Neutropenic Fever | 27 (75%) | 44 (94%) | 0.21 (0.03–0.93), (0.03) |
| Severe sepsis | 0 (0%) | 2 (4%) | 0 (0.00–6.95), (NS) |
| Septic shock | 1 (3%) | 2 (4%) | 0.65 (0.01–12.88), (NS) |
| Bacteremia | 3 (8%) | 11 (23%) | 0.30 (0.08–1.16), (NS) |
| Fungemia | 2 (6%) | 1 (2%) | 2.71 (0.24-31.08), (NS) |
| C. difficile infection | 4 (11%) | 5 (11%) | 1.05 (0.19–5.32), (NS) |
| Mortality | 1 (3%) | 2 (4%) | 0.65 (0.01-12.88), (NS) |
| Time to onset of fever (days) | 12.2 | 11.3 | p = 0.09 |
| Length of Stay (days) | 13.4 | 15.0 | p = 0.306 |
| . | Patients receiving prophylaxis1 . | Patients not receiving prophylaxis . | Odds Ratio for Prophylaxis (95% confidence interval), (p-value) . |
|---|---|---|---|
| Number of pts | 36 | 47 | |
| Neutropenic Fever | 27 (75%) | 44 (94%) | 0.21 (0.03–0.93), (0.03) |
| Severe sepsis | 0 (0%) | 2 (4%) | 0 (0.00–6.95), (NS) |
| Septic shock | 1 (3%) | 2 (4%) | 0.65 (0.01–12.88), (NS) |
| Bacteremia | 3 (8%) | 11 (23%) | 0.30 (0.08–1.16), (NS) |
| Fungemia | 2 (6%) | 1 (2%) | 2.71 (0.24-31.08), (NS) |
| C. difficile infection | 4 (11%) | 5 (11%) | 1.05 (0.19–5.32), (NS) |
| Mortality | 1 (3%) | 2 (4%) | 0.65 (0.01-12.88), (NS) |
| Time to onset of fever (days) | 12.2 | 11.3 | p = 0.09 |
| Length of Stay (days) | 13.4 | 15.0 | p = 0.306 |
33 of 36 pts received moxifloxacin prophylaxis, one patient each received amoxicillin, amoxicillin/clavulanate, and ciprofloxacin.
This retrospective review found a significantly lower rate and delay in onset of neutropenic fever in AML and ALL patients receiving fluoroquinolone prophylaxis, but no difference in patterns of infection compared with non-prophylaxed pts. A trend towards fewer episodes of bacteremia in the prophylaxis group was noted, but did not reach statistical significance. Further data collection and analysis is ongoing for pts undergoing consolidation chemotherapy for AML and ALL or re-induction chemotherapy following relapse for assessment of neutropenic fever episodes and patterns of infection.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
