Efficacy of Short-Course Antibiotics for Bloodstream Infections in Patients With Acute Myeloid Leukemia, Febrile Neutropenia

Bloodstream infection
Bloodstream infection
Researchers conducted a study to compare the effects of short- vs long-course antibiotics for the treatment of bloodstream infections in patients with acute myeloid leukemia and chemotherapy-induced febrile neutropenia.

Among patients with acute myeloid leukemia (AML) and chemotherapy-induced febrile neutropenia (FN) who developed a bloodstream infection (BSI), treatment with a short-course of antibiotics was found to be as effective as long-course antibiotics. These findings were published in the Journal of Infection.

Between 2017 and 2019, researchers conducted a single-center retrospective study among patients (N=71) with AML and chemotherapy-induced FN who were treated for a BSI at a tertiary care center in France. The researchers sought to compare the effects short-course antibiotics (≤7 days) vs long-course of antibiotics (≥7 days) for the treatment of BSI. The primary outcome was the number of recurrent BSIs within 30 days of antibiotic discontinuation among patients who received short-course antibiotics vs those who received long-course antibiotics. The researchers defined FN as an absolute neutrophil count of greater than 500 cells/mm3, and BSI was defined as the detection of a pathogen, other than potential skin contaminants, in 1 or more blood culture specimens.

The researchers analyzed 104 BSI episodes from a total of 71 patients included in the study, of whom 46% received short-course antibiotics and 54% received long-course antibiotics. Among patients in the short- and long-course antibiotic groups, the median duration of FN was 23.5 and 24 days, the median duration of fever was 3 and 7.5 days (P =.02), and the median duration of antibiotic treatment was 6 and 11 days (P <.001), respectively.

Among patients in both the short- (n=48) and long-course (n=56) treatment groups, 43.8% and 37.9% were on induction chemotherapy, 27.0% and 26.8% were on consolidation chemotherapy, and 22.9% and 2.8% were on salvage chemotherapy, respectively. In addition, 6.3% and 8.9% of patients in the short- and long-course treatment groups had a history of multi-drug resistant bacterial infection, and 6.3% and 16.1% required transfer to an intensive care unit (P =.008), respectively.

Of the 104 BSI episodes, 15% were catheter-related, 11% were caused by digestive translocation, 2% were related to cellulitis or soft tissue abscess, and 2% were related to pneumonia. In addition, 65% of episodes were defined as a primary BSI, which included patients with no clinical or radiologic signs of infection.

Among all pathogenic species identified in patients with BSIs, 42% were Enterobacteriaceae, 21% were coagulase-negative Staphylococci, 13%were Streptococci and Enterococci, 9% were non-fermenting bacteria, 8% were polymicrobial, and 7% were miscellaneous.

In patients with catheter-related BSIs, the researchers noted that 63% were caused by coagulase-negative Staphylococci. In addition,  non-fermenting bacteria was most common cause of BSIs among patients with urinary tract infections (40%) and those with cellulitis or soft tissue abscess-associated BSIs (50%). Polymicrobial infections were more common among pneumoniae-associated BSIs (50%).

A total of 7.7% of patients (n=8) developed recurrent BSI within 30 days of antibiotic discontinuation, of whom 5 received short-course antibiotics and 3 received long-course antibiotics. No significant association was found between short-course antibiotic therapy and recurrent BSI risk (P =.37). Of note, only the duration of FN was found to be a significant risk factor for recurrent BSI (odds ratio, 1.04; 95% CI, 1.01-1.06; P =.003). In addition, the median duration of FN was increased among patients who developed recurrent BSI vs those who did not (45 vs 22.5 days; P =.005).

After discontinuation of antibiotic therapy, the researchers noted that the mortality rate at 30 days was 7.7%, which represented 2 patients in the short-course treatment group and 6 and in the long-course treatment group. Of note, only 1 death that occurred among patients who received long-course antibiotics was considered to be of infectious origin.

This study was limited by its retrospective design and the lack of data regarding potential adverse effects from antibiotic therapy.

In regard to patients with AML who develop a BSI, the researchers concluded that “[additional] prospective interventional studies…are needed to assess the efficacy and safety of [short-course antibiotic therapy] in these patients.”


Métais A, Torregrosa Diaz JM, Gallego Hernanz MP, et al. Efficacy of antibiotic short course for bloodstream infections in acute myeloid leukemia patients with febrile neutropenia: a retrospective comparative study. J Infect. 2021;S0163-4453(21)00532-6. doi:10.1016/j.jinf.2021.10.017