Bloodstream infections due to Gram-negative bacteria in patients with hematologic malignancies: updated epidemiology and risk factors for multidrug-resistant strains in an Italian perspective survey

•High mortality rates in patients with hematological malignancies (HM) and bloodstream infections (BSI) caused by Gram-negative bacteria (GNB).•High prevalence of multidrug-resistant (MDR) strains among GNB causing BSI in HM patients.•Significant reduction in use of fluoroquinolone prophylaxis since 2009-2012 survey.•Increased rates of susceptibility to fluoroquinolones in most isolates and to most antibiotics tested among Pseudomonas aeruginosa isolates.•Fluoroquinolone prophylaxis and previous rectal colonization by MDR bacteria important independent risk factors for MDR GNB BSI.•Further studies needed to better understand the role of fluoroquinolone prophylaxis in countries with high prevalence of infections. Bloodstream infections (BSI) caused by Gram-negative bacteria (GNB) in patients with hematological malignancies (HM) have been associated with high mortality rates, particularly with infections caused by antibiotic-resistant strains. A multicenter cohort study including all consecutive episodes of GNB BSI in HM patients was conducted to update the epidemiology and antibiotic resistance patterns (compared to our previous survey conducted between 2009 and 2012) and investigate risk factors for GNB BSI due to multidrug-resistant (MDR) isolates. A total of 834 GNB were recovered in 811 BSI episodes from January 2016 to December 2018. Compared to the previous survey, there was a significant reduction in use of fluoroquinolone prophylaxis and a significant recovery in susceptibility rates to ciprofloxacin among Pseudomonas aeruginosa, Escherichia coli and Enterobacter cloacae isolates. In addition, there was a shift to a significantly increased susceptibility of P. aeruginosa isolates to ceftazidime, meropenem, and gentamicin. A total of 256/834 (30.7%) isolates were MDR. In multivariable analysis, MDR bacteria culture-positive surveillance rectal swabs, previous therapy with aminoglycosides and carbapenems, fluoroquinolone prophylaxis, and time at risk were independently associated with MDR GNB BSI. In conclusion, despite the persistence of a high prevalence of MDR GNB, there was a shift to a reduced use of fluoroquinolone prophylaxis and increased rates of susceptibility to fluoroquinolones in almost all isolates and to almost all antibiotics tested among P. aeruginosa isolates, compared to our previous survey. Fluoroquinolone prophylaxis and previous rectal colonization by MDR bacteria were independent risk factors for MDR GNB BSI in the present study