Management of febrile neutropenia is one of the important points during the management of autologous or allogeneic hematopoietic stem cell transplantation (HSCT). In this study, we prospectively evaluated the safety and effectiveness of an empiric antibiotherapeutic scheme that associated ticarcillin-clavulanic acid (TIM) and isepamycin (ISE) for 7 days followed by a switch at day 8 to TIM and ofloxacin (OFX), additional therapy (glycopeptide or antifungal therapy) was allowed according to clinical features or documented infection. All the steps of the strategy were evaluated. Ninety six patients (73 autologous HSCT and 23 allogeneic HSCT) were enrolled. Fifty three percents of the patients presented a Fever of Unknown Origin. Bacteriologically proven infections were documented in 32 patients mainly on blood cultures (66%) with an high frequency of gram positive bacterias (71%: 38% of coagulase negative staphylococci, 16% of streptococci, 6% of S. aureus). Median time to neutrophil recovery was 14 days and 24 days respectively for autologous and allogeneic HSCT. We observed an overall effectiveness of the strategy in 78% of the cases. First step strategy was successful in 64 patients (65%) but only 26 patients kept the TIM/ISE bi therapy alone (Glycopeptide and antifungal therapy added in 29 and 7 patients). Quinolone switch was performed in 57 patients (59%) and was effective in 35 patients (61%). The success rate of TIM/OFX after success of TIM/ISE was 95%. Forty five percents of the patient could keep an oral form of OFX. Glycopeptide and antifungal therapy were necessary in 58% and 16% of cases with an effectiveness of 79% and 57% respectively. Failures were mainly observed after front-line treatment (TIM/ISE: 34 Failures) with addition of other kind of antibiotics in 95% of the cases. In 11 patients (32%) the additional antibiotic was netromidazole for clinical symptoms conclusive with anaerobic bacteria. A significantly better effectiveness was found in short duration aplasia (<15days, p=0.008) and in autologous HSCT (p=0.0041). Tolerance was good with only 6 grade III toxicities, no severe nephrotoxicity (8 grade I events, 7 in allogeneic setting) and no infectious death was observed. These results showed

(i) that the choice of the first-line beta lactamin molecule should be discussed

(ii) that oral quinolone switch is safe and effective and

(iii) furthermore pointed out the importance of different antibiotic strategies in autologous (and allogeneic HSCT using Reduced Intensity conditioning) in comparison to standard myeloablative allogeneic HSCT.

Author notes

Corresponding author

Sign in via your Institution