A multicenter open-label non-inferiority randomized clinical trial comparing the safety (non-inferiority) of short antibiotic treatment (72 hours) with an anti-pseudomonal carbapenem with regard to treatment failure in comparison with extended treatment (at least 9 days) of high-risk febrile neutropenia in hematology patients receiving standard antimicrobial prophylaxis.

Episodes of fever are very common in patients undergoing intensive chemotherapy treatment for malignant hematological disease. More than 80% of patients experience one or more episodes of fever after their first cycle of chemotherapy. Only 20-30% of these patients have a clinically documented focus and mostly include infections of skin, intestinal tract and lung, while at most 10-25% of these patients have microbiologically proven bacteremia during these episodes. Patients with malignant hematological diseases and intensive chemotherapy induced neutropenia are extremely prone to overwhelming bacterial infections. Therefore, empirical antibiotic treatment is initiated at the first occurrence of fever, even if no apparent cause for the fever is evident. Most protocols advice treatment with very broad-spectrum antibiotics, mostly anti-pseudomonal carbapenems or fourth generation anti-pseudomonal cephalosporins.

Prolonged continuation of treatment may induce bacterial resistance. In view of the possible emergence of bacterial resistance due to prolonged antibiotic administration, continuation until recovery of neutropenia is suboptimal because it is costly because of longer hospital admissions, higher antibiotics costs and more possible adverse reactions.

Recent observational data (Slobbe et al) has showed that in adult hematological patients with febrile neutropenia, discontinuation of empiric antibacterial therapy after three days can be safe if no infectious etiology can be found, even in cases with persistent fever. However no RCT has hitherto been performed to support this observational data.

This study compares the safety (non-inferiority) of short treatment (72 hours) versus extended treatment (at least 9 days) with an anti-pseudomonal carbapenem for hematology patients with unexplained high risk febrile neutropenia. We hypothesize that a more restrictive use of broad-spectrum antibiotic use of three days in unexplained fever in neutropenic hematology patients is non-inferior to the present extended use during at least 9 days which would lead to a more restrictive use of antibiotics and less multiresistant strains of bacteria, costs and hospitalization length in the future.

Inclusion Criteria:

1. Patients with malignant hematological diseases being treated with cytotoxic chemotherapy or stem cell transplantation;
2. High-risk neutropenia (Absolute neutrophil count (ANC) <0.5x109/L which is expected to last longer than 7 days);
3. Fever (One single measured tympanic membrane temperature of >38.5°C or a temperature of >38.0°C during 2 subsequent measurements separated by at least 2 hours);
4. Age 18 years or older;
5. Written informed consent.

Exclusion Criteria:

1. Contraindications to use of imipenem-cilastatin or meropenem such as allergy, previous severe side-effects or previous microbiological cultures with carbapenem-resistant microorganism(s).
2. Corticosteroid use ≥10 mg per day prednisolone or equivalent for more than 3 consecutive day during the previous 7 days.
3. Clinically or microbiologically documented infection.
4. Symptoms of septic shock (systolic blood pressure <90 mm Hg unresponsive to fluid resuscitation and/or oliguria (urine production <500mL/day).
5. Previous enrollment in this study during the same episode of neutropenia.
6. Any critical illness for which Intensive Care Unit treatment is required.
7. Legal incompetency