Sparing Anti-Bacterial Prophylaxis in Acute Myeloid Leukemia during Post Induction Aplasia Does Not Impact on Bacteriemiae Incidence and Early Death: A Retrospective Single Center Study

Introduction:

Acute myeloid leukemia (AML) patients are at high risk of infections during post chemotherapy neutropenia, especially after induction. The anti-bacterial prophylactic approach raised concerns about the emergence of chemo-resistant pathogens and the increased incidence of Gram negative bacteriaemiae during consolidation therapy. The present study aimed at evaluating the need of anti-bacterial prophylaxis or not during induction.

Methods:

To address this point, we set up a retrospective single center study enrolling all consecutively newly diagnosed AML patients admitted in our institution; acute promyelocytic leukemia patients were excluded. All patients received an intensive standard induction chemotherapy. Patients were divided in two groups on the basis of anti-bacterial prophylaxis (levofloxacin 500 mg OD during aplasia and until neutrophil recovery or until intravenous antibiotic treatment was needed) given (Group A) or not (Group B). Group A consisted in consecutive patients treated from June 2001 to December 2016 and group B in those consecutively treated from January 2017 to May 2018. Clinical and microbiological data were collected and compared between the two groups. The primary endpoint was the number of bacteriaemiae. Statistical analysis was performed using Chi-square test for categorical variables while continuous variables were compared by Student's t-test; p value of <0.05 was considered significant.

Results:

A total of 402 consecutive AML patients were enrolled from June 2001 to May 2018. A total of 223 were male while 179 were female, the median age at diagnosis was 54 years (19-76). After induction, a complete remission was achieved in 280 patients (70%), 96 were resistant (24%) and 26 died (6%). Anti-bacterial prophylaxis was administered in 343 patients (group A) while 59 patients experienced aplasia without anti-bacterial prophylaxis (group B). Baseline characteristics were equally distributed among the patients in both groups with the exception of age (median 55 years vs 62 in group A vs B, respectively p<0.0001).

In group A, 313 patients developed fever (91%), while 30 (9%) remained afebrile; in the group B, 56 patients developed fever (95%), while 3 (5%) remained afebrile (p=0.3439). The median number of fever days was similar in both groups (group A vs B: median 6 days, range 0-42 and median 5, range 0-17, respectively, p=0.0873). The incidence of bacteriemiae was 25% (n=87) in group A compared to 32% (n=19) in group B (p=0.2708). Considering the total number of bacteriemiae, the incidence of gram-negative infections were 32% (n=28) vs 47% (n=9), wherease gram-positive infections were 68% (n=59) vs 53% (n=10) in the group A vs B, respectively (p=0.2084). Klebsiella pneumonia carbapenemase-producing (KPC) positive blood cultures were detected in two patients in group A, one with KPC positive rectal swab, while in group B there was a case without evidence of KPC on rectal swab. A septic shock was evidenced in 5% (n=17) vs 11% (n=6) of patients with fever during induction in group A and B respectively (p=0.1320). Early induction deaths were similar in both groups with 22 deaths in group A (6%) and 4 in group B (7%), respectively (p=0.9160). In group A, we observed 80 (23%) pneumonia, 53 (15%) respiratory failure and 35 (10%) neutropenic colitis. In group B we observed 17 (29%) pneumonia, 17 (29%) respiratory failure and 18 (29%) neutropenic colitis.

Conclusions:

Our study showed that omitting levofloxacin prophylaxis did not cause an increase incidence of bacteriaemiae in a homogeneous cohort of AML patients. Neither fever incidence nor duration, septic shocks, early induction deaths and the rate of Gram-negative and Gram-positive cultures were different between the two groups. Our results underline the safety of a prophylaxis sparing approach that may reduce the emergence of drug-resistance Gram-negatives during the consolidation phases. Prospective studies are needed to confirm these data and to assess the impact of prophylaxis during consolidation phase.