pediatric leukemia patients utilizing levofloxacin
STUDY BACKGROUND
Bacterial infection-related morbidity and mortality are high in children treated for acute myeloid leukemia (AML) and relapsed acute lymphoblastic leukemia (rALL). Bacteremia not only contributes to mortality but also prolongs hospitalization, delays chemotherapy, decreases the quality of life, and requires the administration of costly and often toxic antimicrobial drugs. The clinical practice guidelines of the American Society of Clinical Oncology and the Infectious Diseases Society of America Clinical Practice Guidelines recommended that antibiotic prophylaxis with a broad-spectrum fluoroquinolone, either ciprofloxacin or levofloxacin, should be prescribed for adult patients who are prone to prolonged neutropenia. A retrospective analysis, from Saudi Arabia, reported reduced bloodstream infection (BSI) and other infections with ciprofloxacin prophylaxis during delayed intensification.
Although routine in adults, antibacterial prophylaxis is controversial in pediatrics because data supporting efficacy and safety are sparse. There are also concerns about musculoskeletal toxicities, antibiotic resistance and Clostridioides difficile infection. In a multicenter, randomized trial of levofloxacin prophylaxis, Alexander et al. reported a significant reduction in bacterial infection or sepsis in children with AML an rALL. Surprisingly, there was no significant increase in the incidence of multi-drug resistance organisms (MDRO) or C. difficile infection.
This quality improvement study evaluates the impact of a new levofloxacin prophylaxis practice guideline at the Texas Children’s Hospital (TCH) Cancer Center.
METHODS
Patients
Patients aged 6 months to 21 years, diagnosed with AML or rALL at TCH Cancer Center were enrolled. Clinical presentations, demographic features, and microbiological findings were retrospectively reviewed. Exclusion criteria include patients with a prior history of i) hematopoietic stem cell transplant and ii) anaphylaxis to levofloxacin.
Practice guideline
A practice guideline based on the COG protocol ACCL0934 was presented to deliver high-quality care services and to improve health outcomes. Prophylactic treatment with levofloxacin was added to prevent severe neutropenia in AML and rALL patients receiving intensive chemotherapy. Clinical management of fever and neutropenia or any other diagnosed infection was per the institutional evidence-based guideline.
Levofloxacin was discontinued when children were on empiric antibiotics and resumed when the course of antibiotics was completed. Levofloxacin prophylaxis was then discontinued when the patient’s absolute neutrophil count (ANC) was > 200/μl post-nadir and rising.
Evaluation and outcomes
This retrospective cohort study spanned over 4 years from March 1, 2017, to February 28, 2021, consisting of 2 years pre- and 2 years post-implementation. The primary outcomes were central line-associated bloodstream infection (CLABSI) and bloodstream infection (BSI). The BSI events were further categorized based on microorganisms isolated (Gram-negative rods/Streptococcus viridans group. Secondary outcomes included cultures positive for multi-drug-resistant organisms (MDROs), rate of C. difficile infection, blood cultures positive for microbes susceptible to levofloxacin, transfers to ICU, death, and death due to bacterial infection.
Laboratory analysis
Bacteremia was detected using VersaTREK, an Automated Microbial Detection System. Antimicrobial susceptibility test was determined using agar plates and E-TEST strips. Levofloxacin susceptibility of microbes was determined and the results are interpreted qualitatively as intermediate (4 μg/ml) and resistant (8 μg/ml). Stool cultures were obtained from patients with C. difficile infection. Cefepime monotherapy was used for the empirical treatment of fever in patients. Metronidazole was used for intra-abdominal infections and vancomycin for patients with skin infections or a history of Staphylococcus aureus infection. Patients with AML or rALL were not given alternate antibiotic prophylactic regimens. Empirical therapy for Pneumocystis jiroveci pneumonia (PJP) consisted of trimethoprim-sulfamethoxazole or nebulized pentamidine and voriconazole or posaconazole.
Statistical analysis
Patient demographics, underlying diagnosis, days observation, and days of severe neutropenia of pre-and post-implementation period were compared by Fisher’s exact test. The distribution and susceptibility patterns of predominant microbes in the pre-and post-implementation period were analyzed. Bacteremia events due to Gram-negative rods and S. viridans were also analyzed. Total days of levofloxacin prophylaxis for each patient were calculated to check if the cumulative days of levofloxacin prophylaxis correlated with any of the observed outcomes of this study. All the data were performed using STATA 11 software.
RESULTS
- 63 and 72 patients met inclusion criteria for pre- and post-implementation cohorts, respectively. 65 of 72 patients in cohort 2 received levofloxacin prophylaxis.
- Demographics (median age, gender, race, and ethnicity) were similar between groups.
- The pre implementation cohort had a larger proportion of patients with rALL (n = 30; 47.6%); patients with AML were predominant (n = 47; 65.3%) in the post implementation cohort.
- The median observation period for the pre-and post implementation period was 127 and 173 days, respectively.
- Severe neutropenia was reported in both cohorts with a median of 31 vs 47 days in pre-and post-implementation groups, respectively.
Microbiologically confirmed bacteremia
Bacteremia events – Present on Admission (POA) and healthcare-associated NHSN CLABSI – were evaluated in both cohorts. There were 60 BSI events in the pre-groups and 49 in the post-group (P = 0.1). Bacteremia due to Gram-negative rods (RR 0.037; p < 0.001) and NHSN CLABSI (RR 0.62; p =0.01) were significantly low. However, there was an increase in BSI due to non-susceptible Gram-negative rods (RR 3.38; P < 0.001). Additionally, these microbes remained susceptible to extended-spectrum cephalosporins, carbapenems, and aminoglycosides. The incidences of diarrhea due to C. difficile and MDRO were similar in both groups.
Death during levofloxacin prophylaxis was significantly low (RR 0.58; p = 0.04) in the cohort but deaths due to bacterial infection were not statistically significant between the two cohorts (RR 0.38; P = 0.63).
DISCUSSION
The current study examines real-world clinical and microbiologic outcomes of prophylaxis with levofloxacin, a broad-spectrum fluoroquinolone, in pediatric patients treated for AML or rALL. This study has several strengths related to the institutional setting and study design. Compared to previous studies, this study had a longer observation period, extending beyond two chemotherapy cycles. The authors found that the cumulative duration of severe neutropenia was longer in the postintervention group.
The observed reduction in BSI due to Gram-negative rods and NHSN CLABSIs could have also been due to other CLABSI- prevention measures. Similarly, the authors observed that early detection and management of sepsis may have contributed to the fewer BSI mortality in both cohorts.
Previous studies on antibacterial resistance after levofloxacin prophylaxis have reported mixed results. Some authors even reported increased cross-resistance after extended use of antibiotic prophylaxis. Although there was no significant increase in levofloxacin-resistant infection, ten patients did contract bacteremia. The authors speculated that colonization might have been present before prophylaxis. Interestingly, there was no reduction in sepsis due to Gram-negative viridans group Streptococci.
The authors noted several study limitations, including the single-site randomized nature of the study. The findings suggest the need to equate and balance the benefits of levofloxacin prophylaxis with antibiotic resistance patterns.
Oncology Related Tools
Other
Latest Research
