Children with acute myeloid leukemia (AML) had a significantly lower incidence of certain bacterial infections if they received fluoroquinolone prophylaxis after chemotherapy, a small retrospective study showed.

Children treated with levofloxacin had more than a 50% reduction in the incidence ofStreptococcus viridans infections compared with children who received chemotherapy without antibiotic prophylaxis, but did not eliminate infection by the organism. Gram-negative infections also occurred significantly less often with antibiotic prophylaxis.

“Prophylactic use of levofloxacin reduces the incidence of strep viridans infections,” Ferdjallah said. “However, strep viridans resistance is common in patients receiving levofloxacin prophylaxis. Levofloxacin prophylaxis reduces gram negative infections during period of neutropenia.”

Other potential contributors to the lower incidence of infection could not be ruled out, she added. During implementation of antibiotic prophylaxis for patients with AML, a strategy to improve central line care also was implemented. Additionally, chlorhexidine baths and oral hygiene protocols were initiated.

Bacterial sepsis is a leading cause of morbidity and mortality in patients with AML. Contemporary studies have found microbiologically confirmed infections in one-third tothree-fourths of patients, associated with mortality of 6% to 11%.

Studies involving adults with cancer have suggested that fluoroquinolone prophylaxis can reduce infection-associated mortality. One study involving children with acute lymphoblastic leukemia showed an association between antibiotic prophylaxis and lower rates of hospitalization, intensive care admissions, and bacteremia.

The National Comprehensive Cancer Network suggests “consideration of fluoroquinolone prophylaxis” in patients with AML treated with chemotherapy, Ferdjallah said, and theInfectious Diseases Society of America specifically mentions consideration of levofloxacin prophylaxis, including consideration of use in high-risk pediatric patients.

In 2012, Children’s Healthcare of Atlanta implemented a policy of levofloxacin prophylaxis for all patients with newly diagnosed AML. Ferdjallah and colleagues examined the impact of the policy on bacterial infections during induction therapy and episodes of chemotherapy-induced neutropenia. Secondarily, they evaluated the incidence of fungal infections in patients receiving levofloxacin prophylaxis.

Investigators conducted a retrospective chart review of patients with newly diagnosed AML treated from September 2010 to September 2014, covering the 2 years before and after institution of the levofloxacin prophylaxis policy. The protocol stipulates initiation of levofloxacin on the first day of chemotherapy and continuation until cell count recovery.

They defined a bacterial infection as a positive blood, wound, or tissue culture and a fungal infection as a positive culture or CT findings in association with signs and symptoms consistent with infection.

The study population consisted of 39 patients who had a median age of 11 and age range of 15 months to 20 years. The patients received a cumulative total of 132 courses of cytarabine-containing chemotherapy: 80 courses with levofloxacin prophylaxis and 52 courses without levofloxacin.

Overall, gram-positive infections occurred during 23 (28.75%) courses of chemotherapy with levofloxacin prophylaxis and 20 (38.46%) courses of chemotherapy without prophylaxis, a difference that did not achieve statistical significance (P=0.213).

When Ferdjallah and colleagues analyzed infectious by S. viridans status, they found a significant reduction in S. viridans infections, from 28.85% during the period without levofloxacin prophylaxis to 12.5% with prophylaxis (P=0.024). Prophylaxis did not significantly affect the incidence of non-S. viridans infections (16.25% versus 9.6%,P=0.312).

Gram-positive organisms accounted for most of the chemotherapy-associated infections. Nonetheless, no patients who received levofloxacin prophylaxis developed a gram-negative infection as compared with five infections (9.62% of chemotherapy courses) in the no-prophylaxis group (P=0.024).

Levofloxacin prophylaxis was not associated with an increased risk of fungal infections, Ferdjallah said. Fungal infections occurred during two (2.5%) courses of chemotherapy in the prophylaxis group versus five (9.62%) courses in the no-prophylaxis group (P=0.112).

The total infection rate was substantially reduced with levofloxacin prophylaxis, but the difference did not achieve statistical significance (33.75% versus 50.0%, P=0.067).

Three patients died during induction or intensification of chemotherapy, including one infection-related death in a patient who did not receive levofloxacin prophylaxis.

During the discussion that followed her presentation, Ferdjallah acknowledged that a prospective randomized trial would be the optimal way to determine the value of fluoroquinolone prophylaxis during chemotherapy for AML. However, she questioned whether such a study would be feasible, given consistent evidence from observational studies in favor of prophylaxis.