Patients with acute myeloid leukemia (AML) who receive intensive chemotherapy regimens containing high-dose cytarabine are at high risk for bacterial sepsis due to prolonged neutropenia and severe mucositis. To determine if the use of prophylactic antibiotics during periods of neutropenia reduced bacterial infections and viridans streptococcal infections (VSI) in particular, we reviewed the charts of 66 AML patients who were treated on our front line protocol (AML02) from October 2002 to June 2006. During this time, several consecutive prophylactic antibiotic strategies were implemented for use after intensive myelosuppressive therapy. All patients also received prophylactic antifungal therapy. Overall, patients who received any prophylactic antibiotic regimen had an 86% (95% CI: 66–95%) reduction in VSI (P<0.001) and a 76% (95% CI: 62–85%) reduction in all bacterial infections (P<0.001), compared to those who did not receive prophylactic antibiotics. Oral cephalosporins were used initially and were found to be ineffective at preventing overall bacterial infections (P=0.97) or VSI (P=0.70). Intravenous cefepime was then implemented and completely prevented VSI. Furthermore, there was an 84% (95% CI: 52–94%) reduction in all bacterial infections (P<0.001). However, because we observed the emergence of life-threatening resistant gram-negative infections in two patients receiving cefepime, we instituted the prophylactic use of intravenous vancomycin with oral ciprofloxacin. The use of a vancomycin regimen resulted in an 84% (95% CI: 66–92%) decrease in total bacterial infections (P<0.001) and a 97% (95 CI: 78–100%) decrease in VSI (P<0.001). Importantly, fungal infection rates were similar in patients who received prophylactic antibiotics compared to those who did not (0.9 vs. 1.0 infection per 1000 patient days). Patients who received prophylactic antibiotics had shorter lengths of hospital stay (LOS), with a median (range) of 7 days (0–39) vs. 15 days (0–56) for each course of chemotherapy. After adjusting for course of therapy, this represented a 38% (95% CI: 24–50%) reduction in total LOS. Patients who received a vancomycin regimen had a 46% (95% CI: 28–59%) reduction in LOS (P<0.001) and those who received intravenous cefepime a 31% (95% CI: 9–47%) reduction (P=0.008) compared to patients who did not receive prophylactic antibiotics. Overall, there was only one death from septicemia; this occurred in a patient who was not receiving prophylactic antibiotics. In conclusion, this study suggests that supportive care with antibiotic prophylaxis during periods of neutropenia may dramatically decrease the incidence of septicemia and length of hospitalization. While intravenous cefepime was found to be effective prophylaxis for VSI and other bacterial infections, we have avoided its use because of concern for emerging resistance to our front-line empiric therapy for fever and neutropenia. Thus, intravenous vancomycin with oral ciprofloxacin is our prophylactic antibiotic regimen of choice. We did not administer ciprofloxacin alone, as this has previously been shown to increase the risk of infections with gram positive organisms. Further research is needed to determine the most effective antibiotic regimen for prevention of bacterial infection in pediatric patients with AML.
Read full abstract