Garnica, M., Nouer, S.A., Pellegrino, F.L., Moreira, B.M., Maiolino, A., & Nucci, M. (2013). Ciprofloxacin prophylaxis in high risk neutropenic patients: Effects on outcomes, antimicrobial therapy and resistance. BMC Infectious Diseases, 13, 356.

To evaluate the impact of quinolone prophylaxis during neutropenia on outcomes including resistance rates and hospital admissions

Researchers compared data from hospitalizations during two time periods, representing an intervention group and a control group. The intervention group included patients who received quinolone prophylaxis from 2006–2008. Prophylaxis consisted of 500 mg oral ciprofloxacin twice a day or 200 mg IV twice a day if oral medication was not tolerated. The control group included patients who received no antibiotic prophylaxis during 2005. In the event of a fever, patients in both groups began IV cefepime. If patients had a history of a cefepime-resistant gram-negative infection, they were treated with a carbapenem instead. Analysis of demographics and clinical outcomes, including occurrence of fever, duration of empirical antibiotic therapy, duration of hospitalization, and quinolone resistance, were conducted using SPSS software. Chi-square tests and Mann-Whitney tests were used for categorical and continuous variables, respectively. To evaluate patterns of resistance, data from patients outside the intervention cohort, but also hospitalized from 2006–2008, also were analyzed for resistance.

• N = 220 patients (141 in ciprofloxacin prophylaxis group, 79 in control group). Some patients were included more than once if more than 30 days passed between episodes of neutropenia. Thus, in 220 patients, 329 episodes of neutropenia occurred (219 in ciprofloxacin prophylaxis group, 110 in control group).
• AGE: Ciprofloxacin: mean age = 40 years (age range: 14–82 years); Control: mean age = 41 years (age range: 12–66 years)
• MALES: 60% in ciprofloxacin, 66% in control; FEMALES: 40% in ciprofloxacin, 34% in control
• KEY DISEASE CHARACTERISTICS: All patients were receiving intensive chemotherapy for a hematologic malignancy or undergoing hematopoietic cell transplantation (HCT). Diseases for ciprofloxacin and control groups, respectively, are primarily blood cancers, including leukemia (44%, 42%), lymphoma (26%, 21%), and multiple myeloma (26%, 28%). Other diseases represented include germ cell tumors and aplastic anemia.
• OTHER KEY SAMPLE CHARACTERISTICS: More than half of the episodes in each group included patients undergoing autologous or allogeneic HCT. Nearly two-thirds of the episodes in each group included patients with a central venous catheter.

• SITE: Single site   
• SETTING TYPE: Inpatient   
• LOCATION: Brazil

• PHASE OF CARE: Active antitumor treatment

• A retrospective, observational case-control study

For their methods, the researchers defined neutropenia as an absolute neutrophil count (ANC) less than 500/mm³. They defined bone marrow recovery as at least two consecutive days with ANC greater than 500/mm³. All blood cultures were processed using a BacT/ALERT^® system. Susceptibility tests were completed using the VITEK^® automated system. Other laboratory tests on stored bacterial cultures were used to evaluate susceptibility. These included, but were not limited to, disk diffusion, minimal inhibitory concentration, and polymerase chain reaction. Statistical tests, including Chi-square and Mann-Whitney, were completed using SPSS software.

Groups were similar in age and gender. The intervention group was statistically different from the control group in that they experienced slightly shorter periods of neutropenia (9 versus 11 days, p = 0.02), hospitalization (22 versus 24 days, p = 0.002), and antibiotic therapy (8 versus 11 days, p < 0.001); fewer febrile episodes (73 versus 93%, p < 0.001), decreased incidence of any grade mucositis (52% versus 70%, p = 0.003), and bacteremia (22% versus 33%, p = 0.04); and increased use of carbapenem (36% versus 14%, p < 0.001). The intervention group had a higher rate of quinolone-resistant bacteremia (6.77 versus 3.02 per 1,000 patients-day, p = 0.03). Quinolone-resistant enterobacteria was noted in the intervention group and patients outside the intervention cohort but hospitalized during the same time. The rate of extended spectrum beta-lactamase (ESBL)-producing enterobacteria was not significantly increased in the intervention group (0.38 in the control group versus 1.27 in the ciprofloxacin group, p = 0.26).

This study identifies benefits of quinolone prophylaxis in high-risk patients (i.e., patients with hematologic malignancies undergoing chemotherapy with an expected duration of neutropenia longer than seven days, or patients undergoing HCT), including decreased incidence of fever, bacteremia, duration of neutropenia, and length of hospitalization. Risks include an increased incidence of quinolone resistance and bacteremia because of ESBL-producing enterobacteria for the patient and hospital unit.

• Risk of bias (no random assignment)
• Other limitations/explanation: The two cohorts being compared were historical and in different time periods. The authors report no changes in hospital policy or procedures between the two time points; other differences not measured may include different staff or training. Finally, as the data were historical, not all patients included in the cohorts had bacterial samples recovered and saved for further analysis.

Quinolone prophylaxis can reduce the incidence of fever, bacteremia, duration of neutropenia, hospitalization, and duration of antibiotic therapy for select high-risk patients. Nurses should understand these benefits and the risk of quinolone resistance for individual patients in the surrounding hospital unit.