Meropenem Versus Piperacillin-Tazobactam as Empiric Therapy for Febrile Neutropenia in Pediatric Oncology Patients

Despite advances in treatment of cancer with chemotherapy and supportive care, febrile neutropenia (FEN) is a common complication after chemotherapy with a mortality between 2% to 6%. FEN should be managed efficiently and empiric antimicrobial treatment should be started immediately (Santaloya et al., 2007) . FEN management has changed in the recent years and patients have been treated due to some risk factors. There are a few studies in pediatric age group to determine risk factors for febrile neutropenia (Santaloya et al., 2002; Hartel et al., 2007). Initial antibiotic treatment for FEN should have a wide spectrum, be bactericidal and have anti-pseudomonal activity. Institutional bacterial resistance patterns should also be used to guide selection of first-line antibiotics (Hughes et al., 2002; Koh and Pizzo, 2010). Initially, betalactam and aminoglycoside combination have been used in the empiric treatment of febrile neutropenia (Cometta et al., 1995; Aksoylar et al., 2004). Recently, monotherapy with broad spectrum and bactericidal antibiotics replaced combination therapy. Effective monotherapies


Introduction
Despite advances in treatment of cancer with chemotherapy and supportive care, febrile neutropenia (FEN) is a common complication after chemotherapy with a mortality between 2% to 6%.FEN should be managed efficiently and empiric antimicrobial treatment should be started immediately (Santaloya et al., 2007) .FEN management has changed in the recent years and patients have been treated due to some risk factors.There are a few studies in pediatric age group to determine risk factors for febrile neutropenia (Santaloya et al., 2002;Hartel et al., 2007).Initial antibiotic treatment for FEN should have a wide spectrum, be bactericidal and have anti-pseudomonal activity.Institutional bacterial resistance patterns should also be used to guide selection of first-line antibiotics (Hughes et al., 2002;Koh and Pizzo, 2010).Initially, betalactam and aminoglycoside combination have been used in the empiric treatment of febrile neutropenia (Cometta et al., 1995;Aksoylar et al., 2004).Recently, monotherapy with broad spectrum and bactericidal antibiotics replaced combination therapy.Effective monotherapies
Piperacillin/tazobactam (PIP/TAZ) is a beta-lactam/ beta-lactamase inhibitor combination that has a wide range of activity against Gram-positive, Gram-negative, and anaerobic pathogens (Jones et al., 1989).Meropenem is a member of carbapenems, possesses the broadest antibacterial spectrum of any class antibiotic and has the advantage of having activity against extended-spectrum beta-lactamase-(ESBL) producing organisms.Recently, a meta-analysis found that for initial treatment of neutropenic fever, PIP-TAZ resulted in lower all-cause mortality than other beta-lactam antibiotics, including carbapenems in adult cancer patients (Paul et al., 2006).Pediatric data about this issue are very limited.To our knowledge, there is no study comparing meropenem to PIP/TAZ in pediatric hemato-oncology patients with febrile neutropenia.The aim of this retrospective clinical study is to compare the efficiency and toxicity of PIP/TAZ and meropenem monotherapy for the empirical treatment of pediatric cancer patients with fever and neutropenia.

Patients
Between March 2008 and April 2011, all children with FEN who had been treated for hemato-oncological malignancies, who were <18 years of age were identified.Fever was defined as either a single axillary temperature of ≥38.3̊C or sustained temperature over 1h of ≥38.0̊C.Neutropenia was defined as an absolute neutrophil count (ANC) ≤500 cells/mm 3 or ≤1,000 cells/mm 3 , which was expected to be ≤500 cells/mm 3 within 24-48h (Hughes et al., 2002).Patients were identified more than once if they had a distinct episode of FEN and prior antibiotic treatment that had been completed at least 2 weeks earlier.Exclusion criteria were; presence of hypotension and multi-organ failure and intravenous antibiotic treatment within 48 hours of admission.Prophylactic antibiotics were not administered in any of the patients before or during antibiotic treatment.
Electronic and paper records were reviewed.Blood (both peripheral blood and central venous catheter (CVC) if present) and urine cultures, and cultures from any local site suspected to be infected were collected before antibiotic administration.The remission status, the presence of mucositis, the ANC, the duration of neutropenia and fever were recorded.

Classification of the febrile episodes
Infections were defined as fever of unknown origin (FUO) if the infection focus could not be defined, microbiologically documented infection (MDI, microorganism is isolated) and clinically documented infection (CDI, if typical signs of infection were found in physical examination despite no culture growth).

Antibiotic treatment
All patients received treatment on an inpatient basis.PIP/TAZ 360 mg/kg/day or meropenem 60mg/kg/day was started intravenously in three divided doses.Divisional policy was to evaluate the patients after 72 hr of treatment.If fever persisted >38.0̊C at 72 hr, and there was no microbiologically documented infection, amikacin at 15 mg/kg/day was added; if the patient still remained febrile at 96 hr, a glycopeptide (teicoplanin) at 10 mg/kg/day was added.In case of MDI and no clinical improvement, the antibiotic therapy was adjusted according to the antibiogram results.Empirical Amphotericin B at 3 mg/ kg/day was started in patients with persistent fever on the 7 th day of the febrile episode.Antibiotic therapy was continued until fever subsided and the neutrophil count was over 500 cells/mm 3 for 2 days.GCSF was used (5 µcg/kg/day, sc) as secondary prophylaxis when needed.Adverse events were recorded.

Evaluation of the treatment
Modification was defined as change in the initial empirical antimicrobial agent.The treatment was regarded as a success if fever and clinical signs of infection resolved without treatment modification.The treatment was regarded as a failure if another antibiotic or antifungal agent was added or the patient died due to infection during febrile neutropenia.

Statistical evaluation
Statistical analysis was performed using SPSS, version 11.5 (SPSS Inc., Chicago, IL).The data was evaluated using descriptive statistical methods.Statistical differences between study groups were assessed by chisquare test for categorical variables and student t-test for continuous variables.Two-tailed p values were used and p values of <0.05 were defined as significant.

Patient quality
During the study period, 284 episodes in 136 patients (51 female, 85 male; median age 60 months, range, 4-231 months) were documented.Table 1 shows the characteristics of the episodes.There was no statistical difference for sex, underlying disease, number of initial ANC, duration of febrile neutropenia, remission status, grade 3-4 mucositis and number of patients receiving GCSF between the two treatment groups.Patients in the meropenem group were younger than the patients in the PIP/TAZ group (p=0.04).

Characteristics of the febrile attacks
Ninety eight episodes were seen in patients with leukemia 186 episodes were in patients with solid tumors.The mean ANC for the whole group was 108±200 cells/ mm 3 .In 70% of episodes, the neutrophil count was under 100 cells/mm 3 .The median neutropenia duration was 7 days (range 0-80 days).The period of neutropenia was over 10 days in 20% of episodes.There was no difference between groups in terms of gender, remission status,  1).Nineteen percent of the episodes were documented microbiologically.70% of the infections were clinically documented and 30% of the episodes were fever of unknown origin (Table 2).One patient had rhinomaxillary mucormycosis, five patients had invasive aspergillosis infection of the lungs, three patients had CMV pneumonia and two had H1N1 pneumonia (data not shown).In 54% of the microbiologically documented episodes gram negative, in 41% gram positive microorganisms and in 5% fungi were isolated (Table 3).In 35 of the episodes patients had an indwelling catheter; four in the PIP/TAZ and thirty one in the meropenem group (data not shown).

Treatment and response to therapy
Outcome of treatment with two different regimens are presented in Table 4. Meropenem was used in 198 and PIP/TAZ was used in 86 of 284 episodes.There was no significant difference in duration of hospitalization between the two groups.No modification was done in 39% (n=112) of the episodes.Modifications were necessary in 61% of the episodes (172 episodes).The modification rate was not statistically different between the two groups (p=0.58).Success rate without modification was 68% in meropenem group and 32% in PIP/TAZ group (p=0.58).There were twenty two deaths in total during febrile neutropenic episodes and there was no statistical difference betweeen the two groups (p=0.87)(Table 4).The episodes in which patients had neutrophil count less than 100 cells/mm 3 and grade 3 or 4 mucositis, the modification rate was found to be higher (p<0.05)(Table 5).
No adverse effects due to treatment were observed.No patients were admitted with recurrent fever in the 10-day follow-up period.

Discussion
Infection is the main cause of mortality in neutropenic patients with cancer.Infections can be documented  microbiologically and clinically in 50% of the neutropenic febrile episodes.As the documentation of infection is difficult, broad spectrum antibiotics should be instituted as soon as possible (Rossi et al., 1996;Hahn et al., 1997;Hughes et al., 2002;Koh and Pizzo, 2010).Before the early introduction of empirical antibiotics, the mortality rate of Gram-negative infections, especially due to Pseudomonas aeruginosa, Escherichia coli and Klebsiella pneumoniae, was about 80%.Many studies suggest that antibiotics used in the empirical treatment of neutropenic patients should be bactericidal, broad-spectrum and have antipseudomonal activity (Rossi et al., 1996;Koh and Pizzo, 2010).Combinations of beta-lactam and aminoglycosides have been used in pediatric febrile neutropenia for many years (Cometta et al., 1995;Aksoylar et al., 2004).Several studies have shown that combination therapy was not superior to monotherapy.Paul et al. (2010) have shown that ceftazidime, cefepime, PIP/TAZ, and carbapenems were effective as monotherapy in their meta-analysis.The results were in agreement with recent studies including pediatric cancer patients, reporting that cefozopran, cefepime, meropenem,imipenem, sulperazon and piperacillin-tazobactam were effective and safe for empiric treatment of febrile neutropenic episodes (Agaoglu et al., 2001;Corapcioglu et al., 2006;Erbey et al., 2009;Uygun et al., 2009;Vural et al., 2010;Demir et al., 2011;Erbey et al., 2010;Ichikawa et al., 2011;Karaman et al., 2012).Vural et al. (2010) compared PIP/ TAZ versus imipenem in pediatric febrile neutropenia and reported that both antibiotics can be used safely as monotherapy.In a study with adult participants comparing PIP/TAZ with meropenem both antibiotics were found to be effective equally similar to our study (Oztoprak et al., 2010).
In adult patients febrile neutropenic episodes were stratified according to risk groups and there is a tendency to use monotherapy in low-risk groups.Although many studies have reported risk factors in febrile neutropenia in children, a universal scoring system for risk stratification is not available in children.Most important factors for risk groups are severity and duration of neutropenia, and presence of complications.Not being in remission, using high-dose chemotherapy protocols, the presence of severe mucositis, multiorgan failure shock,leukemia are also considered high-risk factors (Blot et al., 1997;Paesmans et al., 2000;Chindaprasirt et al., 2013).Recently, satisfying results with monotherapy in high-risk groups were also informed (Viscoli et al., 2006).In our study, all patients with hemato-oncological malignancies without multiorgan failure or shock received monotherapy regardless of other risk factors.In this study, leukemia constituted 35% of the febrile episodes and there was no statistical difference between the modification rate of solid tumors and leukemia (p=0.97).In 70% of episodes, neutrophil count was under 100/mm 3 , in 20% of them, duration of neutropenia was more than 10 days and in 37% there was severe mucositis which are usually accepted as high-risk factors.The modification ratio was reported to be 20%-50% in several studies and our results were slightly higher than the published work.Factors affecting modification were neutrophil count less than 100 cells/mm 3 and the presence of severe mucositis which is consistent with previous reports (p<0.05)(Uygun et al., 2009;Karaman et al., 2012).Duration of neutropenia and remission status did not have an influence on modification ratio (p>0.05).There was no statistical difference without modification in the meropenem and the PIP/TAZ group.
The microorganisms isolated in febrile neutropenic episodes have changed in the last three decades.In the last decade Gram positive pathogens have been isolated more frequently than Gram negative pathogens (Duncan et al., 2007;Paul et al., 2007).There is an increase of Gram positive pathogens reported recently from pediatric hematology and oncology centers in Turkey (Kebudi et al., 2005).The use of central venous catheters and severe mucositis due to treatment may be the cause for this change (Herwaldt et al., 1992;Koh and Pizzo, 2010).There is still a predominance of Gram negative pathogens where central venous catheters are not frequently used.In our center, central venous catheters are not used routinely.In this study, 19% of febrile episodes were documented microbiologically and Gram negative pathogens were isolated more often than Gram positive pathogens.
In conclusion, in this retrospective study monotherapy with meropenem or PIP/TAZ was found to be equally effective and safe for the initial treatment of febrile neutropenia.

Table 5 . Effects of Patients and Episodes Characteristics on Modification
*FEN: Febrile neutropenia; ANC: Absolute neutrophil count

Table 3 . Microbiologically Documented Infections in FEN Episodes
. Piperacillintazobactam plus amikacin versus ceftazidime plus amikacin as empiric therapy for fever in granulocytopenic patients with cancer.The international antimicrobial therapy cooperative group of the European organization for research and treatment of cancer.Antimicrob Agents Chemother, 39, 445-52.Corapcioglu F, Sarper N, Emine Z, et al (2006).Monotherapy with piperacillin/tazobactam versus cefepime as empirical therapy for febrile neutropenia in pediatric cancer patients: a randomized comparison.Pediatr Hematol Oncol, 23, 177-86.Demir HA, Kutluk T, Ceyhan M, et al (2011).Comparison of sulbactam/cefoperazone with carbapenems as empirical monotherapy for febrile neutropenic children with lymphoma and solid tumors.Pediatr Hematol Oncol, 28, 299-310.Duncan C, Chisholm JC, Freeman S, et al (2007).A prospective study of admissions for febrile neutropenia in secondary paediatric units in South East England.Pediatr Blood Cancer, 49, 678-81.Erbey F, Bayram I, Yilmaz S, Tanyeli A (2009).Imipenem in the treatment of febrile neutropenic children.Asian Pac J