186 Letters to the Editor / International Journal of Antimicrobial Agents 38 (2011) 177–189 Antimicrob Agents 2011, doi:10.1016/j.ijantimicag.2010.11.029 [Epub ahead of print]. [2] Towne TG, Lewis JS, Echevarria K. Efficacy and safety of cefepime. Lancet Infect Dis 2009;9:4–6, author’s reply 6–7. [3] Tamura K, Matsuoka H, Tsukada J, Masuda M, Ikeda S, Matsuishi E, et al. Cefepime or carbapenem treatment for febrile neutropenia as a single agent is as effective as a combination of 4th-generation cephalosporin + aminoglycosides: compar- ative study. Am J Hematol 2002;71:248–55. [4] Espino-Hernández M, Martínez-Motas I, Couto-Ramos MJ, Zuazo-Silva JL. Effectiveness of a combined antimicrobial treatment by means of the draughtboard. In vitro and in vivo assessment. Rev Cubana Farm [online] 2009;43, http://scielo.sld.cu/scielo.php?script=sci arttext&pid=S0034- 75152009000200006&lng=es&nrm=iso [accessed 9 March 2011]. María Espino-Hernández Latin American School of Medicine, Microbiology and Parasitology Department, Carretera Panamericana Km 3½, Santa Fe, Playa, La Habana, Cuba, PC 19108 E-mail addresses: mespino@elacm.sld.cu, mespino@infomed.sld.cu 15 March 2011 doi:10.1016/j.ijantimicag.2011.04.004 On the clinical implications of -lactam–aminoglycoside synergism: author’s reply Sir, Espino-Hernández points to ‘the diversity of clinical scenar- ios analysed and the heterogeneity of the groups’. We rather believe that the diversity of clinical scenarios analysed show- ing the same homogeneous trend in favour of monotherapy strengthens our conclusions. All analyses are presented by patient subgroups. We initiated the current review on the background of several past systematic reviews that showed no benefit of - lactam–aminoglycoside combinations in specific patient subgroups [1–3]. We increased the power of the analysis to show a difference in the current systematic review by looking at the totality of the evidence, but did not find an advantage to combination therapy. Treatment failure for a child with cystic fibrosis is different from treatment failure in sepsis, but compiling the relative effect of com- bination versus monotherapy in these different scenarios is a valid approach to compare their effects. Similarly, the consistency of the results across many years supports the finding that combination therapy offers no clinical benefit. The fact that the results of past and present trials are similar, given major differences in supportive treatment and bacterial resistance, strengths the conclusion of this analysis. All-cause mortality is an essential outcome to assess in ran- domised controlled trials of infection management because it is the only outcome that can be objectively assessed and the purpose of patient care. Certainly deaths may not be caused by infection amongst febrile neutropenic cancer patients (although 30-day mor- tality is usually related to the infectious episode). The purpose of randomisation is to ensure that baseline risk factors for death and deaths unrelated to the effects of the intervention are equally dis- tributed between the study groups. We agree that studies correlating in vitro findings with clinical outcomes might be helpful. Espino-Hernández et al. [4] conducted an interesting study assessing the association between in vitro interactions of -lactam–aminoglycoside combinations and treat- ment failure. They showed that in vitro synergism predicted treatment success, whilst antagonism predicted treatment failure. This might explain the results of clinical studies whereby the ben- efit of -lactam–aminoglycoside combinations was offset by harm of antagonistic interactions between specific antibiotics and bac- teria. However, it is difficult to draw firm conclusions from this study, since antagonism was associated with resistance to the two antibiotics (which probably led to treatment modification), and treatment failure was defined as clinical failure or the need to change antibiotic treatment. Perhaps associating synergistic inter- actions in specific patients with more robust clinical outcomes will improve our understanding of the effects of antibiotic combina- tions and will provide a path for better treatment of patients with sepsis. Funding: No funding sources. Competing interests: None declared. Ethical approval: Not required. References [1] Elphick HE, Tan A. Single versus combination intravenous antibiotic therapy for people with cystic fibrosis. Cochrane Database Syst Rev 2005:CD002007. [2] Paul M, Benuri-Silbiger I, Soares-Weiser K, Leibovici L.  Lactam monotherapy versus  lactam–aminoglycoside combination therapy for sepsis in immuno- competent patients: systematic review and meta-analysis of randomised trials. BMJ 2004;328:668. [3] Paul M, Soares-Weiser K, Leibovici L. -Lactam monotherapy versus  lactam–aminoglycoside combination therapy for fever with neutropenia: sys- tematic review and meta-analysis. BMJ 2003;326:1111. [4] Espino-Hernández M, Martínez-Motas I, Couto-Ramos MJ, Zuazo-Silva JL. Effectiveness of a combined antimicrobial treatment by means of the draughtboard. In vitro and in vivo assessment. Rev Cubana Farm [online] 2009;43, http://scielo.sld.cu/scielo.php?script=sci arttext&pid=S0034- 75152009000200006&lng=es&nrm=iso [accessed 18 April 2011]. Ronit Marcus * Leonard Leibovici Medicine E, Rabin Medical Center, Beilinson Hospital and Sackler Faculty of Medicine, Tel Aviv University, Israel Mical Paul Unit of Infectious Diseases, Rabin Medical Center, Beilinson Hospital and Sackler Faculty of Medicine, Tel Aviv University, Israel * Corresponding author. Tel.: +972 3 937 6501; fax: +972 3 937 6512. E-mail address: gil.ronit@gmail.com (R. Marcus) 7 April 2011 doi:10.1016/j.ijantimicag.2011.04.005 Superior in vitro activity of ertapenem and piperacillin/ tazobactam against recent clinical isolates of Proteus mirabilis from intra-abdominal infections (SMART 2009–2010) Sir, Intra-abdominal infections (IAIs) represent some of the most frequently encountered nosocomial infections in the healthcare setting and are mostly caused by Gram-negative bacilli (GNB) [1]. Proteus spp., notably Proteus mirabilis, comprise an important part of the aetiology of such infections. A wide range of individ- ual antimicrobial agents and combinations of agents is available and recommended for use in the treatment of IAIs, including car- bapenems, piperacillin/tazobactam (PTZ), and cephalosporins or quinolones used in combination with metronidazole, although resistance to these antibiotics is becoming more widespread [1]. In the current study, we report the superior activity of ertapenem and PTZ against P. mirabilis intra-abdominal isolates as part of the Study for Monitoring Antimicrobial Resistance Trends (SMART), which monitors the susceptibility of GNB from IAIs to ertapenem and comparators. A total of 467 isolates of P. mirabilis were studied. All organisms were deemed clinically significant by local participant criteria. Iso- late inclusion was independent of medical history, antimicrobial use, age or gender. All study sites identified each isolate using local