An approach to the design and implementation of clinical trials of empirical antibiotic therapy in febrile and neutropenic cancer patients.

摘要: Abstract The results of many clinical trials on empirical therapy in febrile, neutropenic cancer patients cannot be readily transferred to the practice, because methodology is often flawed and definitions, study endpoints eligibility criteria differ from trial trial. This article critically reviews some issues related design implementation randomised antibiotic patients. Within definition phase III trials, two approaches co-exist, based trial's specific aims: “explanatory” approach “pragmatic” approach. usual “explicit” aim has been that comparing “efficacy” regimens. However, this term more used with reference antibacterial activity regimen under (explanatory aim) than indicate practical benifits it draws overall patient population treated for fever neutropenia (pragmatic aim). These meanings are taken as perfectly interchangeable, while, conversely, they completely distinct (though not independent) treatment effects. Most conducted recent years explanatory though explicitly so, but their have widely applied were pragmatic trials. In an appropriate endpoint success or failure (defined by laboratory data) among those affected infection which drug being given, while survival probably outcome variable, since designed assess febrile can obtain new treatment. Unfortunately, a difficulty assessing cause death and, especially, need very large sample sizes, might render problematic even large, multicentre groups. Both types intention treat analysis, especially crucial should differentiate cases was obtained through multiple modifications who did require any change. Obviously, implies no conclusion drawn about drugs number into account interpretation results, quality life cost evaluations. Information signs infection, age, underlying disease, concomitant administration other antibiotics entry clearly discussed defined. Finally, evaluation toxicity population, due existence toxigenic factors, including type infectious complication, chemotherapy radiotherapy use parental nutrition. All these effects tend ovelap, thus impairing investigator's ability detect drug-related side-effects.