The Surgical Infection Society guidelines on antimicrobial therapy for intra-abdominal infections: evidence for the recommendations.

Revised guidelines for the use of antimicrobial therapy in patients with intra-abdominal infections were recently developed by the Therapeutic Agents Committee of the Surgical Infection Society (Mazuski et al., Surg Infect 2002;3:161-173). These were based, insofar as possible, on evidence published over the past decade. The objective of this document is to describe the process by which the Committee identified and reviewed the published literature utilized to develop the recommendations and to summarize the results of those reviews. English-language articles published between 1990 and 2000 related to antimicrobial therapy for intra-abdominal infections were identified by a systematic MEDLINE search and an examination of references included in recent review articles. If current literature with regard to a specific issue was lacking, relevant articles published prior to 1990 were identified. All prospective randomized controlled trials, as well as other articles selected by the Committee, were evaluated individually and collectively. Data with regard to patient numbers, types of infections, and results of interventions were abstracted. Studies were categorized according to their design, and all included trials were graded according to quality. On the basis of this evidence, the Committee formulated recommendations for antimicrobial therapy for intra-abdominal infections and graded those recommendations. After receiving comments from invited reviewers and the general membership of the Society, the guidelines were finalized and submitted to the Council of the Surgical Infection Society for approval. The final recommendations related to the selection of patients needing therapeutic antimicrobials, acceptable antimicrobial regimens, duration of antimicrobial use, and the identification and treatment of higher-risk patients. Although numerous publications pertaining to these topics were identified, but nearly all of the prospective randomized controlled trials represented comparisons of different antimicrobial regimens for the treatment of intra-abdominal infections. A few prospective trials evaluated the need for therapeutic antimicrobial therapy in patients with peritoneal contamination following abdominal trauma. The quality of these prospective trials was highly variable. Many did not limit enrollment to patients with complicated intra-abdominal infections, lacked blinding of treatment assignment, did not provide a complete description of the criteria used to determine therapeutic success or failure, failed to identify the reasons why patients were excluded from analysis, or did not include an intention-to-treat analysis. For many issues, no prospective randomized controlled trials were encountered, and guidelines had to be formulated using evidence from studies with historical controls or uncontrolled data, or on the basis of expert opinion