Is antibiotic prophylaxis in surgery a generally effective intervention? Testing a generic hypothesis over a set of meta-analyses.

CONTEXT: While the main focus of a meta-analysis is often to assess the effectiveness of a particular intervention in managing or curing a specific condition, there exists a substantial amount of information within published systematic reviews that could be used to assess the validity of a generic hypothesis about the effectiveness of an intervention across a range of different but related conditions.
OBJECTIVE: To systematically search for and then collate the results of meta-analyses for the effectiveness of antibiotic prophylaxis in preventing postoperative wound infection across various types of surgery in order to assess generic hypotheses about the effectiveness of this intervention in surgery as a whole. With the relative risk of wound infection used as the measure of clinical effectiveness, the hypotheses to be assessed were first that antibiotic prophylaxis would be an effective intervention for preventing wound infection over a broad range of different surgical procedures, and second that there would be a substantial difference in the effectiveness of antibiotic prophylaxis between "clean" and "contaminated" surgical procedures. DATA SOURCES: Medline and the Cochrane Database of Systematic Reviews. STUDY SELECTION: Eligible meta-analyses were meta-analyses published between 1990 and 2006 of randomized controlled trials that looked at the effectiveness of prophylactic antibiotics versus no antibiotic or placebo in preventing postoperative wound infections. DATA EXTRACTION: Independent data extraction by multiple observers.
RESULTS: The first hypothesis was strongly supported by the data as evidenced by the fact that the estimates of the relative risk of infection for the 23 types of surgery that were included in the study were all less than 1. However, there was no real evidence supporting the second hypothesis that the relative risk of wound infection would substantially vary over different levels of surgery cleanliness.
CONCLUSION: : As well as antibiotic prophylaxis being a generally effective intervention for preventing postoperative wound infection, the level of this effectiveness would appear to be reasonably independent of what type of surgery is being considered. Therefore, the general prevailing attitude that antibiotic prophylaxis should be assumed to be ineffective unless its effectiveness has been experimentally proven beyond doubt for the specific type of surgery being considered, perhaps should be revised. In particular, perhaps a sensible philosophy would be to assume that antibiotic prophylaxis is effective in reducing the risk of wound infection for all types of surgery, even ones where no clinical trial data exists and make exceptions to this rule if, for certain types of surgery, it can be proved to the contrary.