Background: The prevention of surgical site infection guidelines issued by the Centers for Disease Control and Prevention (CDC) recently recommended that only a single dose of preoperative antibiotics be administered to patients undergoing clean-contaminated procedures based on data from a variety of surgical disciplines. For orthopaedic procedures, where postoperative infections can have significant consequences, the existing evidence for this recommendation is widely debated.
Questions/purposes: Is there a difference in postoperative infection risk when utilizing a single dose of preoperative antibiotics compared with multiple doses of perioperative antibiotics for orthopaedic procedures where implants are placed?
Methods: MEDLINE, EMBASE, Google Scholar, and Cochrane were systematically reviewed for randomized controlled trials (RCTs) of a single dose of preoperative antibiotics compared with pre- and postoperative prophylaxis from 1980 to 2017 for all orthopaedic procedures where implants were being placed. Infection (both superficial and deep) as a primary outcome through all available followup was required for inclusion. Fourteen RCTs detailing 9691 orthopaedic procedures were included for analysis, including seven arthroplasty, one spine, and six general orthopaedic trials (two specific to hip fracture fixation). Pooled infection outcomes were analyzed with random-effects modeling in light of study heterogeneity. Bias was evaluated using the Cochrane risk of bias tool as well as a funnel plot for publication bias, and quality of evidence was evaluated using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach. Bias was largely uncertain; however, a high risk of bias was noted in four studies. No significant overall publication bias was noted. The quality of evidence was determined to be very low based on the GRADE tool, downgraded based on risk of bias, inconsistency, and imprecision. Despite the quality of evidence, the data were pooled in light of the current recommendations from the CDC to critically evaluate the recommendation that a single dose of antibiotics be utilized.
Results: There were no differences in infection risk between single- versus multiple-dose groups (single: 83 of 4263 [2%], multiple: 101 of 5428 [2%]; odds ratio, 0.92 [95% confidence interval, 0.56-1.51]; p = 0.740, I = 36% for statistical heterogeneity).
Conclusions: There is no difference in infection risk between a single dose and multiple doses of perioperative antibiotics for orthopaedic procedures where implants are utilized, consistent with recent recommendations. However, the quality of evidence for orthopaedic procedures is low, and a randomized study with a sufficient sample size is needed to examine the issue before universal adoption of a single antibiotic dose.
Level of evidence: Level I, therapeutic study.