Intraoperative interventions for preventing surgical site infection: an overview of Cochrane Reviews.

Surgical site infection (SSI) rates vary from 1% to 5% in the month following surgery. Due to the large number of surgical procedures conducted annually, the costs of these SSIs can be considerable in financial and social terms. Many interventions are used with the aim of reducing the risk of SSI in people undergoing surgery. These interventions can be broadly delivered at three stages: preoperatively, intraoperatively and postoperatively. The intraoperative interventions are largely focused on decontamination of skin using soap and antiseptics; the use of barriers to prevent movement of micro-organisms into incisions; and optimising the patient's own bodily functions to promote best recovery. Both decontamination and barrier methods can be aimed at people undergoing surgery and operating staff. Other interventions focused on SSI prevention may be aimed at the surgical environment and include methods of theatre cleansing and approaches to managing theatre traffic.

To present an overview of Cochrane Reviews of the effectiveness and safety of interventions, delivered during the intraoperative period, aimed at preventing SSIs in all populations undergoing surgery in an operating theatre.

Published Cochrane systematic reviews reporting the effectiveness of interventions delivered during the intraoperative period in terms of SSI prevention were eligible for inclusion in this overview. We also identified Cochrane protocols and title registrations for future inclusion into the overview. We searched the Cochrane Library on 01 July 2017. Two review authors independently screened search results and undertook data extraction and 'Risk of bias' and certainty assessment. We used the ROBIS (risk of bias in systematic reviews) tool to assess the quality of included reviews, and we used GRADE methods to assess the certainty of the evidence for each outcome. We summarised the characteristics of included reviews in the text and in additional tables.

We included 32 Cochrane Reviews in this overview: we judged 30 reviews as being at low risk of bias and two at unclear risk of bias. Thirteen reviews had not been updated in the past three years. Two reviews had no relevant data to extract. We extracted data from 30 reviews with 349 included trials, totaling 73,053 participants. Interventions assessed included gloving, use of disposable face masks, patient oxygenation protocols, use of skin antiseptics for hand washing and patient skin preparation, vaginal preparation, microbial sealants, methods of surgical incision, antibiotic prophylaxis and methods of skin closure. Overall, the GRADE certainty of evidence for outcomes was low or very low. Of the 77 comparisons providing evidence for the outcome of SSI, seven provided high- or moderate-certainty evidence, 39 provided low-certainty evidence and 31 very low-certainty evidence. Of the nine comparisons that provided evidence for the outcome of mortality, five provided low-certainty evidence and four very low-certainty evidence.There is high- or moderate-certainty evidence for the following outcomes for these intraoperative interventions. (1) Prophylactic intravenous antibiotics administered before caesarean incision reduce SSI risk compared with administration after cord clamping (10 trials, 5041 participants; risk ratio (RR) 0.59, 95% confidence interval (CI) 0.44 to 0.81; high-certainty evidence - assessed by review authors). (2) Preoperative antibiotics reduce SSI risk compared with placebo after breast cancer surgery (6 trials, 1708 participants; RR 0.74, 95% CI 0.56 to 0.98; high-certainty evidence - assessed by overview authors). (3) Antibiotic prophylaxis probably reduce SSI risk in caesarean sections compared with no antibiotics (82 relevant trials, 14,407 participants; RR 0.40, 95% CI 0.35 to 0.46; moderate-certainty evidence; downgraded once for risk of bias - assessed by review authors). (4) Antibiotic prophylaxis probably reduces SSI risk for hernia repair compared with placebo or no treatment (17 trials, 7843 participants; RR 0.67, 95% CI 0.54 to 0.84; moderate-certainty evidence; downgraded once for risk of bias - assessed by overview authors); (5) There is currently no clear difference in the risk of SSI between iodine-impregnated adhesive drapes compared with no adhesive drapes (2 trials, 1113 participants; RR 1.03, 95% CI 0.66 to 1.60; moderate-certainty evidence; downgraded once for imprecision - assessed by review authors); (6) There is currently no clear difference in SSI risk between short-term compared with long-term duration antibiotics in colorectal surgery (7 trials; 1484 participants; RR 1.05 95% CI 0.78 to 1.40; moderate-certainty evidence; downgraded once for imprecision - assessed by overview authors). There was only one comparison showing negative effects associated with the intervention: adhesive drapes increase the risk of SSI compared with no drapes (5 trials; 3082 participants; RR 1.23, 95% CI 1.02 to 1.48; high-certainty evidence - rated by review authors).

This overview provides the most up-to-date evidence on use of intraoperative treatments for the prevention of SSIs from all currently published Cochrane Reviews. There is evidence that some interventions are useful in reducing SSI risk for people undergoing surgery, such as antibiotic prophylaxis for caesarean section and hernia repair, and also the timing of prophylactic intravenous antibiotics administered before caesarean incision. Also, there is evidence that adhesive drapes increase SSI risk. Evidence for the many other treatment choices is largely of low or very low certainty and no quality-of-life or cost-effectiveness data were reported. Future trials should elucidate the relative effects of some treatments. These studies should focus on increasing participant numbers, using robust methodology and being of sufficient duration to adequately assess SSI. Assessment of other outcomes such as mortality might also be investigated as part of non-experimental prospective follow-up of people with SSI of different severity, so the risk of death for different subgroups can be better understood.