Updated Recommendations for Control of Surgical Site Infections

J. Wesley Alexander, MD, ScD; Joseph S. Solomkin, MD; Michael J. Edwards, MD


Annals of Surgery. 2011;253(6):1082-1093. 

In This Article

Prophylactic Topical Antimicrobials

The use of alcohol (wine) to prevent infections in wounds was described by Hippocrates in the fourth and fifth century BC and later documented in biblical writings (Luke 10:33-34). The fact that any benefit was caused by an antimicrobial effect of alcohol was, of course, not known. It was not until Listerian times that other antimicrobial agents, such as carbolic acid, were discovered to have benefits in protecting against wound infection by killing bacteria in contaminated wounds. With the introduction of systemic antibiotics, it was quickly reasoned that these same antibiotics might be effective when applied topically in wounds. However, the first prospective randomized studies were not done until 1956 when it was shown that irrigation of the operative site with tetracycline reduced the incidence of SSI after appendectomy from 8.1 to 1.2% (SDC-101). By 1977, there were 11 randomized clinical trials and 6 retrospective studies, which supported the clinical effectiveness of topical antibiotics (SDC-102). Neomycin, ampicillin, and tetracycline were the drugs most commonly used. Several animal studies have shown that application of antibiotics into the incision will significantly reduce the incidence of infection to the same extent as systemic prophylaxis, but combined systemic and topical applications provided no additional benefit when the same antibiotics were used (SDC-103-105). In clinical studies, topical irrigation with an antibiotic is also more effective when the type of topical antibiotic is different from the systemic antibiotic being given, with evidence that they are additive (SDC-106-113). In a recent study in patients undergoing inguinal herniorrhaphy, topical gentamicin was equivalent to systemic gentamicin in preventing wound infections (SDC-114). In another study in which cefazolin was measured in both serum and wound drain fluid after intravenous or topical application, concentrations of cefazolin in the wound drain fluid were over 4,000 μg/mL, which remained high for 24 hours after topical application (SDC-115). When ampicillin was given beneath the fascia and into the subcutaneous space during wound closure, initial wound fluid concentrations exceeded 1000 μg/mL and were 73 μg/mL, and 14 μg/mL after 14 and 20 hours, respectively (SDC-116).

The timing of topical antibiotics is a crucial factor (SDC-117). There were no deep 30 day SSI when irrigation of the surgical site was performed by irrigation with a needle and a syringe containing kanamycin with or without cephalothin on an average of every 4.7 minutes throughout the operative procedures compared to 0.73% without antibiotic lavage.[8] In a later randomized double blind study of GI and biliary procedures by the same group, 3% of patients with antibiotic irrigation developed infections compared to 9% in the control group (SDC-118). Use of a spray of a solution of cephalothin and gentamicin into the surgical site several times during cardiac surgery resulted in no deep SSI in 502 patients undergoing cardiopulmonary bypass.[9] In another study, irrigations with penicillin in 368 clean general surgical procedures reduced the incidence of postoperative SSI from 3.5% to 0.27% (SDC-119). In a more recent study, kanamycin was infused into the subcutaneous wound space by means of a catheter, allowing it to dwell for 2 hours before extraction of the fluid by activating the closed suction device to which it was attached.[10] Of 772 morbidly obese patients undergoing initial gastric bypass, there was only one primary deep SSI (0.13%) and two secondary deep infections (after spontaneous drainage of a seroma), totaling 0.39% for any deep infection.

With the demonstration that topical antibiotics could have a significant effect on the development of wound infections, there has been interest in developing techniques in which the antibiotics could be released over a longer period of time. Experimental studies have shown that the use of carriers such as polymethylmetacrylate (SDC-120), polyglycolic acid (SDC-121); glyceryl monostearate (SDC-122), poly-(dl-lactide-co-glycolide) (SDC-123), glycerylmono-stearate (SDC-124), and PMMA could be used as carriers for a variety of antibiotics including cefazolin (SDC-122-123), gentamicin (SDC-120, SDC-125), tobramycin (SDC-120), minocycline (SDC-121), and vancomycin (SDC-124). All of the animal models showed that the antibiotic-impregnated implants reduced bacterial counts in contaminated wounds and significantly protected against infections which were independent of and additive to systemic antibiotics (SDC-120-121, SDC-125). Significantly higher concentrations of the antibiotics were present at the location of the test sites. These or similar products have now been evaluated in a variety of clinical situations. An observational study in 1993 used tobramycin-impregnated PMMA beads to successfully prevent infections in compound fractures (SDC-126-127). Gentamicin containing collagen sponges were placed randomly in 221 colorectal surgery patients, and this was associated with a significant reduction in the rate of infection of the wounds (18.4%–5.6%) (SDC-128). Collagen-gentamicin implanted sponges were evaluated for their effect in preventing sternal wound infections in a study of 2000 cardiac patients randomized to treatment and control arms (SDC-129). The incidence of wound infection was 4.3% in the treatment group and 9.0% in the control group (relative risk, RR = 0.47).

If topical antibiotics are effective in controlling wound infections, it seems reasonable that irrigation of the wound with other antimicrobials would also be effective. Several of these, such as povidone iodine solutions and chlorhexidine have been studied extensively but, in general, their uses have not been effective or have actually been deleterious (SDC-130-134).

As an extension of the use of topical antimicrobials to prevent infection, mupirocen (pseudomonic acid) has been shown to be as much as 97% effective in reducing nasal carriage of S. aureus (SDC-135). A recent analysis including the Cochrane Data Base cited 9 randomized clinical trials involving 3396 participants (SDC-136). When mupirocen was compared to either placebo or no treatment, there was a reduction of S. aureus infections associated with intranasal mupirocen (RR = 0.55). However, the infection rate caused by other types of bacteria was significantly higher in patients treated with mupirocen (RR = 1.38).


Topical antibiotics are clearly effective in reducing wound infections and may be as effective as the use of systemic antibiotics. The combined use of systemic antibiotics and topical antibiotics may have additive effects, but this is lessened if the same antibiotic is used for both topical and systemic administration. Topical antibiotics work best if they are used throughout the procedure compared to simple irrigation of the wound at the time of closure. However, the most effective time interval for application has not been established. High concentrations of the antibiotic may be assured when it is injected into the wound and retained after closure or with the use of sustained release antibiotic containing materials which, when biodegradable, may not need removal. The use of such materials in clinical studies, however, has been very restricted. Using other antimicrobials such as povidone iodine solutions to decontaminate wounds has not been effective and has been shown to inhibit wound healing and/or increase wound infection. They should not be used.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.