Prophylactic gentamicin-collagen sponges do not appear to significantly decrease infection rate compared with no intervention in patients undergoing colorectal surgery, according to the findings of a large randomized phase 3 trial.
Elliott Bennett-Guerrero, MD, from the Duke Clinical Research Institute, in Durham, North Carolina, and colleagues reported their findings online August 4 in The New England Journal of Medicine.
According to the article, the gentamicin-collagen sponge is approved for surgical implantation in 54 countries, and more than 1 million patients have been treated with the sponges since its approval in Germany in 1985.
The sponges, when inserted, provide high localized concentrations of antibiotics, avoiding systemic concentrations that can damage the kidney, and have demonstrated encouraging data in previous studies.
Dr. Bennett-Guerrero and colleagues conducted the current phase 3 study to confirm the findings from those earlier trials and to support regulatory approval in the United States.
The trial included 602 patients undergoing open surgery or laparoscopy for colorectal surgery at 39 centers across the United States. Patients were randomly assigned to receive either no intervention or the insertion of 2 sponges above the fascia on wound closure. All patients also received standard prophylactic systemic antibiotics.
More surgical site infections were reported in patients assigned to receive sponges compared with those undergoing no intervention (30% vs 20.9%; P = .01). Superficial surgical site infections occurred in 20.3% of patients in the sponge group vs 13.6% of patients in the control group (P = .03). In addition, compared with control patients, more patients receiving the sponges were likely to seek medical care because of wound-related symptoms (19.7% vs 11.0%; P = .004).
The incidence of deep surgical site infections was not significantly different between groups (8.3% vs 6.0% for sponge vs control; P = .26), nor was the number of rehospitalizations resulting from surgical site infections (7.0% vs 4.3% for sponge vs control; P = .15).The frequency of adverse events was comparable between the 2 groups.
"Our data do not allow us to identify with certainty the cause of the lack of efficacy we observed," the authors note. "However, we can speculate that several factors may have been operational."
According the researchers, factors that contributed to these findings may have included the presence of more-resistant bacteria in the sponge group. In addition, the sponge may fail to provide adequate local concentrations of gentamicin, and a sponge depleted of antibiotic may even harbor bacteria and increase the risk for infection. Another reason, the authors note, is that collagen used to construct the sponges could have stimulated a deleterious local effect or have provided a barrier to rapid and effective wound closure.
Previous studies showing a benefit for the sponges may have potentially lacked adequate follow-up and/or quality control measures, and may also have had inherent differences in patient populations.
The study was supported by Innocoll Technologies. Several authors have disclosed funding from Innocoll Technologies in the form of grant support, travel accommodations, and other payment and expenses. In addition, some authors have disclosed funding support in the form of personal or institutional grants, consultancy fees, stocks, travel support, and other payments from one or more of the following companies: Alexicon, Applied Medical, Eli-Lilly, Ethicon, Helsinn, Genzyme, Life Cell, Medicure, Medtronic, Novadeq, NTII, Proctor and Gamble, Sanofi-Aventis, Springer, Transentric, and others.
N Engl J Med. Published online August 4, 2010.
