CORR Insights®: Does an Antimicrobial Incision Drape Prevent Intraoperative Contamination? A Randomized Controlled Trial of 1187 Patients

Where Are We Now?

Periprosthetic joint infection (PJI) is a devastating complication that poses substantial risks to our patients. Patients who undergo revision TKA for PJI are at a considerably greater risk of perioperative complications, readmission, and mortality compared to patients who undergo revision for aseptic causes [9]. Unfortunately, infection is now the most common reason for early revision TKA and the second most common reason for revision TKA overall [21]. The costs associated with PJI continue to increase and are projected to exceed USD 1.6 billion by 2020 [16].

One intraoperative intervention for primary prevention that has been widely adopted is the routine use of iodine-impregnated adhesive surgical drapes. Many surgeons use these drapes to create a physical barrier between sterile and nonsterile regions and to prevent migration of drapes during the procedure. In their randomized trial, Hesselvig et al. [13] found that iodine-impregnated adhesive drapes reduced the rates of bacterial contamination of surgical wounds during TKA. However, their study did not investigate whether this difference in the contamination rate was correlated with differences in the rate of subsequent PJI.

Guidelines for the prevention, diagnosis, and treatment of PJI are continuously challenged and refined by emerging evidence [20]. A consensus meeting that discussed skin preparation and surgical draping found no evidence that these iodine-impregnated drapes reduce the rate of subsequent PJI [4]. A comparative study [18] and systematic review [22] also could not find an effect of these drapes on the rate of surgical-site infection, and some authors have suggested that their use for prophylaxis against surgical-site infection is not necessary in modern total joint arthroplasty [17].

Although an effect on PJI has not been proven, some reports have found that iodine-impregnated drapes do reduce the rates of bacterial wound contamination [11, 19]. A recent randomized, single-surgeon clinical trial of 101 nonimplant hip preservation surgeries found a lower rate of bacterial contamination (12%) in a group with iodine-impregnated adhesive drapes compared to a control group that did not use adhesive drapes (27%) [19]. This study by Hesselvig et al. [13] included five centers, 24 surgeons, 1187 patients, and all common knee arthroplasty procedures (including TKA, unicompartmental arthroplasty, and patellofemoral arthroplasty). Despite regional differences in skin preparation, perioperative antibiotic administration, and inherent variations between surgical centers and surgeons, Hesselvig et al. [13] found that the bacterial contamination rate was lower (10%) in procedures that used these drapes compared to those that did not (15%).

Where Do We Need To Go?

There are few randomized controlled trials on this subject in orthopaedic surgery. Hesselvig et al. [13] have completed one, and they should be applauded for this, but their study has several limitations. A future multicenter study could use a centralized clinical laboratory instead of relying on multiple site-specific laboratories with heterogenous standards and protocols. Additionally, future studies could standardize agents for perioperative antibiotic prophylaxis, and rigorous implementation and data collection could reduce the exclusion rate of future studies.

Although Hesselvig et al. [13] found a correlation between iodine-impregnated drapes and reduced contamination rates, this study does not address the clinically important question of whether these drapes actually prevent infections. Many PJI isolates are skin flora, and for this reason, it is likely that a decreased rate of procedural wound contamination is associated with a decreased rate of subsequent PJI. However, most PJIs are a multifactorial consequence [1] of host factors [8, 14], surgical factors [2, 5, 12], and environmental exposure [3, 7]. Wound contamination is an imperfect surrogate for PJI, and more evidence is needed to prove the efficacy and cost-effectiveness of iodine-impregnated adhesive drapes in orthopaedic surgery.

How Do We Get There?

Hesselvig et al. [13] stated that they will follow their cohort to determine whether observed contamination is correlated with later infection. I hope that Hesselvig et al. will follow their entire cohort to determine whether the use of iodine-impregnated drapes is correlated with later PJI, regardless of whether culture swabs confirmed contamination at the index procedure. Institutional data regarding the real cost of iodine-impregnated drapes and the costs of treating patients with PJI could then be used to determine whether this intervention was cost-effective.

Other surgical subspecialties are leading the way in this regard. In cardiac surgery, a propensity-matched study using longitudinally maintained data evaluated iodine-impregnated drapes versus non-iodine-impregnated drapes and found a lower rate of surgical-site infection with iodine-impregnated drapes at an average cost savings of nearly 800 Euros per patient [6]. Perhaps Hesselvig et al. could replicate that analysis in a follow-up study of their cohort [13]. Alternatively, an analysis of institutional costs could be used to determine the absolute risk reduction necessary for routine use of iodine-impregnated drapes to be cost-effective [15], and this information could be useful for a power analysis in a de novo trial. At most centers, the cost of iodine-impregnated adhesive drapes is relatively low and the absolute risk reduction necessary to find their use cost-effective is similarly low. Given this theoretically small target, the number of patients needed to appropriately power a definitive trial might not be feasible [10]. If the effect is small, Hesselvig et al.’s follow-up study ultimately may not find any statistical difference, and recent proof that the routine use of these drapes reduces the rates of wound contamination may be the best evidence that surgeons will get.