Skin Preparation for Prevention of Surgical Site Infection After Cesarean Delivery: A Randomized Controlled Trial

    loading  Checking for direct PDF access through Ovid


I read with interest the article by Ngai et al1 describing a randomized controlled trial (RCT) of preoperative skin antisepsis with chlorhexidine with alcohol, povidone–iodine with alcohol, and both sequentially in cesarean deliveries. The trial was reasonably large and showed no significant differences in surgical site infections. The authors made a pertinent statement, which is that, before their study, chlorhexidine-alcohol and povidone-iodine-alcohol had not been compared in a meaningful way in an RCT. I wholeheartedly agree.
However, the article does not contain any information concerning preparations (brands) of antiseptics and their concentrations (eg, % weight/weight, % volume/volume) of active ingredients. This is important, because it is crucial to know whether all active ingredients were within known microbicidal concentration ranges. Some preparations indeed contain ingredients far below active concentrations, such as 23% isopropanol in one study.2 In such a case, and with unknown concentrations, a participation in the microbicidal process simply cannot be assumed. In a previous systematic review,3 we had to exclude RCTs for that reason. One would even want to know the performance of the antiseptic comparators in standardized microbicidal tests, which are a common requirement for product registration.
The question assumes new relevance because of the recent World Health Organization guidelines for the prevention of surgical site infections.4 Therein, chlorhexidine-alcohol is recommended with a Strength of Recommendation of “Strong” and a Quality of Evidence of “Low to Moderate.” The basis for this recommendation is a meta-analysis of six RCTs of chlorhexidine-alcohol compared with povidone-iodine-alcohol. However, two of six included trials contained unknown (and unretrievable) ingredient concentrations, one contained 23% isopropanol, and two were small trials with only a single surgical site infection. Another trial published in 20165 was ad hoc included, but the Ngai trial,1 published in 2015, was not. An updated meta-analysis from which the trials with unknown or inadequate ingredients were excluded, assuming a hypothetical scenario that the Ngai trial contained sufficient ingredients, eliminated the statistical significance (data not shown). Thus, I conclude that the World Health Organization recommendation for chlorhexidine-alcohol is premature and the question of antiseptic preparations in the Ngai trial becomes highly relevant for the worldwide medical community.

Related Topics

    loading  Loading Related Articles