Dear Editor,

The potential clinical benefits of the perioperative use of high inspired oxygen fraction (FIO2) for preventing surgical site infections (SSIs) have attracted great interest in recent years. Trials by Greif et al. [1] and Belda et al. [2] demonstrated that SSIs decreased significantly following colon surgery in patients who received 80% oxygen intraoperatively and for the first hours following surgery.

In the sphere of cardiac surgery, SSIs are serious complications associated with extended hospital stay, increased hospital costs, and higher mortality and morbidity rates [3]. Thus, in 2005 our Department of Anesthesiology and Reanimation adopted a clinical strategy of administering 50% oxygen without nitrous oxide during anesthesia and for the first 6 postoperative hours in an effort to decrease SSIs.

In contrast to the findings of Belda et al. [2], clinical trials by Pryor et al. [4] and, more recently, by Meyhoff et al. [5], found no difference in SSI risk when 80% oxygen rather than 30% oxygen was administered during abdominal surgery and for 2 hours postoperatively. Their findings suggested that perioperative hyperoxia was not effective in reducing SSIs. These reports add to the evidence base surrounding the potential role of high FIO2 in SSI prevention.

The rationale for administering high FIO2 to prevent SSIs is to produce a high PaO2 and thereby increase the PsqO2 (tissue oxygen partial pressure), since oxidative killing by neutrophils is the primary defense against surgical pathogens. The risk of infection is thus inversely related to PsqO2[3]. Our aim in this study was to analyze the relationship between PaO2 values and SSIs.

We designed a prospective study that analyzed the data from 1,024 consecutive patients who underwent cardiac surgery with extracorporeal circulation at our institution from January 30, 2007 to June 30, 2009. Transplant patients were excluded. The patients were categorized according to the presence or absence of SSIs. The study was approved by the hospital's Research Commission, and all participants provided informed written consent. The Center for Disease Control and Prevention (CDC) criteria [6] were used to define SSIs. The SPSS software package (version 15) was used for statistical analysis. A p ≤ 0.05 was considered significant.

To assess risk factors for SSI, we used one-way analysis of variance for univariate continuous variables and the chi-square test for categorical variables. In addition, we conducted Fisher's exact test whenever the chi-square expected value of at least one cell was less than 5.

We avoided multicollinearity among the explanatory variables by performing collinearity diagnostic analyses. We performed the stepwise selection of variables from the models with the following criteria: Tolerance greater than 0.4 or variance inflation less than 2.5, condition number less than 10, and a variance of two or more variables no greater than 0.5.

SSIs developed after cardiac surgery in 54 (5.3%) patients, 28 (2.8%) superficial or deep incision SSIs and 26 (2.5%) organ/space SSIs. The intraoperative and postoperative PaO2 values were not associated with an increased risk of SSI either by univariate or multivariate analysis (Table 1). The 30-day mortality rate was similar in both groups: patients without SSIs, n = 72 (7.4%) vs. patients with SSIs, n = 4 (7.4%); (P = .11).

Table 1 Characteristics and preoperative, intraoperative, and postoperative data for patients with and without surgical site infections (SSIs).
Full size table

Our results agree with the results of the trials conducted by Pryor et al. [4] and Meyhoff et al. [5] in that perioperative hyperoxia was not effective in reducing SSIs. PsqO2 is typically lower than the PaO2 level by a factor of two to four. As might be expected, tissue oxygenation improves much less than arterial oxygen in response to supplemental oxygen administration. Sternal wound oxygenation increased by an average of 4 mm Hg (from 23 to 27 mm Hg) with supplemental oxygen at 50% [3].

The data from prior studies [4, 5], as well as the present results, leads us to question our policy to routinely administer a high inspired oxygen fraction to cardiac surgery patients in order to prevent SSIs. In summary, the PaO2 in adult cardiac surgery patients is not related to SSI rate. The strategy of administering supplemental inspired oxygen to reduce the incidence of SSIs does not appear to be clinically useful.