Surgical site infection, which includes wound infection and endometritis, is one of the most common complications of major surgery.1 Patients who develop this complication have a twofold increase in the length of hospital stay2 and cost the U.S. health care system approximately $1.8 billion per year.3 Rates of surgical site infections after cesarean delivery have been reported in the literature to range anywhere from 0.7% to 23.5%.4–6 These infections represent significant morbidity for patients undergoing cesarean delivery as well as increased costs for hospitals providing care for these patients.7
Various aspects of perioperative care have been evaluated and studied to find interventions that can decrease incidence of surgical site infection. Even with completely clean procedures, most surgical incisions contain some bacteria by the end of the procedure,8 and prevention of surgical site infection is aimed at enhancing native immune functioning and preventing the incisional bacterial load from converting into infection. Subcutaneous perfusion and adequate tissue oxygenation are two other important components of immunity to wound infection.9
In recent years, there has been much interest in the concept of preventing surgical site infection by increasing tissue oxygenation. Adequate tissue oxygenation at wound sites may have a clinically important effect on wound infection by enhancing bactericidal function of leukocytes.10 When combined with adequate subcutaneous perfusion, supplemental oxygen may decrease postoperative wound infection.9 Two large, randomized controlled studies done in Europe on patients undergoing colorectal surgery showed that increasing the concentration of supplemental oxygen from 30% to 80% was associated with a 39–50% decrease in surgical site infection.8,11
Given this background, the purpose of the current study was to evaluate whether supplemental perioperative oxygen could reduce the occurrence of surgical site infection or endometritis in women undergoing cesarean delivery.
MATERIAL AND METHODS
This was a double-blind, randomized controlled trial conducted at Santa Clara Valley Medical Center from August 2006 to August 2010. The study was approved by the institutional review board at our institution, and all participants gave written informed consent to participate in the study. Women were eligible if they were to undergo elective or emergency cesarean delivery. Exclusion criteria included fever (temperature of 38°C or higher), chorioamnionitis (temperature of 38°C or higher with fetal or maternal tachycardia), patients who were group B streptococci-positive and had been started on antibiotics, immunocompromised or HIV-positive patients, planned general anesthesia, age younger than 18 years, and incarcerated patients.
A computer-generated randomization table was used to randomly allocate eligible women to either 30% or 80% oxygen during surgery (after umbilical cord clamping) and for 1 hour postoperatively. Randomization was done after cord clamping because the effect of an increased concentration of oxygen on the fetus is controversial.12,13 The anesthesiologist was the only person aware of the concentration of oxygen given to the patient. An anesthesia form noted the study dose of oxygen used. The information of the exact dose of oxygen used, along with the patient's name, medical record number, and randomization number, was kept in a binder in a locked cabinet. The patients and physicians examining the wound postoperatively were blinded to the oxygen concentration used. It is routine at our institution for the cesarean deliveries to be performed by an attending obstetrician and second-year resident.
Despite the cesarean deliveries being performed by different health care providers, the cesarean delivery itself was standardized as much as possible, particularly with respect to factors that directly influence the rate of postoperative infections. All patients received prophylactic antibiotics, usually cefazolin, 2 g intravenously, after cord clamping. If the patient was allergic to penicillin, she received 900 mg clindamycin. If the subcutaneous tissue was more than 4 cm in thickness, it was closed with 2-0 plain catgut. The decision regarding method of skin closure was made by the operating surgeons. Care was taken to make sure the patients received adequate fluid resuscitation and that optimal room temperature was maintained in the operating room to ensure normothermia and adequate tissue perfusion. A warming blanket was used at the discretion of the anesthesiologist. An aerosol face mask was used to deliver oxygen at 30% or 80% concentration during surgery and for 1 hour after surgery. The flow rate was 10 L per minute in both groups. Preoperative and postoperative fingerstick blood glucose levels were done on the day of surgery to ensure normoglycemia during surgery.
Preoperative data collected included the patient's age, race, parity, body mass index (BMI, calculated as weight (kg)/[height (m)]2), number of previous cesarean deliveries, history of diabetes, cardiac or pulmonary disease, anemia, smoking, or chronic steroid use. The volume of intravenous fluids administered, the American Society of Anesthesiologists score, and duration of surgery were recorded in all patients. Other parameters noted were intraoperative factors such as blood loss, transfusion, type of skin incision, duration of surgery, anesthesia class, medications given during anesthesia, and postoperative hematocrit and fingerstick blood glucose.
On the first postoperative day, the patient had a complete blood count, and the incision was examined daily during her postoperative stay in the hospital. After discharge from the hospital, the patient was seen in the clinic 2 weeks postoperatively for any evidence of surgical site infection or endometritis. If she failed to keep her 2-week appointment, an attempt was made to reach her by phone or a home visit by a public health nurse. If this could not be achieved, the 6-week postpartum visit record was reviewed to look for any historical evidence of wound infection.
The primary outcome was a composite outcome of either surgical site infection or endometritis. Surgical site infection was defined per the Centers for Disease Control and Prevention guidelines.14 At least one of the following criteria was required: 1) purulent discharge from the incision site; 2) organisms isolated from an aseptically obtained culture or tissue from the superficial incision; 3) at least one of the following signs or symptoms of infection: pain or tenderness, localized swelling, redness or heat, and the superficial incision was opened by the surgeon unless the incision culture was negative; and 4) diagnosis of superficial incisional surgical site infection by the surgeon.
Postpartum endometritis was diagnosed by the clinical finding of a temperature of more than 38°C associated with uterine tenderness without any other source of fever identified. Each of the infectious outcomes was examined separately as well. Additional stratified analysis was conducted in women with diabetes mellitus.
A previous review of our own data showed our baseline surgical site infection rate was 8%. To demonstrate a 50% difference in infection rate, with a two-sided α of 0.05 and a power of 80%, we planned to recruit 1,202 patients in a one-to-one randomization. However, after we had 98 patients over a 9-month period, we estimated that the study would take 9 years to accomplish. We reexamined our primary outcome to combine two of the infectious outcomes associated with cesarean delivery, surgical site infection or endometritis. This increased the baseline rate to 12% and led to a revised sample size estimation of 778 patients. This did not change our study protocol in any way because we were already collecting data on endometritis. We used an intention-to-treat analysis. Dichotomous outcomes were compared using the χ2 or Fisher’s exact test and continuous outcomes were compared with the Student’s t test. All analyses were conducted with Stata 11 software. The study is registered with Clinicaltrials.gov (NCT00876005).
A total of 831 participants were recruited for the study (Fig. 1). Most of the recruitment (80%) for the study took place in the preoperative clinic for patients undergoing scheduled cesarean deliveries. We found it difficult to adequately counsel patients on the labor and delivery unit. A total of 415 women received 30% oxygen and 416 women received 80% oxygen intraoperatively and for 1 hour postoperatively. The two groups were well matched for age, race, parity, BMI, number of previous cesarean deliveries, elective cesarean deliveries, history of diabetes, cardiac or pulmonary disease, anemia, smoking, or chronic steroid use (Table 1).
All women were followed for evidence of surgical site infection or endometritis during their postoperative hospital stay by the surgical team caring for the patient. Most patients were discharged home on postoperative day 3. There was no significant difference in the rate of surgical site infection or endometritis in the two groups at the time of discharge, 2.4% compared with 2.9% (P=.70) (Table 2).
All women were given an appointment for a postoperative visit for wound check at 2 weeks postpartum. There were 325 patients (39.1%) who did not show for their 2-week postoperative evaluation. A vigorous attempt was made to reach the women who did not follow up. Thirty women were seen at home by a public health nurse, and 10 women were reached by phone and a detailed history of their wound status was taken. Of the 285 participants who could not be contacted by any means at 2 weeks, 150 were seen at their 6-week postpartum visit and were questioned about any wound infection. There were 135 participants (16.2%), 65 in the 30% group and 70 in the 80% group, who were not seen or contacted after their discharge from the hospital. The number of participants who could not be contacted postoperatively did not differ significantly between the two groups.
The primary composite outcome of surgical site infection or endometritis was not different between the two groups, 8.2% in women who received 30% oxygen compared with 8.2% in women who received 80% oxygen (P=.89) (Table 2). In all the participants who were seen postpartum, the rate of surgical site infection was 5.5% in the patients who received 30% oxygen and 5.8% in the patients who received 80% oxygen (P=.98). The rate of endometritis was 2.7% in the women who received 30% oxygen and 2.4% in those who received 80% oxygen (P=.66).
The diabetic patients were analyzed separately. There were 87 diabetic patients who received 30% oxygen and 90 who received 80% oxygen (Table 3). There was no difference in the class of diabetes, BMI, or number of elective cesarean deliveries in the two groups. Of these, 28 patients (15.8%) could not be contacted for an assessment by 6 weeks postpartum. The rate of surgical site infection was 6.9% in those who received 30% oxygen and 14.4% in those who received 80% oxygen (P=.11) (Table 4).
Our study showed that there was no difference in surgical site infection or endometritis in patients who received either 30% or 80% oxygen during cesarean delivery and for 1 hour afterward. There has been much interest in the use of supplemental perioperative oxygen to reduce the incidence of surgical site infections after Greif et al11 showed the wound infection rate was reduced from 11.2% to 5.2% in patients who received 80% oxygen compared with 30% oxygen during colorectal surgery and for 2 hours postoperatively. A second multicenter trial published in 2005 also showed a significant reduction in infection rates in patients undergoing colorectal surgery who received 80% oxygen.8 However, another study in a similar general surgical population showed that a high perioperative fraction of inspired oxygen did not reduce the overall significance of surgical site infection and may even have deleterious effects.15 We did not find a clinically or statistically significant difference in the rate of postoperative infections by providing 80% oxygen in women undergoing cesarean delivery. Some of the major differences between our study and the ones that showed a positive result were that the patients undergoing colorectal surgery were sicker, had greater operative times, and more likely to have received general anesthesia.
To our knowledge, there have not been many studies looking at supplemental perioperative oxygen in patients undergoing cesarean delivery. In 2008, Gardella et al16 published a randomized trial on 143 women undergoing cesarean deliveries suggesting that high-concentration perioperative oxygen did not decrease the risk of postcesarean surgical site infection. Similarly, in 2011, Scifres et al17 published a nonblinded, randomized trial on 585 women undergoing cesarean deliveries suggesting there is no benefit from high-concentration perioperative oxygen with regard to postoperative infectious morbidity. Our study is consistent with these two prior studies.
The concept of using oxygen as an antimicrobial agent was first considered in 1980.1 One of the mechanisms of early bacteria eradication involves oxidative processes in white cells that depend on oxygen tension at the wound site.1 Studies by Hopf et al9 have shown that increasing inspired oxygen leads to increased oxygen tension at the wound site. However, reactive oxygen species produced by a high oxygen partial pressure can produce tissue injury and inhibit antibacterial mechanisms.15 Increased partial oxygen pressure can produce opposing physiologic consequences and the results of previous studies have been conflicting.17 Thus, it might be that a potential detrimental effect counterbalanced any benefit from the increased oxygen yielding a null finding. Perhaps a lower oxygen concentration for a longer period of time would be beneficial.
A major strength of this current study was that it was randomized and both the surgeon and the patient were blinded to the concentration of oxygen used. However, our study has several limitations. An attempt was made to standardize the surgical procedure, although different surgeons performed the procedures in both groups and there may have been minor variations in surgical technique. Our primary outcome infection rate was lower than what we had assumed, thus lowering the statistical power of the study. For example, we had only 34% power to detect a one-third lowering in the surgical site infection rate. Indeed, as detailed in the “Material and Methods” section, our lower than anticipated surgical site infection rate led us to broaden our primary outcome to include endometritis. A further limitation of the study was that we had a relatively high lost-to-follow-up rate of 16.2%. The study was conducted at a single site, which had a population that was generally Hispanic, obese, and had elective cesarean deliveries, so the results may not be broadly applicable to other populations.
In conclusion, the findings of our study suggest that increasing the concentration of oxygen in the inspired air in women undergoing cesarean deliveries under regional anesthesia does not decrease the rate of surgical site infection or endometritis.
1. Dellinger EP. Increasing inspired oxygen to decrease surgical site infection. JAMA 2005;294:2091–2.
2. Kirkland KB, Briggs JP, Trivette SL, Wilkinson WE, Sexton DJ. The impact of surgical-site infections in the 1990s: attributable mortality, excess length of hospitalization, and extra costs. Infect Control Hosp Epidemiol 1999;20:725–30.
3. Bratzler DW, Houck PM, Richards C, Steele L, Dellinger EP, Fry DE, et al.. Use of antimicrobial prophylaxis for major surgery: baseline results from the National Surgical Infection Prevention Project. Arch Surg 2005;140:174–82.
4. Cardoso Del Monte MC, Pinto Neto AM. Postdischarge surveillance following cesarean section: the incidence of surgical site infection and associated factors. Am J Infect Control 2010;38:467–72.
5. Ghuman M, Rohlandt D, Joshy G, Lawrenson R. Post-caesarean section surgical site infection: rate and risk factors. N Z Med J 2011;124:32–6.
6. Gong SP, Guo HX, Zhou HZ, Chen L, Yu YH. Morbidity and risk factors for surgical site infection following cesarean section in Guangdong Province, China. J Obstet Gynaecol Res 2012;38:509–15.
7. Olsen MA, Butler AM, Willers DM, Gross GA, Hamilton BH, Fraser VJ. Attributable costs of surgical site infection and endometritis after low transverse cesarean delivery. Infect Control Hosp Epidemiol 2010;31:276–82.
8. Belda FJ, Aguilera L, Garcia de la Asuncion J, Alberti J, Vicente R, Ferrandiz L, et al.. Supplemental perioperative oxygen and the risk of surgical wound infection. JAMA 2005;294:2035–42.
9. Hopf HW, Hunt TK, West JM, Blomquist P, Goodson WH, Jensen JA, et al.. Wound tissue oxygen tension predicts the risk of wound infection in surgical patients. Arch Surg 1997;132:997–1004; discussion 1005.
10. Allen DB, Maguire JJ, Mahdavian M, Wicke C, Marcocci L, Scheuenstuhl H, et al.. Wound hypoxia and acidosis limit neutrophil bacterial killing mechanisms. Arch Surg 1997;132:991–6.
11. Greif R, Akca O, Horn E-P, Kurz A, Sessler DI. Supplemental perioperative oxygen to reduce the incidence of surgical-wound infection. N Engl J Med 2000;342:161–7.
12. Khaw KS, Wang CC, Ngan Kee WD, Pang CP, Rogers MS. Effects of high inspired oxygen fraction during elective caesarean section under spinal anaesthesia on maternal and fetal oxygenation and lipid peroxidation. Br J Anaesth 2002;88:18–23.
13. Backe SK, Lyons G. Oxygen and elective caesarean section. Br J Anaesth 2002;88:4–5.
14. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections, 1988. Am J Infect Control 1988;16:128–40.
15. Pryor KO, Fahey TJ III, Lien CA, Goldstein PA. Surgical site infection and the routine use of perioperative hyperoxia in a general surgical population. JAMA 2004;291:79–87.
16. Gardella C, Goltra LB, Laschansky E, Drolette L, Magaret A, Chadwick HS, et al.. High-concentration supplemental perioperative oxygen to reduce the incidence of postcesarean surgical site infection. Obstet Gynecol 2008;112:545–52.
17. Scifres CM, Leighton BL, Fogertey PJ, Macones GA, Stamilio DM. Supplemental oxygen for the prevention of postcesarean infectious morbidity: a randomized controlled trial. Am J Obstet Gynecol 2011;205:267.e1–9.