The prevalence of obesity (defined as body mass index [BMI, calculated as weight (kg)/[height (m)]2] 30 or higher) continues to increase, affecting approximately 40% of adult females in the United States according to Centers for Disease Control and Prevention 2013–2014 statistics.1 Obesity has a significant effect on many aspects of pregnancy, including increased need for cesarean delivery. In addition, obesity is a well-recognized risk factor for the development of postoperative surgical site infections and other wound complication.2,3 In the late 1990s, Tran et al4 found a 9.8% surgical site infection rate among obese women after cesarean delivery with risk of infection doubling for every five-unit increment of BMI.
Prophylactic negative pressure wound therapy as a means of reducing wound infection has been explored as primary dressings in the orthopedic and cardiothoracic fields, but limited data have been published in the peer-reviewed literature report on use with abdominal incisions. Two meta-analyses recently combined published peer-reviewed manuscripts, meeting abstracts, and data published on intranet websites to determine the effectiveness of prophylactic negative pressure wound therapy at the time of cesarean delivery.5,6 Unfortunately, these two meta-analyses reported conflicting data on the effectiveness of prophylactic negative pressure.
Therefore, the objective of this randomized controlled trial was to compare the occurrence of superficial surgical site infections in women with class II or III obesity as defined by the Centers for Disease Control and Prevention using prophylactic negative pressure wound therapy compared with standard dressings after cesarean delivery.
MATERIALS AND METHODS
We conducted a randomized controlled, nonblinded, multicenter study of prophylactic negative pressure wound therapy compared with standard surgical dressings placed at the time of primary closure in a cesarean delivery. Institutional review board approval was obtained from the Dartmouth Committee for the Protection of Human Subjects on April 21, 2015 (#00005211) and from the Southern New Hampshire Medical Center Clinical Trials Office (#2015-01). This trial was registered with clinicaltrials.gov (Clinical Trial Registration: NCT02390401). Women were recruited and consented to participate in this study before the onset of active labor during any routine prenatal visit or inpatient admission.
Women were recruited from two hospitals, Dartmouth-Hitchcock Medical Center (a rural tertiary care center with a level III neonatal intensive care unit) and Southern New Hampshire Medical Center (a community-based hospital with a level II neonatal intensive care unit).
Women were eligible for inclusion if they underwent cesarean delivery for a viable neonate and their BMI on admission to the labor and delivery floor was 35 or higher. All women undergoing cesarean delivery were eligible for inclusion. Women were approached only for participation and consent in this study before active labor Additionally, women who were younger than 18 years old, did not speak English, had an allergy to silver or adhesives products, or who had a skin incision that would not fit the device or standard dressing (eg, “T” skin incision) were excluded.
Abdominal preparation for cesarean delivery was performed with 2% chlorhexidine alcohol solution. All women received intravenous antibiotics before skin incision consistent with American College of Obstetricians and Gynecologists’ guidelines7 with the majority of women receiving cefazolin or clindamycin with gentamicin if allergic. Surgical techniques were left to the discretion of the surgeon. Women were discharged postoperatively by the primary team, usually on the third or fourth postoperative day, irrespective of study assignment. All women were scheduled to return for a wound assessment on postoperative day 5–7.
If randomized to prophylactic negative pressure wound therapy, the device was applied at the time of primary skin closure at cesarean delivery. It was placed over the closed surgical incision under sterile conditions and was removed between postoperative day 5 and 7 at the time of incision check.
The standard dressing after cesarean delivery was also applied using a sterile technique. If subcuticular closure was used, sterile slim adhesive strips (also known as Steri-Strips) were applied. For both subcuticular and staple closure, the dressing consisted of a sterile nonadherent wound dressing (also known as Telfa), a sterile gauze, and a waterproof transparent adhesive dressing (also known as Tegaderm). The standard dressing was removed on postoperative day 2.
A randomization program (www.randomization.com, Alberta, Canada) was used to generate sealed opaque envelopes with study assignment. Women were randomized at the conclusion of the cesarean delivery, during skin closure, when the envelopes were opened by a circulating operating room nurse. Two randomization strata were created using permuted blocks with varying block sizes for women with BMIs from 35 to less than 40 and women with BMIs of 40 or higher and for each site to ensure equal distribution of study allocation across these two separate BMI categories and sites.
Our primary outcome was occurrence of superficial surgical site infection, defined according to Centers for Disease Control and Prevention criteria8 as infection involving only the skin or subcutaneous tissue occurring within 30 days of surgery with at least one of the following: 1) purulent drainage from the wound or 2) organism identified by culture or wound deliberately opened by the surgeon. Patients also needed to have at least one of the following symptoms: pain, swelling, or erythema. Our secondary outcome was a composite wound complication, including superficial, deep, or organ-space surgical site infection; wound dehiscence; seroma; or hematoma that occurred within 30 days of surgery. Other secondary outcomes included 30-day readmission, 30-day reoperation, and need for antibiotic treatment for any reason (eg, mastitis, urinary tract infection, pyelonephritis). We also examined length of stay after cesarean delivery and rates of breastfeeding at discharge this seems to have been erroneously duplicated. We had already stated this earlier in the text. Additional outcomes were determined by chart review.
Standard statistical methods such as frequency distributions and measures of central tendency were used to evaluate the demographic and clinical characteristics of the participants. Pearson χ2, Fisher exact test, and Student t test for parametric distributed variables and Kruskal-Wallis for nonparametric distributions were used as appropriate. We first completed an observed outcome analysis based on observed outcomes for the women who completed their 1-week postoperative incision check. We then calculated an unadjusted relative risk and 95% CI for the primary outcome. We then completed a second outcome analysis for all women who were randomized and received their intended treatment despite lack of follow-up at the 5- to 7- day incision check by imputing missing data. We assumed that all women who were lost to follow-up had developed superficial surgical site infections. For the secondary outcome of composite wound complication, we assumed that complications that would not require readmission happened (eg, seroma, hematoma, or wound dehiscence) and we assumed that complications that would generally require inpatient readmission did not happen (eg, deep surgical site infections or organ-space surgical site infections). Finally, we conducted two subgroup analyses of women with BMIs of 40 to 50 and BMIs of 50 or higher to see whether these groups had differences in superficial surgical site infections not observed in the larger cohort.
At the time of study initiation (April 2015), based on the current literature, we assumed an occurrence of superficial surgical site infection of 20% of obese women undergoing cesarean delivery.4,9,10 We determined that 400 women (200 prophylactic negative pressure wound therapy, 200 standard dressing) would need to be recruited to have an 80% power to detect a 50% decrease in superficial surgical site infection (assuming P<.05). We anticipated that enrollment across the two sites would take 24 months.
At the completion of the anticipated 2 years of recruitment, 166 women out of a planned 400 (42%) had been enrolled in the study. This was likely the result of the limitations of women's ability to return for device removal and incision evaluation 5–7 days after delivery (many had been transported to the tertiary recruiting site and lived a long distance from this site). As a result of this low enrollment, the study's data safety monitoring board recommended an unplanned interim analysis with futility calculation to determine whether the study should continue.
We anticipated that to reach our original recruitment goals, we would need to continue our study for 34 additional months. If we assumed that observed outcomes in the remaining 244 women would develop the primary outcome, superficial surgical site infection, as predicted in the original sample size calculation (20% superficial surgical site infection in women with standard dressing and 10% in women with prophylactic negative pressure wound therapy), our study's power to detect a difference between the two interventions would be 29% at completion of enrollment of our 400 participants. If we assumed that the primary outcome in the remaining 244 women would be the same as prevalence observed in the first 166 women who had already completed the study, our study's power to detect a difference between the two interventions would be less than 0.1%. Therefore, the determination of the data safety monitoring board was that continued enrollment to original recruitment targets was futile and the recommendation was made to halt further enrollment and disseminate our findings.
Women were recruited between January 5, 2015, and January 7, 2017. A total of 166 women were randomized, 163 received their allocated treatment, and 161 (80 prophylactic negative pressure wound therapy and 81 standard dressing) returned for their postoperative wound assessment approximately 1 week after cesarean delivery (Fig. 1). The two randomization groups were similar in regard to maternal characteristics and intrapartum and intraoperative events (Tables 1 and 2).
Twelve percent of women (20/161) completing the 1-week incision check experienced superficial surgical site infection. There were no statistically significant differences in the occurrence of observed superficial surgical site infections between women using prophylactic negative pressure wound therapy (12/80 [15%]) compared with women who received the standard dressing (8/81 [10%], P=.35; Table 3; relative risk 1.52, 95% CI 0.66–3.52). Assuming that all women who were lost to follow-up developed superficial surgical site infection, there still would not have been a statistically significant difference in the occurrence of superficial surgical site infection between groups (12/80 [15%] vs 13/86 [15%], P=.98).
There were no significant differences in the occurrence of composite wound complications between women using prophylactic negative pressure wound therapy (25/80 [31%]) compared with women who received the standard dressing (24/81 [30%], P=.87). In a subgroup analysis of the 71 women with BMIs of 40 to 50, there were no statistically significant differences observed in superficial surgical site infections in the women using the prophylactic negative pressure wound therapy (7/31 [23%]) compared with women receiving standard dressings (7/40 [18%], P=.77). In a subgroup analysis of the 34 women with BMIs higher than 50, there were no differences in the occurrence of observed superficial surgical site infections in women using the prophylactic negative pressure wound therapy (21% [4/19]) compared with women receiving standard dressings (20% [3/15], P>.99).
Risk factors for superficial surgical site infection are explored in Table 4. In addition to maternal BMI category, number of prior cesarean deliveries and preoperative anemia (serum hematocrit less than 36%) were both associated with increased postcesarean superficial surgical site infection (Table 4).
There were no statistically significant differences observed for length of stay after delivery and breastfeeding at the time of hospital discharge between women with prophylactic negative pressure wound therapy and standard dressings. Fifteen percent of women (12/80) with the prophylactic negative pressure wound therapy device had it removed early: five women for local skin irritation, six women because the pump stopped working, and one additional woman requested early removal.
In this randomized controlled trial, we did not observe a decrease in superficial surgical site infection after cesarean delivery in obese women with the use of prophylactic negative pressure wound therapy. Even in a subgroup analysis of morbidly obese (BMI of 40 to 50) and the most obese women (BMI higher than 50), who are considered to be at significantly higher risk of surgical site infection and wound complication based on published literature (Tran et al), we did not find a difference in the occurrence of superficial surgical site infection with the use of prophylactic negative pressure wound therapy; however, the results of these subgroup analyses are limited by small numbers.
Our study was limited by slow enrollment and therefore failure to reach our planned sample size. The two limiting factors in our enrollment were the need to consent women before active labor to ensure women could give informed consent free from pain and the need for patients to return for an incision check 5–7 days after their delivery. As a tertiary center for a large rural geographic catchment area, many women delivering at our facility could not feasibly return for an incision check after discharge.
In conclusion, we did not observe a statistically significant difference in postcesarean superficial surgical site infection or composite wound complications between women who used the prophylactic negative pressure wound therapy and women who received a standard dressing. There is currently a large multicentered National Institutes of Health-funded trial (ClinicalTrials.gov Identifier: NCT03009110) to examine the use of prophylactic negative pressure wound therapy for the reduction of wound complications in obese women after cesarean delivery with a goal of 2,850 women. Until the results of this or other trials are published, we recommend against use of prophylactic negative pressure wound therapy after cesarean delivery.
1. Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, Ogden CL. Trends in obesity among adults in the United States, 2005 to 2014. JAMA 2016;315:2284–91.
2. El-Chaar D, Finkelstein SA, Tu X, Fell DB, Gaudet L, Sylvain J, et al. The impact of increasing obesity class on obstetrical outcomes. J Obstet Gynaecol Can 2013;35:224–33.
3. Myles TD, Gooch J, Santolaya J. Obesity as an independent risk factor for infectious morbidity in patients who undergo cesarean delivery. Obstet Gynecol 2002;100:959–64.
4. Tran TS, Jamulitrat S, Chongsuvivatwong V, Geater A. Risk factors for postcesarean surgical site infection. Obstet Gynecol 2000;95:367–71.
5. Smid MC, Dotters-Katz SK, Grace M, Wright ST, Villers MS, Hardy-Fairbanks A, et al. Prophylactic negative pressure wound therapy for obese women after cesarean delivery: a systematic review and meta-analysis. Obstet Gynecol 2017;130:969–78.
6. Yu L, Kronen RJ, Simon LE, Stoll CRT, Colditz GA, Tuuli MG. Prophylactic negative-pressure wound therapy after cesarean is associated with reduced risk of surgical site infection: a systematic review and meta-analysis. Am J Obstet Gynecol 2018;218:200–10.e1.
7. Use of prophylactic antibiotics in labor and delivery. Practice Bulletin No. 120. American College of Obstetricians and Gynecologists. Obstet Gynecol 2011;117:1472–83.
9. Leth RA, Uldbjerg N, Nørgaard M, Møller JK, Thomsen RW. Obesity, diabetes, and the risk of infections diagnosed in hospital and post-discharge infections after cesarean section: a prospective cohort study. Acta Obstet Gynecol Scand 2011;90:501–9.
10. Alanis MC, Villers MS, Law TL, Steadman EM, Robinson CJ. Complications of cesarean delivery in the massively obese parturient. Am J Obstet Gynecol 2010;203:271.e1–7.