Materials and Methods
Women who had cesarean deliveries at the Medical University of South Carolina between June 1996 and August 1997 were eligible. The study was approved by the Institutional Review Board. Written informed consent was obtained from all participants. All operations were done by obstetric residents with the assistance of one of the coauthors. Pfannenstiel incisions were done on all patients except for those with previous vertical incisions, in whom repeat vertical incisions were used. Hair on the abdominal skin was shaved immediately before preparation of the surgical site. The skin was prepared with either povidone-iodine or 4% chlorhexidine gluconate solution, depending on iodine sensitivity. The same scalpel was used for incising the skin and subcutaneous tissue. The maximum thickness of the subcutaneous tissue was measured in centimeters before we entered the peritoneal cavity. Antibiotic prophylaxis of 1 g of intravenous cefazolin was given to each woman when the umbilical cord was clamped. The placenta was removed with gentle traction and exterior massage of the uterus, and removed manually only if it was retained for more than 5 minutes. Whenever possible, the uterus was exteriorized for hysterotomy repair. Wounds were irrigated thoroughly with sterile saline before closure. Sutures and drains were not used in the subcutaneous tissue. Pfannenstiel skin incisions were closed with subcuticular 3-0 vicryl, whereas vertical skin incisions were closed with staples.
Wound infection was diagnosed when a wound drained purulent material or serosanguinous fluid, associated with induration, warmth, and tenderness. Suspected wound infections were opened for confirmation. Wounds were examined twice daily during hospitalization for evidence of infection. After discharge, women were instructed about the signs and symptoms of wound infection, given written instruction sheets with a diary for symptom documentation, and instructed to contact one of the coauthors immediately if any listed symptoms appeared. Women who contacted the investigators about symptoms were examined by one of the coauthors within 12 hours. All participants were examined 2 and 6 weeks after surgery.
We also collected demographic data and intrapartum data, including presence or duration of ruptured membranes, number of vaginal examinations, presence of chorioamnionitis, reason for cesarean, and skin incision types. Chorioamnionitis was defined as three of four of the following signs without other causative factors: maternal fever (at least 38C), uterine tenderness, maternal tachycardia (more than 120 beats per minute), and fetal tachycardia (more than 160 beats per minute).
To detect a 15% difference in the frequency of wound infection between women with less than 3 cm of subcutaneous tissue and those with at least 3 cm, with α = .05 and β = .1, we determined that 120 subjects would be needed in our sample. Discrete data were tested for significance using χ2 or Fisher exact tests. Continuous data were tested for significance with a two-tailed t test with P < .05 considered significant. To determine the potential confounding effect of multiple variables, a multiple logistic regression analysis was done. Variables considered risk factors for wound infection after cesarean delivery were included in univariate and logistic regression analyses.
Wound infection was diagnosed in 11 of 140 women (7.8%) after cesarean. The demographic and intrapartum variables of infected and uninfected groups are listed in Table 1. The groups were similar in selected demographic characteristics, presence or duration of ruptured membranes, number of vaginal examinations, and frequency of chorioamnionitis. Perioperative data are shown in Table 2. Univariate analysis showed a significantly greater subcutaneous tissue thickness, maternal weight, and body mass index (BMI) in infected women compared with uninfected women. The groups were similar in type of skin and uterine incisions, elective or unscheduled cesareans, and operation times. The median length of hospital stay for all subjects was 4 days, ranging from 2 to 7 days.
Multiple logistic regression analysis confirmed that only a single independent risk factor—maximum thickness of subcutaneous tissue—was significantly associated with abdominal wound infection after cesarean, with a relative risk (RR) of 2.8 (95% confidence interval [CI] 1.3, 5.9). Whereas nine of 11 infected women (81.8%) had subcutaneous tissue depths greater than 3 cm, 38 of 129 women (29.5%) who were not infected had thicknesses greater than 3 cm.
Aerobic cultures were available from the exudate of five of 11 cases that met the criteria for wound infections. Escherichia coli was isolated from two cultures, Enterococcus species from two cultures, and both E coli and Enterococcus species from one culture. Anaerobic and mycoplasma cultures were not collected.
Obesity has long been regarded as a risk factor for abdominal wound infection.6–10 However, obesity has been equated with an overall increase in weight and not specifically with the patient's habitus. When obesity was calculated anthropometrically using skin-fold thickness, no significant association with wound infection could be found.11 Our data suggest that increased weight or BMI does not specifically increase patients' risks of wound infection, but that the thickness of subcutaneous tissue at the site of the incision does. Thus, a larger and heavier woman with the same subcutaneous tissue thickness of a smaller and lighter woman would have the same risk of wound infection. Our findings are similar to those reported by Soper et al5 from a sample of women who had abdominal hysterectomies, in which subcutaneous tissue thickness was the most significant risk factor for wound infection.
It is widely accepted that ischemic wounds heal poorly because of limited perfusion and delivery of fibroblasts and leukocytes. Previous investigators found that wounds with large unapproximated areas or “dead space” remained relatively hypoxic.12 Large unapproximated surfaces also might accumulate serosanginous fluid, which can act as an ideal culture medium for contaminating pathogens. Greater subcutaneous tissue thicknesses increase the likelihood of such an environment and might explain the association with wound infection seen in our study.
Other identifiable risk factors for wound infection after cesarean delivery have included increased vaginal examinations, prolonged rupture of membranes, and emergency surgeries.13,14 However, we were unable to confirm such associations in our population. The predominant pathogens isolated from a few of the wound infections in our population were E coli and Enterococcus species, which are consistent with the findings of previous investigators.15,16
Identifying subcutaneous tissue thickness as a significant risk factor for postoperative wound infection after cesarean is not useful as a diagnostic test. In our study, the positive predictive value of subcutaneous tissue thickness of at least 3 cm was only 19% (nine of 47), with a negative predictive value of 98% (91 of 93). Most women with subcutaneous tissue measurements of at least 3 cm will not have wound infections more than 80% of the time.
Several mechanical prevention strategies for wound infection have been studied, including closed-suction drainage and closure of the subcutaneous tissue, with varying degrees of success.3,17 Soisson et al18 conducted a randomized trial of subcutaneous retention sutures versus skin closure alone in women with subcutaneous tissue thickness of at least 5 cm after gynecologic laparotomy. In that study, the investigators found a significant reduction in the frequency of superficial wound separation in women with retention sutures.18 Perioperative antibiotic prophylaxis during cesarean has been widely accepted, but increasing resistance among bacterial pathogens soon might limit obstetricians' options for wound-infection prophylaxis. Perhaps with the evolution of many antibiotic-resistant pathogens, the focus of wound-infection prevention might change to mechanical strategies in which subcutaneous tissue thickness will be important. On the basis of the findings in the present study, we are currently conducting a randomized trial of closed-suction drainage versus observation after cesarean for the prevention of wound infection in women with a subcutaneous tissue depth of at least 4 cm.
1. Emmons SL, Krohn M, Jackson M, Eschenbach DA. Development of wound infections among women undergoing cesarean section. Obstet Gynecol 1988;72:559–64.
2. Webster J. Post-caesarean wound infection: A review of the risk factors. Aust N Z J Obstet Gynaecol 1988;28:201–6.
3. Loong RLC, Rogers MS, Chang AMZ. A controlled trial on wound drainage in caesarean section. Aust N Z J Obstet Gynaecol 1988; 28:266–9.
4. Gibbs RS, Blanco JD, St Clair PJ. A case-control study of wound abscess after cesarean delivery. Obstet Gynecol 1983;62:498–501.
5. Soper DE, Bump RC, Hurt WG. Wound infection after abdominal hysterectomy: Effect of the depth of subcutaneous tissue. Am J Obstet Gynecol 1995;173:465–71.
6. Pitkin RM. Abdominal hysterectomy in obese women. Surg Gynecol Obstet 1976;142:532–6.
7. Davidson AIG, Clark C, Smith G. Postoperative wound infection: A computer analysis. Br J Surg 1971;58:333–7.
8. Garrow JS, Hastings EJ, Cox AG, North WR, Gibson M, Thomas TM, et al. Obesity and postoperative complications of abdominal operation. BMJ 1988;297:181–6.
9. Cruse PJE, Foord R. A five-year prospective study of 23,649 surgical wounds. Arch Surg 1973;107:206–10.
10. Beattie PG, Rings TR, Hunter MF, Lake Y. Risk factors for wound infection following caesarean section. Aust N Z J Obstet Gynaecol 1994;34:398–402.
11. Shapiro M, Munoz A, Tager IB, Schoenbaum SC, Polk BF. Risk factors for infection at the operative site after abdominal or vaginal hysterectomy. N Engl J Med 1982;307:1661–6.
12. Kivisaari J, Vihersaari T, Renvall S, Niinikoski J. Energy metabolism of experimental wounds at various oxygen environments. Ann Surg 1975;181:823–8.
13. Martens MG, Kolrud BL, Faro S, Maccato M, Hammill H. Development of wound infection or separation after cesarean delivery. Prospective evaluation of 2,431 cases. J Reprod Med 1995;40:171–5.
14. Nice C, Feeney A, Godwin P, Mohanraj M, Edwards A, Baldwin A, et al. A prospective audit of wound infection rates after caesarean section at five West Yorkshire hospitals. J Hosp Infect 1996;33:55–61.
15. Roberts S, Maccato M, Faro S, Pinell P. The microbiology of post-cesarean wound morbidity. Obstet Gynecol 1993;81:383–6.
16. Sherman D, Lurie S, Betzer M, Pinhasi Y, Arieli S, Boldur I. Uterine flora at cesarean and its relationship to postpartum endometritis. Obstet Gynecol 1999;94:787–91.
17. Del Valle GO, Combs P, Qualls C, Curet LB. Does closure of the camper fascia reduce the incidence of post-cesarean superficial wound disruption? Obstet Gynecol 1992;80:1013–6.
18. Soisson AP, Olt G, Soper JT, Berchuck A, Rodriguez G, Clarke-Pearson DL. Prevention of superficial wound separation with subcutaneous retention sutures. Gynecol Oncol 1993;51:330–4.