In the United States in 2010, 33% of all deliveries were performed by cesarean, ie, 1.3 million cesarean deliveries per year.1 The cesarean delivery rate in China is as high as 46%.2 Despite being the most common major operation performed annually in the United States, there is a paucity of data to guide best practices for closure of the cesarean delivery skin incision.
Most commonly, a cesarean delivery is performed through a suprapubic low-transverse skin incision.3,4 At the end of the operation, the skin incision is typically closed with the placement of either a continuous subcutaneous suture that dissolves over time or multiple metal staples that are removed at a later date.3,4 A recent Cochrane review comparing suture with staples for cesarean delivery skin closure concluded that future studies are needed that take into consideration the effect of body mass index (BMI, calculated as weight (kg)/[height (m)]2) and primary compared with repeat cesarean delivery in the setting of optimization of operative techniques to decrease the risk of wound complications, eg, preoperative antibiotic prophylaxis, spontaneous placental removal, and closure of the subcutaneous space when it is at least 2 cm in depth.3,5–7 Our objective was to evaluate whether the wound complication rates after cesarean delivery differ between women whose skin incisions are closed with suture as compared with staples.
MATERIALS AND METHODS
This was a prospective, multicenter, randomized clinical trial conducted at Thomas Jefferson University Hospital in Philadelphia, Pennsylvania, Lankenau Medical Center in Wynnewood, Pennsylvania, and Yale-New Haven Hospital in New Haven, Connecticut. Institutional review board approval was obtained from each hospital. Active enrollment for the study was from 2010 to 2012; the trial was registered on ClinicalTrials.gov (Identifier: NCT01211600).
Women who were candidates for a cesarean delivery through a low-transverse skin incision for delivery of a neonate greater than or equal to 23 weeks of gestation were offered inclusion by an obstetric provider either during their prenatal visits (typically for those undergoing planned cesarean delivery) or when they presented to Labor and Delivery. Women were excluded for poorly controlled diabetes (defined as greater than or equal to 50% of fasting or 2-hour postprandial blood glucose fingersticks greater than 120 and 160, respectively, within the week before delivery), current cesarean delivery through a vertical skin incision, chronic steroid use, active lupus flare, human immunodeficiency virus or acquired immunodeficiency syndrome, current treatment for cancer, a history of radiation to the abdomen or pelvis, current treatment with immunosuppressant medications, or hypersensitivity to either closure material. At Yale, age younger than 18 years was also considered an exclusion criterion as deemed by the institutional review board. Women who signed institutional review board-approved consent preoperatively were randomized during the cesarean delivery.
Randomization was based on a one-to-one computer-generated schema in random-sized blocks (size 4, 6, and 8) stratified within each center by current BMI and primary compared with repeat cesarean delivery: BMI less than 30 with primary cesarean delivery, BMI less than 30 with repeat cesarean delivery, BMI 30 or greater with primary cesarean delivery, and BMI 30 or greater with repeat cesarean delivery. Sequentially numbered opaque envelopes were used to mask the allocation until fascial closure, at which time randomization and allocation were performed.
Consenting patients were randomized to closure of the skin incision with subcuticular continuous 4-0 suture (poliglecaprone or polyglactin) or stainless steel staples. Staples were removed on postoperative days 4–10 at the discretion of the health care provider. Regardless of the material used for skin closure, skin manipulation was performed with Adson forceps. All women had Steri-Strips placed at the time of suture placement or staple removal. The patient was instructed to remove them herself if they did not fall off within 7 days.
Before the cesarean delivery skin incision, all women received abdominal preparation with surgical solution (chlorhexidine or iodine solution) and intravenous antibiotics. Entry into the abdomen was by a low-transverse skin incision. Typically, the placenta was delivered spontaneously. The abdominal closure technique was performed per the discretion of the attending surgeon and typically, the peritoneum was not closed. Once the fascia had been reapproximated, the circulating nurse retrieved the first envelope from the randomization schema. Subcutaneous tissue was closed with plain or chromic gut interrupted suture if the depth of this layer was at least 2 cm.
The primary outcome was the rate of wound complications. Patients with any wound concerns were evaluated by their obstetric provider. Wound complications were evaluated as a composite of: wound infection requiring antibiotics, hematoma, seroma (at least moderate serous drainage, as documented in the medical record), 1 cm or longer disruption in incision (separation), and readmission to the hospital for a wound concern. Incidence of wound complications was determined by review of all office and emergency department records for every patient at 6 weeks postpartum to assess whether the patient was diagnosed with any of the aforementioned wound complications. This was performed by a core group of masked investigators who did not review the operative report and were not aware of how the patient's incision was closed, unless it was documented in the postpartum visit. Additionally, patients were surveyed at their 4- to 6-week postpartum visit to assess if they were evaluated for a wound concern. For patients who did not present for the postpartum visit or who followed up at an outside institution, a phone interview was conducted for survey completion. Wound complication was listed as present if a patient had one or more complications, eg, a patient with a hematoma and separation would be listed once in the category of total patients with at least one wound complication. Prespecified secondary outcomes were total operative duration and duration of skin closure, number of suture or staple packets required for skin closure, length of hospital stay, diagnosis of endomyometritis, preoperative to postoperative hemoglobin difference, and number of wound-related visits within 6 weeks postoperatively. The duration of skin closure was defined as the time from completion of fascial closure to completion of the skin closure.
Prespecified subgroup analyses included: cesarean delivery as related to labor (before labor compared with after the onset of labor, stratified by primary and repeat cesarean delivery); type of suture material used (braided suture compared with monofilament suture); placement of incision with regard to the pannus (above compared with below); and after exclusion of patients with chorioamnionitis.
For the first year of the study, women who declined randomization were asked to provide consent for review of their demographic factors to help assess whether enrollment bias was present. After enrollment ceased, the data were explored for data entry errors and any such findings were corrected. To further assess the accuracy of the data entry, a random sample of 60% of the charts was double-entered to ensure that data entry errors were at a minimum.
The data analyses for this study were generated using SAS 9.3. All analyses were performed based on the intention-to-treat principle and a P value <.05 was considered significant. Unadjusted statistical differences between study groups with regard to details about the labor and delivery process were compared by Pearson χ2 and Fisher's exact test for categorical variables and Student's t test and Mann–Whitney U test for continuous variables, as appropriate for the data distribution. All categorical variables were reported as count and percent; and continuous data were expressed as median and interquartile range, unless otherwise noted.
All general linear models were stratified by randomization group and adjusted for hospital by including it as a covariate. Exact logistic regression was used for the analysis of wound complications, so adjusted odds ratios are reported as the measure of effect from that analysis. The odds ratio approximates the relative risk in this setting. Poisson regression was used for the analysis of number of wound-related visits and a generalized estimating equation with a robust variance estimator was used for the analysis of length of separation.
An estimation of wound complication rate postcesarean delivery of 11%8 was used to calculate a sample size that would detect a 50% reduction in wound complications with an α of 0.05 and a power of 80%. This required a sample size of 850 women total. We planned to enroll up to 468 women per study arm to account for a potential 10% of study participants to be lost to follow-up. However, as a result of relocation of the study principal investigator and decreasing enrollment, the study was stopped early, after enrolling 746 women. This decision was determined by the site principal investigators in collaboration with biostatisticians. We did not analyze any data until after the decision was made to stop the study.
Of 1,671 patients screened for participation, 32 were excluded and 269 declined. Therefore, 1,370 women consented to enrollment, of which 746 delivered by cesarean and were randomized (Fig. 1). Three hundred seventy women were allocated to closure with suture (49 Vicryl or Vicryl PLUS; 310 Monocryl or Monocryl PLUS; 11 suture unknown) and 376 to closure with staples.
Demographics were comparable between the two groups (Table 1). Forty-four percent of the participants were African American and 42% Caucasian. The vast majority of women were carrying singleton (94%), term (87%) gestations; 66% were parous, 63% had a BMI 30 or greater, and 48% did not have any prior cesarean deliveries. There were no demographic or medical history differences between women who declined randomization and those who participated (data not shown). There were no material differences between the groups with respect to indication for cesarean delivery, preoperative chorioamnionitis, treatment with prophylactic antibiotics, location of incision as compared with the pannus (when present), estimated blood loss, closure of subcutaneous tissue, or number of staple or suture packets used (Table 1). Of note, all patients received prophylactic antibiotics preoperatively; the majority (89%) had 1–2 g of cefazolin intravenously (Table 1). None of the patients underwent cesarean hysterectomy or had incisional drains placed. The overall incidence of wound complications was 7.8%. Those closed with suture were 57% (4.9% compared with 10.6%; adjusted odds ratio [OR] 0.43, 95% confidence interval [CI] 0.23–0.78) less likely to have a postoperative wound complication compared with those closed with staples (Table 2). There were no significant differences between the groups with respect to occurrence of a wound infection, hematoma, seroma, or readmission for a wound complication; however, those closed with suture were 80% less likely to have a wound separation of 1 cm or greater.
Duration of operation and duration of closure were significantly longer (by approximately 8 and 9 minutes, respectively) in those closed with suture (Table 3). There were no statistically significant differences between groups with respect to hospital stay, treatment with anticoagulation within 24 hours of the cesarean delivery, diagnosis of endomyometritis, or change in hemoglobin levels from preoperative values to postoperative values (Table 4).
Two hundred sixteen (57.6%) patients returned for a visit to remove the staples; the remainder had staples removed before hospital discharge. The median number of days at which staples were removed was 6 (interquartile range 4–9 days). Although women closed with suture were 40% less likely to require an additional visit as those closed with staples, even when not diagnosed with a wound complication (Table 2), when the visit for staple removal was not included in the overall wound-related visits, there was no significant difference between the groups in wound-related visits (P=.94).
There were no differences in the distribution of complications when comparing those cesarean deliveries that were performed prelabor compared with those that were performed after the onset of labor (P=.96, Breslow–Day test) nor were there any differences when timing of labor was stratified by primary and repeat cesarean delivery (Table 1). Additionally, there were no significant differences when comparing wound complications between women with two compared with three prior cesarean deliveries or three compared with four prior cesarean deliveries. There were no differences when comparing braided suture with monofilament suture (6.1% compared with 4.5%, respectively, P=.70, Pearson's χ2). Only 17 women were reported to have cesarean delivery incisions placed above the pannus, so we did not perform this prespecified secondary analysis. When we analyzed only those women who did not have chorioamnionitis, there remained a significant difference (P<.01, Pearson's χ2) in wound complication rates between those closed with suture (5.1%) and those closed with staples (10.1%). Although likely not clinically significant, the median length of separation in each group was 1.5 cm for sutures and 2.0 cm for staples (P=.04, using a generalized estimating equation model on log-transformed length of separation and adjusting for hospital and group). The majority of patients who had a wound separation either healed spontaneously or had Steri-Strips placed. Two patients required wound packing (one in the suture group, one in the staple group). One patient had a wound vacuum placed. One patient had an evisceration and one had a wound dehiscence; both had initially been closed with staples and both were readmitted for management of these complications.
In our study, in which prophylactic antibiotics were consistently administered before skin incision, suture closure of the skin at cesarean delivery was associated with a significant (57%) decrease in the incidence of wound complications compared with staples, from 10.6% with staples to 4.9% with suture. In particular, wound separation of the skin was significantly decreased to 1.6% from 7.4% in women whose incisions were closed with suture compared with those closed with staples. The decreases in the individual rates of wound infection, hematoma, seroma, or readmission for a wound complication in the suture compared with the staple group were not statistically significant; however, our study was not powered to assess differences in these individual complications. Duration of time required for skin closure was 9 minutes longer with suture compared with staples.
To adequately compare our study to prior literature, we performed a systematic review of the literature through MEDLINE through PubMed, EMBASE, ClinicalTrials.gov, and Cochrane using the search terms “cesarean,” “suture,” “staples,” “wound infection,” and “wound complications” in April 2013 without any restrictions on language. We reviewed the results of our search to identify randomized clinical trials that compared nonabsorbable staples compared with absorbable suture; we identified seven such trials (see Appendix 2, available online at http://links.lww.com/AOG/A485, which depicts an overall summary of the published trials).9–15
The majority of the initial trials compared patient satisfaction, pain perception, and cosmetic outcome of the closure techniques. A recent Cochrane review of six of the seven trials determined that closure with suture compared with staples may have a different effect based on whether the skin incision was vertical or low-transverse: five of the seven studies excluded vertical skin incisions (see Appendix 2, http://links.lww.com/AOG/A485).6 Despite these studies, a recent survey shows that many practitioners still use staples for closure of cesarean delivery incisions.16
The multicenter nature of this study increases its external validity; the study population was diverse with respect to race, marital status, and insurance source.
In conclusion, our study showed a 57% decreased incidence of wound complications when the cesarean delivery skin incision was closed with suture rather than with staples. Although closure with staples is significantly faster than closure with suture, in cases in which speed is not a serious concern, we recommend that the horizontal cesarean delivery skin incision be closed with suture in an effort to decrease cesarean delivery morbidity related to wound complications.
1. Martin JA, Hamilton BE, Ventura SJ, Osterman MJK, Wilson EC, Mathews TJ. Births: final data for 2010. Atlanta (GA): National Center for Health Statistics; 2012. p. 61.
2. Lumbiganon P, Laopaiboon M, Gulmezoglu AM, Souza JP, Taneepanichskul S, Ruyan P, et al.. Method of delivery and pregnancy outcomes in Asia: the WHO global survey on maternal and perinatal health 2007-08. Lancet 2010;375:490–9.
3. Berghella V, Baxter JK, Chauhan SP. Evidence-based surgery for cesarean delivery. Am J Obstet Gynecol 2005;193:1607–17.
4. Dahlke JD, Mendez-Figueroa H, Rouse DJ, Berghella V, Baxter JK, Chauhan SP. Evidence-based surgery for cesarean delivery: an updated systematic review. Am J Obstet Gynecol 2013;209:294–306.
5. Anorlu RI, Maholwana B, Hofmeyr GJ. Methods of delivering the placenta at caesarean section. The Cochrane Database of Systematic Reviews 2008, Issue 3. Art. No.: CD004737. DOI: 10.1002/14651858.CD004737.pub2.
6. Mackeen AD, Berghella V, Larsen ML. Techniques and materials for skin closure in caesarean section. The Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD003577. DOI: 10.1002/14651858.CD003577.pub3.
7. Costantine MM, Rahman M, Ghulmiyah L, Byers BD, Longo M, Wen T, et al.. Timing of perioperative antibiotics for cesarean delivery: a metaanalysis. Am J Obstet Gynecol 2008;199:301.e1–6.
8. Johnson A, Young D, Reilly J. Caesarean section surgical site infection surveillance. J Hosp Infect 2006;64:30–5.
9. Frishman GN, Schwartz T, Hogan JW. Closure of pfannenstiel skin incisions: staples vs subcuticular suture. J Reprod Med 1997;42:627–30.
10. Gaertner E, Burkhardt T, Beinder E. Scar appearance of different skin and subcutaneous tissue closure techniques in caesarean section: a randomized study. Eur J Obstet Gynecol Reprod Biol 2008;138:29–33.
11. Rousseau JA, Girard K, Turcot-Lemay L, Thomas N. A randomized study comparing skin closure in cesarean sections: staples vs subcuticular sutures. Am J Obstet Gynecol 2009;200:265.e1–4.
12. Cromi A, Ghezzi F, Gottardi A, Cherubino M, Uccella S, Valdatta L. Cosmetic outcomes of various skin closure methods following cesarean delivery: a randomized trial. Am J Obstet Gynecol 2010;203:36.e1–8.
13. Basha SL, Rochon ML, Quinones JN, Coassolo KM, Rust OA, Smulian JC. Randomized controlled trial of wound complication rates of subcuticular suture vs staples for skin closure at cesarean delivery. Am J Obstet Gynecol 2010;203:285.e1–8.
14. Rengerink KO, Mol BW, Pajkrt E, de Graaf I, Wiersma I, Donker M. Techniques for wound closure at caesarean section: a randomized controlled trial. Am J Obstet Gynecol 2011;204:S267.
15. Figueroa D, Jauk VC, Szychowski JM, Garner R, Biggio JR, Andrews WW, et al.. Surgical staples compared with subcuticular suture for skin closure after cesarean delivery: a randomized controlled trial. Obstet Gynecol 2013;121:33–8.
16. Mackeen AD, Devaraj T, Baxter JK. Cesarean skin closure preferences: a survey of obstetricians. J Matern Fetal Neonatal Med 2013;26:753–6.
17. Schulz KF, Altman DG, Moher D; CONSORT Group. CONSORT 2010 statement: updated guidelines for reporting parallel group randomized trials. Obstet Gynecol 2010;115:1063–70.