Introduction

Cesarean delivery is the most common major surgical procedure among women in the United States, and 5–12% are complicated by surgical site infection. Infections and other wound complications after cesarean are a substantial burden to both patients and the health care system. While costs may vary depending on the severity, the average attributable hospital cost is estimated at $3529 per wound infection after cesarean. When costs of home health services, indirect costs, and intangible costs are taken into account, the total cost of wound complications after cesarean is much higher.

Several randomized trials and metaanalyses have shown lower rates of wound infection and disruption after cesarean with subcuticular skin closure than with staple closure. However, the optimal choice of suture material for subcuticular skin closure after cesarean is unclear. Vicryl (polyglactin 910, a braided multifilament synthetic absorbable suture composed of a copolymer of glycolide and L-lactide; Ethicon, Somerville, NJ) and Monocryl (poliglecaprone 25, a monofilament synthetic absorbable suture made from a copolymer of glycolide and epsilon-caprolactone; Ethicon) are the commonly used suture materials for subcuticular closure of transverse skin incisions at cesarean in the United States. Whereas in vitro and animal studies suggest multifilament suture materials are considerably more susceptible to bacterial infiltration and may be associated with a higher risk of surgical site infection than monofilament sutures, clinical data on their relative effectiveness are limited.

Therefore, we conducted this analysis to test the hypothesis that subcuticular skin closure with Vicryl is associated with a higher rate of surgical site infection and other wound complications after cesarean compared to Monocryl.

Materials and Methods

This is a secondary analysis of data from a randomized trial in which pregnant women undergoing scheduled or unscheduled cesareans were randomly assigned to preoperative skin preparation with either chlorhexidine–alcohol or iodine–alcohol ( ClinicalTrials.gov no. NCT01472549 ). The study was approved by the institutional human studies review board prior to enrollment. All participants provided written informed consent. Pregnant women undergoing cesarean delivery at Washington University Medical Center in St Louis, MO, from September 2011 through June 2015 were eligible. We excluded women who had known allergy to chlorhexidine, alcohol, iodine, or shellfish, or had skin infection adjacent to the operative site. For this analysis we included only women with low transverse skin incisions who had subcuticular skin closure with either 4-0 Monocryl or 4-0 Vicryl on a Keith or PS-2 needle. Women who had vertical skin incisions, were closed with staples, or had no follow-up after discharge were excluded.

Patients received standard infection prevention measures including weight-based preoperative antibiotic prophylaxis. Per our institutional protocol, patients preoperatively received 2 g of cefazolin if they weighed <120 kg and 3 g if >120 kg. Patients with severe penicillin or cephalosporin allergy received 2 mg/kg of gentamicin plus 900 mg clindamycin. Choice of suture material was at the discretion of the operating physician. Patients were followed up daily until discharge from the hospital. Patients generally had a postoperative visit at approximately 2 weeks and a routine postpartum visit at 4–6 weeks. They were also contacted at approximately 30 days after delivery to assess symptoms of surgical site infection and inquire if they had a physician office or emergency room visit for wound symptoms. Medical records were obtained from physician office, emergency room visits, or hospital admissions and reviewed by the principal investigator, blinded to the study arm, to determine the diagnosis of surgical site infection or other wound complication at each postoperative visit or readmission. We collected demographic information, obstetric and medical history, and details of the surgical procedure. Data were collected by direct patient interview and supplemented with data abstraction from patient charts.

The primary outcome for the trial and this analysis was superficial or deep surgical site infection within 30 days of cesarean delivery based on the Centers for Disease Control and Prevention National Healthcare Safety Network definitions. Superficial surgical site infections were those that involved only the skin or subcutaneous tissue, while deep infections involved deeper soft tissues. Secondary outcomes were other wound complications including skin separation, seroma, hematoma, and cellulitis. The diagnoses of these complications were made by the treating physician and verified by chart review by the principal investigator.

We defined the comparison groups by subcuticular skin closure with 4-0 Monocryl or 4-0 Vicryl, and compared baseline characteristics using the unpaired Student t test or the Mann-Whitney U test for continuous variables and the χ ² or Fisher exact test for categorical variables. We calculated rates and unadjusted odds ratios (OR) with 95% confidence intervals (CI) for the primary and secondary outcomes. Multivariable logistic regression was used to calculate adjusted OR. The final models adjusted for antiseptic allocation, schedule/unscheduled cesarean, and subcutaneous layer closure. Model fit was assessed with the Hosmer-Lemeshow goodness-of-fit test.

We used time-to-event analysis to compare the timing of diagnosis of surgical site infections in the 2 groups. Kaplan-Meier estimates of freedom from surgical site infection were compared in the 2 groups using the log-rank test. We conducted subgroup analyses for the primary outcome and employed the Mantel-Haenszel test of homogeneity to assess if the relative effectiveness of suture material differed across subgroups (obese vs nonobese, diabetic vs nondiabetic, scheduled vs unscheduled cesarean, primary vs repeat cesarean, closure vs nonclosure of the subcutaneous layer). We included all patients from the primary study who met inclusion criteria; no a priori sample size estimation was performed. All tests were 2-tailed with P < .05 considered significant. Analyses were conducted using a software package (STATA, Version 12, Special Edition; Stata Corp, College Station, TX).

Materials and Methods

This is a secondary analysis of data from a randomized trial in which pregnant women undergoing scheduled or unscheduled cesareans were randomly assigned to preoperative skin preparation with either chlorhexidine–alcohol or iodine–alcohol ( ClinicalTrials.gov no. NCT01472549 ). The study was approved by the institutional human studies review board prior to enrollment. All participants provided written informed consent. Pregnant women undergoing cesarean delivery at Washington University Medical Center in St Louis, MO, from September 2011 through June 2015 were eligible. We excluded women who had known allergy to chlorhexidine, alcohol, iodine, or shellfish, or had skin infection adjacent to the operative site. For this analysis we included only women with low transverse skin incisions who had subcuticular skin closure with either 4-0 Monocryl or 4-0 Vicryl on a Keith or PS-2 needle. Women who had vertical skin incisions, were closed with staples, or had no follow-up after discharge were excluded.

Patients received standard infection prevention measures including weight-based preoperative antibiotic prophylaxis. Per our institutional protocol, patients preoperatively received 2 g of cefazolin if they weighed <120 kg and 3 g if >120 kg. Patients with severe penicillin or cephalosporin allergy received 2 mg/kg of gentamicin plus 900 mg clindamycin. Choice of suture material was at the discretion of the operating physician. Patients were followed up daily until discharge from the hospital. Patients generally had a postoperative visit at approximately 2 weeks and a routine postpartum visit at 4–6 weeks. They were also contacted at approximately 30 days after delivery to assess symptoms of surgical site infection and inquire if they had a physician office or emergency room visit for wound symptoms. Medical records were obtained from physician office, emergency room visits, or hospital admissions and reviewed by the principal investigator, blinded to the study arm, to determine the diagnosis of surgical site infection or other wound complication at each postoperative visit or readmission. We collected demographic information, obstetric and medical history, and details of the surgical procedure. Data were collected by direct patient interview and supplemented with data abstraction from patient charts.

The primary outcome for the trial and this analysis was superficial or deep surgical site infection within 30 days of cesarean delivery based on the Centers for Disease Control and Prevention National Healthcare Safety Network definitions. Superficial surgical site infections were those that involved only the skin or subcutaneous tissue, while deep infections involved deeper soft tissues. Secondary outcomes were other wound complications including skin separation, seroma, hematoma, and cellulitis. The diagnoses of these complications were made by the treating physician and verified by chart review by the principal investigator.

We defined the comparison groups by subcuticular skin closure with 4-0 Monocryl or 4-0 Vicryl, and compared baseline characteristics using the unpaired Student t test or the Mann-Whitney U test for continuous variables and the χ ² or Fisher exact test for categorical variables. We calculated rates and unadjusted odds ratios (OR) with 95% confidence intervals (CI) for the primary and secondary outcomes. Multivariable logistic regression was used to calculate adjusted OR. The final models adjusted for antiseptic allocation, schedule/unscheduled cesarean, and subcutaneous layer closure. Model fit was assessed with the Hosmer-Lemeshow goodness-of-fit test.

We used time-to-event analysis to compare the timing of diagnosis of surgical site infections in the 2 groups. Kaplan-Meier estimates of freedom from surgical site infection were compared in the 2 groups using the log-rank test. We conducted subgroup analyses for the primary outcome and employed the Mantel-Haenszel test of homogeneity to assess if the relative effectiveness of suture material differed across subgroups (obese vs nonobese, diabetic vs nondiabetic, scheduled vs unscheduled cesarean, primary vs repeat cesarean, closure vs nonclosure of the subcutaneous layer). We included all patients from the primary study who met inclusion criteria; no a priori sample size estimation was performed. All tests were 2-tailed with P < .05 considered significant. Analyses were conducted using a software package (STATA, Version 12, Special Edition; Stata Corp, College Station, TX).