Alexis O-Ring Wound Retractor for the Prevention of Post-cesarean Surgical Site Infections

The study design was a randomized controlled study to evaluate wound infection rates in patients undergoing a cesarean delivery. The patients were randomized into two sub- groups; the first group received routine hand-held metal retractors as needed by the surgical team, and the second group received an Alexis Wound Retractor. Inclusion criteria included: maternal age of at least 18 years, a gestational age of at least 24 0/7 wks by best obstetric estimate, a viable gestation, a non-emergent cesarean delivery, and a planned transverse skin incision. Deliveries included: primary cesarean for failed induction, maternal request, or other obstetric indications; and repeat cesarean deliveries that were elective or had obstetric indications (such as spontaneous labor in a subject with a term gestation and with history of 2 or more prior cesarean deliveries). Exclusion criteria included chorioamnionitis, chronic corticosteroid therapy, a prior abdominal vertical skin incision scar or planned vertical incision, history of a prior wound infection or separation, a planned cesarean hysterectomy, or maternal or fetal complications requiring an emergent delivery.

Randomization was done using a web-based randomization generator. Variables were allocated as: 1 for Alexis, and 0 for conventional surgical retractors. Packages including consent forms, data sheets, and the randomized allocation were prepared for the entire study and were stored in one office. Study packets were prepared to include a sealed envelope indicating randomization to be opened after the consent form was signed. Only after randomization, the Alexis retractor was opened and placed on the surgical field.

Prior to initiation of the study, in-service sessions were conducted with OB surgical teams to facilitate the use of the retractors. Information provided included optimal use of the retractor and selection of the appropriate retractor size. The primary surgeon of each case was previously trained by the principle investigator in placement of the retractor on actual patients.

Once a decision was made to proceed with cesarean delivery, the subject was approached and offered to participate in the study. After the patient had been enrolled and consent forms signed, an allocated envelope for each study subject determined which arm of the study to which the patient would be randomized; no other management decisions was made as part of the RCT. Surgical decisions were independent of the patient's designated study arm and retractor type. Such decisions included: type of anesthesia, length of transverse skin incision , method of tissue dissection (blunt vs. instrumental), creation of bladder flap, size and nature of uterine incision (transverse, low vertical, classical, J-type, T-incision), surgical material used, exteriorization of the uterus, manual vs. spontaneous delivery of placenta, layer closure techniques (bladder flap, peritoneum, subcutaneous tissue), and skin closure technique (staples, subcuticular sutures). Skin preoperative preparation was performed as standard with ChloraPrep® Applicator (2% Chlorhexidine gluconate and 70% isopropyl alcohol). Intravenous antibiotics were given preoperatively in accordance to national guidelines (12). The primary outcome of this study was the development of a surgical site infection after cesarean delivery within two weeks after cesarean.

Incisions were checked daily in-house. However, the exams that were used in this study to determine the occurrences of SSI were the first exam done in the outpatient setting for removal of staples (post op days 5 - 7), and the two week post partum visit (post op days 14 - 18). SSIs were diagnosed according to the Centers for Disease Control and Prevention (CDC) criteria.(13). The most commonly seen characteristics were purulent drainage or localized swelling and/or redness of the surgical wound.

In addition, other wound complications were also assessed and included: seroma (wounds with clear drainage) and/or hematomas. Other secondary outcomes included skin time to uterine incision (minutes) , exteriorization of the uterus(yes, no), total operative time (minutes), estimated blood loss (mL) estimated by the primary surgeon, and postoperative pain based on a subjective 1-10 pain scale, assessed 4 times daily while hospitalized and at the outpatient visits previously mentioned.

An initial sample size estimate was performed using the frequentist approach with an assumption of power 80% and Type 1 error = 5% (2-sided). The primary outcome of the study is expected to occur in 15% of women in the "usual care" group. In order to detect a 50% reduction (15% to 7.5%) in the rate of wound complications with the use of the Alexis retractor, a total sample size of 564 women will be needed (282 per group). The planned study was to be carried out at two distinct sites: the University of Texas Medical Branch in Galveston (UTMB) and the University of Texas in Houston. The Institutional Review Board of UTMB-Galveston IRB # 10-188 approved the study.