Wound Closure Techniques

Introduction:

Much of the practice of surgical wound closure and subsequent care is based upon tradition and individual preference rather than science. Most patients undergoing cesarean deliveries at Hartford Hospital have a subcuticular running closure. Occasionally, patients have a staple closure, at the discretion of the operating surgeon. With newly introduced wound closure techniques, we intend to compare an approved closure technique (Dermabond) with the more traditional suture closure via a prospective, observer blinded, randomized, controlled trial.

Methods:

Study participants will include all women who provide informed consent prior to undergoing cesarean section at Hartford Hospital. Patients will be excluded from the study if they will require a vertical skin incision; are allergic to Dermabond; or have a temperature > 100.2 F preop or intraop. Qualified patients will then be randomized to skin closure with either the Dermabond wound closure technique or suture.

Closure method randomization will be performed prospectively by placing closure modality options in opaque sealed envelopes prior to the initiation of the study. The envelopes will be available in the operating room. The circulating nurse will open the envelope to reveal the randomization arm prior to skin closure and the surgeon will utilize the indicated technique. The circulating nurse will record the length of time needed to close the skin incision. The surgeon will complete a short survey immediately post operatively regarding ease of use of the closure modality as well as expected cosmetic outcome.

If a patient is randomized to suture, surgeon preference will dictate whether 4-0 vicryl or 4-0 monocryl will be used as well as whether Keith or curved needle will be utilized. Patient demographics and contact information will be obtained prior to initiation of the procedure.

Prior to initiation of surgery, the operative field will be prepared by dry shaving the site with a disposable razor then cleansing the abdomen with Betadine using sterile technique. Hibiclens soap will be substituted for patients who are sensitive or allergic to iodine. A first generation Cephalosporin will be administered ~ 30 minutes preop for patients undergoing scheduled cesarean delivery. If the procedure is unscheduled, a first generation Cephalosporin will be administered after cord clamp if the patient has not already received antibiotics preoperatively.

The patients' post operative pain medicine requirements and pain self assessment scores will be monitored during their post operative hospital course as documented in the medication logs on their charts. Most patients will be discharged on postoperative day three or four unless complications arise.

All patients will be seen in the office for a 6-week post partum follow up. A medical professional will assess their wounds. The patient and the medical professional will be queried regarding their satisfaction with cosmetic outcome of the wound. The patient will also be queried regarding continued pain medicine requirements. If the patient does not show up for their post partum visit, they will be called to obtain this information and to arrange a time for follow-up.

Each medical professional will be provided with a disposable camera. The camera will be used to take pictures of each patient's wound at the 6-week follow up visit. Specific guidelines will be provided regarding the distance and angle at which the pictures should be taken to ensure optimal and uniform photo clarity. The pictures will be reviewed at a later time by a panel of observers who will be blinded to the type of wound closure that the patient received. The observer will assess each picture for wound cosmesis and healing to provide standardized objective evaluation of wound closure outcomes.

Power and Sample Size:

The primary outcomes will be surgical time (wound closure only), patient satisfaction and blinded observer-rated cosmesis. Using a 1:3 randomization scheme, group sample sizes of 25 (test/Dermabond) and 75 (control/suture) would achieve 80% power to detect a difference of 1.0 between the null hypothesis that both group means for satisfaction and cosmesis are 3.0 ("average") and the alternative hypothesis that the mean of the Dermabond group is 2.0 ("above average") with estimated group standard deviations of 1.5 and with a significance level (alpha) of 0.05 using a two-sided, two-sample t-test.

Assuming the average closure time for the suture group is 6 minutes (SD +/-1), and hypothesizing an average closure time in the Dermabond group of 1.5 minutes (SD +/- 0.5 minutes), these sample sizes would afford better than 99% power to detect differences in surgical (closure) time between the two groups.

The Women's Health Department at Hartford Hospital performs an average of 12 procedures per week (range 10-15). If 70% of the eligible women consent to participate, it should take approximately 12 weeks to accrue the minimum number of subjects. However, if the participation rate is 80%-85%, only 10-11 weeks should be required for accrual.

Data Analysis / Statistics:

Descriptive statistics on continuous variables (e.g., height, weight, maternal age, gestational age, surgical (closure) time) will consist of means, standard deviations and 95% confidence intervals. Frequencies of race and ethnicity also will be run. To evaluate the primary outcomes between treatment arms, differences in closure time, patient satisfaction and observed healing, t-tests will be utilized. Subgroup comparisons (e.g., within the suture group between curved and Keith needles) also will be performed with t-tests. Secondary analyses of dichotomous (yes/no) variables, including surgeon responses to "Did you find this method effective?" and "Did you find this method burdensome?" will be evaluated with a c2 test. Although infections may occur and will be noted, the rate is anticipated to be so low that statistical assessment will not be necessary. Statistically significant differences will be considered as analyses resulting in p<0.05. All analyses will be performed with SPSS v. 11.5 (Chicago, IL, 2003).