Open or closed? Management of skin incisions after emergent general surgery laparotomies

Author(s):
Brett Tracy; Julia Coleman; Holly Baselice; Shruthi Srinivas; Sara Scarlet; Rondi Gelbard

Background:

The management of midline abdominal skin incisions following an emergent laparotomy varies.  We sought to determine if there was a relationship between incisional management and postoperative wound complications among patients undergoing emergent laparotomy for general surgery.

Hypothesis:

We hypothesize that skin closure technique is not associated with adverse wound outcomes.

Methods:

We performed a retrospective review of emergency general surgery patients (>18 years) who underwent an exploratory laparotomy within 6 hours of surgical consultation.  Patients whose fascia was not closed during the index operation were excluded.  Patients were divided into groups: open skin (OS) and closed skin (CS).  Open skin included negative pressure wound therapy or wet-to-dry gauze; closed skin included closure with staples or sutures.  We collected data on demographics, hemodynamics, laboratory values, hollow viscus perforation (HVP), and drain placement.  Our primary outcome was rate of postoperative intra-abdominal abscess (IAA), superficial surgical site infection (SSSI), and incision dehiscence.

Results:

The cohort comprised 388 patients: 42.3% OS (n=164) and 57.7% CS (n=224).  There was no difference in age, BMI, or sex between groups, but there were more HVPs in the OS group (71.3% vs 20.5%, p<.0001).  OS compared to CS patients had greater heart rates (95.9 vs 88.2, p<.0001), WBC (15.3 vs 11.7, p=0.0001) but lower albumin (3.3 vs 3.7, p<.0001).  There was no difference in complications, IAA, SSSIs, or dehiscence between groups; however, OS patients had a greater rate of drain placement (51.2% vs 17.9%, p<.0001).  In a subgroup analysis of patients with HVP (n=163), there was no difference in complications but a greater rate of drain placement in the OS group (43.5% vs 60.7%, p=0.04).  For the overall cohort and subgroup of HVP, there was no difference in overall complications, IAA, SSSI, or dehiscence between patients with or without drains.  On multivariable logistic regression, temperature (OR 1.5, 95% CI 1.1-2, p=0.005), white blood cell count (OR 1.04, 95% CI 1.01-1.08, p=0.02), HVP (OR 6.7, 95% CI 3.7-12, p<.0001) and drain placement (OR 2.1, 95% CI 1.2-4, p=0.02) were significant predictors of OS (AUC 0.829).

Conclusions:

Open skin management occurs often after emergency general surgery laparotomies and is more common among patients with HVP, leukocytosis, and drain placement. However, rates of SSIs and wound dehiscence are similar between OS and CS patients.  Our results suggest that skin closure is not contraindicated in the presence of sepsis and HVP.