N. Narueponjirakul1,2, K. A. Breen1, M. El Hechi1, N. Kongkaewpaisarn1, G. C. Velmahos1, D. R. King1, M. B. Tabrizi1, P. J. Fagenholz1, N. N. Saillant1, A. E. Mendoza1, H. M. Kaafarani1, M. G. Rosenthal1 1Massachusetts General Hospital,Division Of Trauma, Emergency Surgery & Surgical Critical Care / Department Of Surgery / Harvard Medical School,Boston, MA, USA 2King Chulalongkorn Memorial Hospital,Department Of Surgery/ Faculty Of Medicine/ Chulalongkorn University,Pathumwan, BANGKOK, Thailand
Introduction: Obesity is a known risk factor for SSI, however, body mass index (BMI) as a metric of obesity, does not reliably predict SSI. Surgical site infections (SSIs) in emergency colon operations are common and lead to poor outcomes, increased healthcare costs, and decreased quality of life. We hypothesize that abdominal wall obesity as measured by abdominal wall thickness (AWT), and not BMI would serve as a better predictor of SSI in emergency colon operations.
Methods: We retrospectively evaluated our institutional Emergency Surgery Database (2007-2018). Emergency colon operations for any indication were included. ‘Emergency’ status was defined as procedures occurring within 12 hours of arrival to the emergency department or procedures booked for the operating room as emergent or urgent. Elective operations, procedures without preoperative imaging, patients requiring reoperations, temporary abdominal closure, presence of enterocutaneous fistula or active infection of abdominal wall, mortality within 7-days of operation, and skin wounds left open were excluded. AWT was measured in five locations on pre-operative CT scans: 1) right anterior superior iliac spine, 2) midpoint between xyphoid and umbilicus, 3) 2-cm inferior to the umbilicus, 4) midpoint between the umbilicus and pubic symphysis, and 5) lateral border of the rectus muscle at umbilical level. An average AWT among the five locations was calculated. Only superficial and deep SSI were considered as SSI in the analysis. Univariate and multivariable analyses were used to determine predictors of SSI.
Results: 241 patients out of 14,753 patients in the database were included. The median BMI was 25.8kg/m2 [22.4-30.1]. SSI occurred in 61 patients (25.3%). Median AWT between patients with (2.1cm [1.4, 2.8]) and without (1.8cm [1.2, 2.5]) a SSI was significantly different (p=0.047). SSI occurred in 25 (20%) patients with average AWT<1.8cm and 36 (30%) patients with average AWT ≥1.8cm. BMI and AWT were correlated (Pearson’s r=0.6821). On univariate analyses, wound classification, elevated SGOT level >40U/L, and AWT ≥1.8cm at 2 locations (2 cm below umbilicus and lateral border of rectus muscle at level of the umbilicus) were associated with SSI (p= 0.001, 0.018, 0.013, and 0.029, respectively), whereas BMI was not significantly associated (p=0.326). On multivariate analysis, wound classification, elevated SGOT level >40U/L, diabetes, and AWT ≥1.8cm at 2 cm inferior to umbilicus were independent predictors of SSI (OR 1.74, 95% CI 1.03-2.94, p=0.039; OR 3.23; 95% CI 1.31-7.97, p=0.011; OR 3.58, 95% CI 1.53-8.39, p=0.003; OR 2.47,95% CI 1.16-5.29, p=0.020, respectively).
Conclusion: AWT is a better predictor of SSI than BMI. Preoperative imaging of AWT may direct intraoperative decision making regarding wound management. Future clinical outcomes research in emergency surgery should include abdominal wall thickness as an important patient variable.