Abstract

Create and validate diverticulitis surgical site infection prediction scale. Surgical site infections cause significant morbidity after colorectal surgery. An infection prediction scale could target infection prevention bundles to high-risk patients. Prospectively collected National Surgical Quality Improvement Program and electronic medical record data obtained on diverticulitis colectomy patients across a Healthcare Network-wide Colorectal Surgery Collaborative (5 hospitals). Patients with and without surgical site infections were compared. Predictive variables were identified using logistic regression model; model estimates obtained through 1000 bootstrap replications for scale validation. A total of 1737 colectomies were performed (2010-2016): mean age 59.9 years (SD 12.7), 56.4% female; 93.4% Caucasian; smokers 16.3%, diabetics 7.7%, steroid use 6.0%. Two hundred thirty-one (13.3%) were presented to operating room emergently and 138 (7.9%) with abscess at time of disease admission. Two hundred ninety-six patients underwent Hartman procedures, and 113 (6.5%) received diverted primary anastomosis. Average length of stay was 6.9 days (standard deviation 7.01), 30-day mortality was 1.5%, anastomotic leak rate was 3.1%. Twenty-one percent of patients (n = 366) developed a surgical site infection. Several predictors for infection were identified: obesity (body mass index >30), advanced age (>70 years), diabetes mellitus, preoperative abscess, open surgery, emergent operations, and prolonged operations (>3 h). Creation of protected anastomosis in emergent settings was associated with increased infection rates. Presence of more than 5 risk factors was associated with infection rates of 45.8% (c = 0.69). Patients with diverticulitis have high surgical site infection rates due to nonmodifiable risk factors. Our Prediction and Enaction of Prevention Treatments Trigger scale can risk stratify patients for targeting surgical site infection prevention bundles and outcomes risk adjustments.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call