A New Scoring System Using Preoperative Factors and Contour Mapping for Predicting Postoperative Complications of Laparoscopic Gastrectomy

Background: Surgical risk assessment is becoming more important in the field of laparoscopic gastrectomy given the increasing complexity and technical demands of oncological procedures undertaken. However, no validated method for preoperative risk (PR) assessment has been reported. Methods: Two-hundred fourteen patients who underwent laparoscopic gastrectomy between October 2011 and August 2014 were reviewed. Independent risk factors were examined by multivariate analysis and a PR score model was established. Results: Thirty-six (16.8%) patients experienced postoperative complications. Multivariate analysis revealed that age ≥75 years, American Society of Anesthesiologists score ≥2, severe pulmonary disease, Brinkman index ≥600, history of upper abdominal surgery and body mass index (BMI) were independent factors. Among these, BMI was strongly correlated with postoperative morbidity rate (r = 0.960). To permit risk prediction using BMI and the remaining 5 factors, a new PR score was established using 5 categorical variables. The new score was significantly correlated with the postoperative complication rate (r = 0.948), and its predictive value was superior to conventional risk scores such as E-PASS and POSSUM. A contour map was then created using BMI and this new PR score to more accurately stratify patients according to surgical risk prior to laparoscopic gastrectomy. Conclusions: A 2-dimensional risk estimation model was created to better estimate surgical risk and is a promising new tool for adequate surgical management.