Background: Body weight loss (BWL) is frequently observed in gastric cancer patients who underwent gastrectomy for gastric cancer. Recently, we reported that severe BWL after gastrectomy was a significant risk factor for continuation of S-1 adjuvant chemotherapy. However, risk factors of BWL after surgery remain unclear. Methods: The present study retrospectively examined the patients who electively underwent curative gastrectomy for gastric cancer between January of 2012 and June of 2014. All patients received peri-operative care of ERAS protocol. %BWL was calculated by percentile of body weight at one month after surgery to preoperative body weight. Severe BWL was defined as %BWL over 10%. Risk factors for severe BWL were determined by both univariate and multivariate logistic regression analyses. Results: Two-hundred seventy eight patients were examined. Median age (range) was 68 years (27-86). Median body mass index (range) was 22 (13.4-33.5). Thirty patients had received neoadjuvant chemotherapy. Operative procedure was total gastrectomy (TG) in 97 patients (Open in 61 and laparoscopic in 36) and distal gastrectomy (DG) in 181 patients (Open in 94 and laparoscopic in 87). Median operation time (range) was 268.5 minutes (95-554). Median blood loss (range) was 115 mL (0-1600). Morbidity of grade 2 or more defined by Clavien-Dindo classification was observed in 37 patients including pancreatic fistula in 8, anastomotic leakage in 5, and abdominal abscess in 3. No mortality was found. Completion of ERAS protocol without any variance was 95.3%. Median %BWL was 6% (-4.3% to 19.5%). Both univariate and multivariate logistic analyses demonstrated that morbidity (odds rate 3.56, p=0.001), blood loss over 300ml (odds rate 2.04, p=0.0356), and total gastrectomy (odds rate 2.1, p=0.0258) were significant risk factors for severe BWL. Conclusions: Nutritional intervention trial to inhibit BWL after gastrectomy should be focused on the patients who developed morbidity, showed blood loss over 300ml, or received total gastrectomy.