Objectives: Current evidence regarding metabolic surgery suggests that different types of digestive tract reconstruction can result in differences in postoperative glucose tolerance. This study evaluated the impact of Billroth I (B-I), Billroth II (B-II), and Roux-en-Y (R-Y) procedures on peri-operative glucose tolerance in patients with gastric carcinoma who had diabetes mellitus. Methods: A single-institution, retrospective cohort study was conducted using data from patients who underwent totally laparoscopic distal gastrectomy. These patients were grouped according to the type of reconstruction (B-I, B-II, or R-Y). After the operation, we addressed the changes in glucose tolerance—including changes in HbA1c levels, remission of diabetes, and overall effects of the treatment. Results: We studied 57 patients (B-I, n=32; B-II, n=17; R-Y, n=8). B-II and R-Y reconstruction improved HbA1c levels more than B-I. Notably, R-Y improved tolerance the most (B-I vs. B-II, p<0.001; B-I vs. R-Y, p<0.001; B-II vs. R-Y, p<0.001). The type of reconstruction (B-II and R-Y vs. B-I) and a pre-operative HbA1c ≥7% were the two significant independent contributing factors determining postoperative improvement in HbA1c, with odds ratio (OR) 8.437, 95% confidence interval (CI) 1.635–43.527, p=0.011; OR 16.5, 95% CI 3.361–81.011, p=0.001, respectively. Conclusions: Either R-Y or B-II should be considered the primary option for patients with gastric carcinoma and diabetes when glycemic control is insufficient.