Background: There is no established insulin regimen in T2DM patients receiving parenteral nutrition. Aims: To compare the effectiveness (metabolic control) and safety of two insulin regimens in patients with diabetes receiving TPN.

Almost one third of patients do not achieve type 2 diabetes remission after bariatric surgery or are unable to sustain this effect long term. Our objective was to delve further into the dynamic responses of diabetes after bariatric surgery and to evaluate the "time-within-remission range" as a variable of metabolic control. A descriptive cohort study was done using a computerised multicentre and multidisciplinary registry. All data were adjusted by propensity score. A total of 1186 subjects with a follow-up of 4.5 ± 2.5 years were included. Type of surgery, diabetes remission, recurrence of diabetes, "time-within-remission range" and key predictors of diabetes outcomes were assessed. All patients (70% women, 51.4 ± 9.2 years old, body mass index (BMI) 46.3 ± 6.9 kg/m 2 ) underwent primary bariatric procedures. "Time-within-remission range" were 83.3% (33.3-91.6) after gastric bypass, 68.7% (7.1-87.5) after sleeve gastrectomy and 90% (83.3-92.8) after malabsorptive techniques (p < 0.001 for all). Duration of diabetes, baseline HbA1c and insulin treatment were significantly negatively correlated with the "time-within-remission range". The association of bariatric techniques with "time-within-remission range", using gastric bypass as a reference, were: odds ratio (OR) 3.70 (2.34-5.84), p < 0.001 for malabsorptive techniques and OR 0.55 (0.40-0.75), p < 0.001 for sleeve gastrectomy. Characteristics of type 2 diabetes powerfully influence the outcomes of bariatric surgery. The "time-within-remission range" unveils a superiority of gastric bypass compared to sleeve gastrectomy.