inhibition of hepatic gluconeogenesis; however, clinical and pre- clinical data suggest that metformin has a direct endothelium- protective action(s). Methods: We investigated the effects of metformin on endothelium- dependent vasodilation (EDV) in blood vessels from control and dia- betic mice as well as on endothelial nitric oxide synthase (eNOS) in endothelial cells in culture. Results: 50 μM metformin significantly (P<0.05) enhanced acetylcholine-mediated EDV in thoracic aortae from db/db dia- betic mice and in aortae from control mice maintained in high glucose (HG; 40 mM) Krebs. Data from mouse microvascular endothelial cells (MMECs) demonstrate that the endothelial protective effects of a 3-hour exposure to metformin against HG-induced reduction of serine 1177 phosphorylated eNOS (p-eNOS) are associated with a signifi- cant enhancement of p-eNOS via a mechanism that, based on a siRNA (AMPKa1/2 siRNA) protocol, is partially dependent on AMPK. Expo- sure of MMECs to HG for 72 hours reduced expression of sirtuin 1, the protein product of the anti-ageing “longevity” gene, SIRT1 and enhanced senescence-associated β-galactosidase activity; however, metformin prevented these pro-ageing effects of HG. HG altered the expression of a number of microRNAs (miRNA), including miR- 34a, and a miR-34a inhibitor mimicked the effects of metformin and enhanced the expression of sirtuin 1 whereas the overexpression of miR-34a prevented the protective actions of metformin. Conclusions: Metformin has a direct protective effect on the endo- thelium that is dependent on the anti-ageing gene SIRT1 and involves regulation by miR-34a. 200 Evaluation of the Anti-Glucotoxic Effects of Leech Saliva Extract (LSE) on Mouse and Human Pancreatic Islets SANAZ MAHMOUDI, MAHSHID EBTIA, AHMED MERZOUK, SUZANNE STRATTON, REZA JALILI, AZIZ GHAHARY Vancouver, BC Introduction: Leech saliva extract (LSE), which has anticoagulant prop- erties as well as many peptides/proteins, has been shown to improve a variety of medical conditions including diabetes. Our research is designed to evaluate antiglucotoxic effects of LSE, derived from Hirudo medicinalis, on mouse and human pancreatic islets in vitro. Hypothesis: Treatment of pancreatic islets with LSE prevents/ reverses detrimental effects of glucotoxicity. Methods: Human pancreatic islets were received from the Ike Barber Human Islet Transplant Center (Vancouver, BC) in accordance with approved procedures and guidelines. Mouse pancreatic islets were isolated in our laboratory from 6- to 8-week-old male B6 mice. Results: MTT assays revealed ~ 20% increase in the viability of human islets treated with high concentrations (1 ug/mL) of LSE under glucotoxic conditions, compared to low or no LSE treatments. Simi- larly, mouse islets treated with continuous high concentrations of LSE showed ~25% increase and those treated simultaneously with high concentrations of LSE showed ~15% increase in viability. Moreover, GSIS was increased by more than 2-fold in mouse and human islets treated with high concentrations of LSE under high glucose conditions. Conclusion: Our results suggest that LSE partially prevents the effects of glucotoxicity on islets, in a dose-dependent manner, but does not reverse already established toxic effects. Both cell viability and insulin secretion are partially preserved after preventative LSE treatment under glucotoxic culture conditions. With further research, LSE might serve as a novel approach to prevention of diabetes mellitus pro- gression, namely at the pre-diabetic stage. 201 Withdrawn 202 Metabolic Outcomes Following Islet Autotransplantation VEENA AGRAWAL*, TATSUYA KIN, SHARLEEN IMES, A.M. JAMES SHAPIRO, PETER A. SENIOR Edmonton, AB Background: Loss of both insulin and glucagon after total pancreatectomy leads to brittle, insulin-dependent diabetes, with high risk of hypoglycemia. Islet autotransplantation (IAT) can prevent or ameliorate diabetes, but poor islet yields from dis- eased pancreata contribute to variable outcomes, and long-term data are lacking. Methods: We performed a retrospective chart review of all patients who underwent pancreatectomy with IAT at our institution between 2002 and 2016. Information regarding mortality, insulin use and metabolic factors was collected. Results: Forty patients received an IAT between 2002 and 2016. Mean age was 41 (range 3 to 84) years. Thirty-six underwent total or completion pancreatectomies. Indications for pancreatectomy were chronic or recurrent pancreatitis (57.5%), benign or malig- nant tumours (37.5%) and trauma (5%). Mean total islet equivalents (IE) transplanted was 286,768 (range 20,468 to 710,224) IE, with a mean yield of 4057 (range 301 to 10,276) IE/kg. Nine patients died during the follow-up period. Data for insulin use was avail- able for 22 patients; 55% of these patients were insulin independent at follow up ranging from 1.5 months to 14 years. Islet yield was not significantly associated with insulin independence. At most recent follow up, 52% of patients had diabetes on the basis of insulin use, FBG >7.0 mmol/L or HbA1c >6.5%. BETA-2 scores were calculated for 16 patients; 7 had scores >15. Conclusion: Outcomes following IAT are highly variable, with a high mortality rate predominantly in patients with malignancy. However, IAT prevents diabetes and avoids insulin in half. Further research is needed to clarify long-term outcomes. 203 Self-Efficacy, Self-Care and Glycemic Control among Canadian South Asians with Type 2 Diabetes ROWSHANAK AFSHAR*, FARDOWSA YUSUF, TRICIA S. TANG Vancouver, BC The purpose of this study was to evaluate the relationships between self-efficacy, self-care behaviours and glycemic control in South Asian adults with type 2 diabetes living in Canada. This study used a cross-sectional design of 41 South Asian adults with type 2 diabetes. Blood samples were collected to determine HbA1c and lipid profile values. The participants were asked to complete a self-report questionnaire consisting of items assessing sociodemo- graphic characteristics, diabetes self-care behaviours using the revised Summary of Diabetes Self-Care Activities Measure (SDSCA- revised) and diabetes-specific self-efficacy using the Stanford Self- Efficacy Scale. Participants’ blood pressure, height, weight and waist circumference were also measured. Bivariate correlations found positive relationships between physical activity self-care items and the total self-efficacy score (r=0.7104753, p<0.001). Some of the self-care behaviours including testing blood sugar, taking diabetes pills, eating fruits and vegetables, eating high fat foods and physical activity were related to sociodemographic and clini- cal measures. Checking feet had a positive relationship with total cholesterol (r=0.35, p=0.028). However, there were no significant relationships between self-efficacy, sociodemographic variables, HbA1c and other clinical measures. Abstracts / Can J Diabetes 40 (2016) S27–S74 S71