Data were published that superseded this study.

Gastric bypass surgery has been proposed as an effective treatment strategy for type 2 diabetes mellitus (T2DM), but thus far algorithms for the management of T2DM immediately after surgery have not been established. The investigators designed an algorithm to achieve good glycaemic control in patients with insulin-requiring T2DM upon discharge after gastric bypass surgery, and evaluated this algorithm for feasibility, efficacy, and safety.

This is a prospective randomised controlled intervention study on obese human subjects with Type 2 Diabetes Mellitus undergoing obesity surgery (n=80). Patients will be recruited from those already due to undergo bariatric surgery within the context of the obesity clinic (Imperial College London). Patients will be randomised to either of two treatment protocols.

Protocol A-Conservative Management: On admission to hospital all pre-operative oral antidiabetic agents and insulin are stopped and glucose monitoring is carried out every four hours at the same time as routine observations. Actrapid insulin is administered with glucose as per standard sliding scale (Appendix A). Just prior to discharge, total insulin administered over the previous 24 hour post-operative period is calculated. Based on the results of ward based glucose monitoring, post-operative patients can be divided in two broad categories.

The first group includes patients who are able to maintain euglycaemia without exogenous insulin administration. These patients are discharged home on their preoperative Metformin dose and are advised to continue monitoring their fasting glucose levels at home. If fasting glucose levels of less than 4 or more than 10mmol/l are recorded at home, the patients are encouraged to contact the obesity clinic physicians for advice.

The second group includes patients experiencing hyperglycemic episodes in the immediate post-operative period. These patients are discharged home on their preoperative Metformin dose and once daily insulin Glargine as decided by the discharging doctor (or based on half their last 24 hour requirements) and are advised to continue monitoring their fasting glucose levels at home. If fasting glucose levels of less than 4 or more than 10mmol/l are recorded at home, the patients are encouraged to contact the obesity clinic physicians for advice.

Protocol B-Intensive management: On admission to hospital all pre-operative oral antidiabetic agents and insulin are stopped and glucose monitoring is carried out every four hours at the same time as routine observations. Actrapid insulin is administered as per standard sliding scales (appendix A). Just prior to discharge, total insulin administered over the previous 24 hour post-operative period is calculated. Based on the results of ward based glucose monitoring patients are discharged home on metformin 1000mg BID and/or the equivalent amount of Glargine insulin requirement in previous 24 hours prior to discharge. These patients are also advised to continue glucose monitoring at home and will be contacted every day and their medication adjusted to maintain a fasting glucose between 5 and 7mmol/l. When this is consistently achieved the investigators will stop calling, but patients will be encouraged to continue monitoring their blood sugars and contact the investigators should they experience any problems with their glycaemic control.

Venesection will take place at 10 days, 3, 6 and 12 months post-operatively and yearly thereafter for fasting glucose and HbA1c as part of the routine care of patients of the obesity clinic. As part of clinical management we will ensure that all patients with diabetes will have a kidney (2 samples of morning urine albumin creatinine ratio), nerve (nerve conduction study, including thermal threshold testing) and eye (retinal photograph) tests once before and within 1, 2 and 3 years after surgery. These tests are all non invasive and except from the nerve tests are performed routinely for all patients with diabetes at least yearly and if performed in the community they will not be repeated.Therefore, other than the nerve conduction studies, there will be no extra burden to the patients as a result of the study.

Inclusion Criteria:

• Body mass index (BMI) > 35,
• Insulin-requiring T2DM,
• Age >18yr,
• Having undergone gastric bypass surgery,
• HbA1C >8%

Exclusion Criteria:

• Non-insulin requiring T2DM,
• HbA1C <8%