The primary objective of this trial is to measure changes in hepatic lipid content using three common alternative therapeutic strategies to improve glycemic control in patients with type 2 diabetes who are not controlled with metformin alone.

In patients with type 2 diabetes whose glycemic control is not accomplished with Metformin alone, there are several options for additional hypoglycemic agent. As per recent National guidelines at least three "second line" agents can be added to metformin: thiazolidinediones, Lantus insulin, and DPP4 inhibitors. All three approaches have been proven to help accomplishing goals of therapy for glycemic control (HbA1c\<7%) in clinical trials. Whether one approach is superior to the other in improving associated non-glycemic metabolic abnormalities and risk for future morbidity and mortality in patients with type 2 diabetes remains a matter of intense debate. Among these abnormalities, excessive triglyceride content in the liver (fatty liver or liver steatosis) is a major predictor of risk for non-alcoholic steato-hepatitis (NASH) and liver cirrhosis. Fatty liver is highly prevalent in patients with type 2 diabetes. There are no clinical studies done to determine if any of these three therapeutic options is superior at reducing the liver triglyceride content in patients with type 2 diabetes and fatty Liver. Therefore, in this study we propose to measure and compare the effects on liver triglyceride content when either pioglitazone, lantus insulin, or DPP4 inhibitors are added to metformin in patients with poorly controlled type 2 diabetes (HbA1c\>7.6%) and fatty liver (presence of "bright" liver in abdominal ultrasound). The primary objective of this trial is to measure changes in hepatic lipid content using three common alternative therapeutic strategies to improve glycemic control in patients with type 2 diabetes who are not controlled with metformin alone.