This is a pilot study to show that it is possible to identify the specific types of fats in blood, adipose tissue, and liver tissue. The study doctors hope to use the analysis of these fats to learn more about nonalcoholic fatty liver disease (NFLD). Nonalcoholic fatty liver disease is the accumulation of fat in the liver of people who have minimal alcohol exposure. Nonalcoholic Fatty Liver Disease is associated with obesity and insulin resistance, and predicts development of Type 2 Diabetes. The study doctors are interested in looking at the relationship between liver fat and insulin resistance.

Inclusion Criteria:

Subjects eligible to participate in the study will be:

1. Adults over age 18;
2. Patients enrolled in the bariatric surgery clinic who are planning to undergo and have insurance approval for bariatric surgery;
3. Insulin sensitive (based on HOMA-IR calculated from screening labs) OR insulin resistant (based on HOMA-IR and or personal history of diet controlled diabetes)
4. Able and willing to give informed consent.

Exclusion Criteria

The following will exclude subjects from study participation:

1. Clinically significant medical issues outside of diabetes (eg, pregnancy, cardiovascular disease, uncontrolled pulmonary disease). If female and of child bearing potential, a pregnancy test will be performed and confirmed negative prior to participation in the study and prior to the scheduled glucose clamp procedure.
2. A history of hematologic (platelets <100 x107/L), hepatic (liver function tests [LFTs] >2X upper limit of normal), renal (Creatinine >1.5 mg/dL), pulmonary/ cardiac abnormalities (ie, abnormal electrocardiogram [EKG]).
3. Use of anti-diabetes medication, anticoagulants, or medication that might alter lipids. If the subject has a history of diabetes, it should be controlled by diet alone.
4. Inflammatory or celiac intestinal disease;
5. Untreated thyroid disease;
6. Excessive alcohol use, illicit drug use, or active untreated psychiatric disease.

Data too small (n=4) to be shared publicly, although we will share deidentified data upon case-by-case request

2 years