How do you measure insulin resistance?

The most accurate method is the euglycaemic hyperinsulinaemic clamp. In this procedure, insulin is administered intravenously while blood sugar is kept artificially stable. The amount of glucose the body then requires directly reflects how sensitive it is to insulin. The method is the most reliable, but takes several hours and requires specialised equipment. One practical point to keep in mind: the result must be corrected for muscle mass, not total body weight, otherwise insulin sensitivity in people with excess weight will be underestimated.

One step less burdensome is the oral glucose tolerance test, in which you drink a sugar solution and then have blood drawn at multiple time points. An index is calculated from the glucose and insulin values, such as the Matsuda index. This provides a reasonable estimate of how the body responds after a meal, but is less accurate than the clamp. Mixed meal tests are also being studied increasingly often for this purpose.

In day-to-day practice, HOMA and QUICKI are more popular. Both calculate insulin resistance from fasting blood values, are inexpensive and straightforward to perform. QUICKI has been compared extensively with the clamp and also predicts the risk of developing diabetes later in life. The drawback of both is that they say nothing about how the body handles insulin after a meal. In addition, the cut-off values are not universal: they differ by age, sex and population group, so a 'normal value' that applies to one group does not necessarily apply to another.

The TyG index, calculated from fasting triglycerides and glucose, is even less expensive and in one small study (99 participants) showed reasonable agreement with the clamp. This is an interesting indication, but it is too early for broad use as a diagnostic tool.