SAN FRANCISCO — Surrogate measures of insulin resistance, while not nearly as reliable, may be far more practical for routine clinical use than the hyperinsulinemic-euglycemic clamp and the insulin suppression test, which can take 3 hours or more.
At the Third World Congress on Insulin Resistance Syndrome, Dr. Sun H. Kim, of Stanford (Calif.) University, discussed available surrogate tests. Other speakers at the congress discussed two new instrumental measures likely to become available soon.
One problem with using any measure of insulin resistance is its continuous distribution in the general population. There is no absolute criterion by which to classify individuals as insulin resistant or insulin sensitive. Dr. Kim follows the general practice of defining individuals who fall in the upper tertile of insulin resistance as having insulin resistance syndrome.
Available surrogate measures fall into three categories—routine measures, additional measures, and calculated measures.
The routine measures include fasting glucose and components of the lipid panel, especially HDL cholesterol and triglycerides.
Compared with the steady-state plasma glucose level (SSGL) derived from an insulin suppression test, fasting glucose has a correlation coefficient of just 0.38 and HDL has a correlation coefficient of just −0.41.
Triglyceride level is a somewhat better predictor of SSGL, with a correlation coefficient of 0.60.
Further, these three routine measures have an additional complication—their correlation coefficients vary depending on the patient's level of obesity.
The additional measures include fasting plasma insulin and the area under the curve for insulin in a 2-hour oral glucose tolerance test (OGTT). Fasting plasma insulin has a fairly good correlation with SSGL—0.61—but it, too, varies with the patient's obesity.
Among the routine and additional measures, the area under the curve of insulin in an OGTT has the highest correlation with SSGL (0.79), and that correlation does not vary depending on the patient's level of obesity.
The calculated measures include the ratio between triglycerides and HDL cholesterol, the homeostasis model assessment of insulin resistance (HOMA-IR), and the quantitative insulin sensitivity check index (QUICKI).
A triglyceride to HDL ratio above 3 predicts SSGL with a correlation coefficient of 0.61, and this varies only modestly with obesity. This is likely to be the most frequently used measure in clinical practice. HOMA-IR and QUICKI make use of fasting insulin and fasting glucose levels. Both correlate well with insulin sensitivity, but not much better than fasting insulin alone.
The two new instrumental measures, however, promise to provide more sensitive and specific measures of insulin sensitivity.
The deuterated-glucose disposal test (
Deuterated water is measured with an isotope ratio mass spectrometer, a large instrument that costs about $250,000. But Dr. Marc Hellerstein of the University of California, Berkeley, hopes to be able to employ a smaller, less expensive laser spectrometer that would have the additional advantage of a high throughput.
The second instrumental measure is a
According to Dr. Richard Z. Lewanczuk of the University of Alberta, Edmonton, human studies have demonstrated that the breath test has a sensitivity of 78% and a specificity of 96%, compared with standard measures. The Food and Drug Administration recently approved this technique for use in clinical trials in the United States.
Dr. Lewanczuk disclosed that he is a major stockholder in Isotechnika Diagnostics, the company that hopes to market the