ABSTRACT
The evaluation of insulin resistance (IR) in clinical practice is based on the determination of fasting insulin (I0) and insulin level after 2 hours in an oral glucose tolerance test (OGT). However, there are not adequate cutoff points to discriminate IR patients. Objectives: to evaluate the reliability of insulin levels in the diagnosis of IR using the intravenous insulin tolerance test (IVITT) as the gold standard. Patients and Method: The OGT and IVITT of patients who participated as cases or controls in research protocols were analyzed. We excluded those cases with fasting glycemia over126 mg/dl. Results: 128 cases, 111 F, 17 M; Age: 40.3 +/- 14.8 years; BMI: 33 +/- 8 kg/m2; Waist circumference, M: 100.3 +/- 9.4 cm, F: 96 +/- 15 cm. According to IVITT (KITT), 103 (80.5 percent) were IR (KITT < 4.5 percent) and 25 (19.5 percent) were non IR (KITT > 4.5 percent). Fasting (G0) and 120 minutes after glucose challenge glycemia (G120), I0 and I120, HOMA and area under the glycemia and insulin curve, were significantly higher in the IR, as the same as, hypertension and acanthosis nigricans features (p < 0.05). According to G120, 45 cases (35.2 percent) had glucose intolerance, 9 (7 percent) diabetes and 74 (57.8 percent) were normals. In addition to G0, only IVITT was significantly different among the 3 groups (p = 0.025), identifying most insulin resistant subjects. The sensitivity and specificity for a cutoff point of I120 at 60 µIU/mL, were 30 percent and 88 percent, respectively. Conclusion: Baseline and 120 minutes post glucose charge insulin levels and HOMA, do not discriminate insulin resistant subjects, especially when there is fasting or post-stimulus hyperglycemia. Therefore, they are not recommended for individual diagnosis or therapeutic decisions