OBJECTIVE: The purpose of this study was to evaluate the interobserver variability of ultrasound in the measurement of subcutaneous, visceral and perirenal fat through standard technique. MATERIALS AND METHODS: From November 2006 to January 2007, 50 patients were evaluated. The subcutaneous fat thickness was measured with a 7.5 MHz linear transducer transversely positioned 1 cm above the umbilical scar. For the visceral fat, a 3.5 MHz transducer was also positioned 1 cm above the umbilical scar, considering the distance between the internal surface of the abdominal rectus muscle and the posterior aortic wall in the abdominal midline. The perirenal fat was measured in the middle third of the right kidney, with the transducer positioned at the axillary midline. RESULTS: The t-Student test was utilized to analyze the interobserver reproducibility with significance level of 95%. No statistically significant difference was observed among mean values for subcutaneous, visceral and perirenal fat (p = 0.7141, 0.7286 and 0.6368, respectively). Mean and standard deviation corresponded to 2.64 ± 1.37 for subcutaneous thickness, 6.84 ± 2.38 for visceral fat, and 4.89 ± 2.6 for perirenal fat. CONCLUSION: Ultrasound presented a good interobserver reproducibility in the evaluation of abdominal fat based on measurement of subcutaneous, visceral and perirenal fat as parameters.

Keywords: Abdominal fat, Ultrasound, Metabolic syndrome