AbstractObjectives:Questions about reliability of bioimpedance analysis (BIA) in morbidly obese subjects have curtailed its use in this setting, but metabolic implications might reignite the debate. In a prospective study, it was aimed to analyze anthropometric and clinical associations.Methods:Bariatric candidates (n = 94) with or without metabolic syndrome were consecutively investigated. Age was 34.9 ± 10.4 years (68.1% females), and BMI was 40.8 ± 4.6 kg m−2. Methods included single‐frequency BIA, anthropometrics, inflammatory indices, and general biochemical profile.Results:Body composition results (water, fat) in females, but not in males, were entirely consistent with the literature. In both genders good association was observed with anthropometrics (BMI, waist circumference), inflammatory indices (ferritin, C‐reactive protein) and general biochemical variables. Anthropometric measurements also displayed comparable associations. Multivariate tests including the two sets of measurements indicated no predominance of one method over the other, one complementing the other as metabolic marker.Conclusions:BIA limitations were mostly relevant for males, not females. Despite such discrepancies, good associations with anthropometry were demonstrated for both genders. Correlations with liver enzymes, and indices of protein, carbohydrate, and lipid metabolism could be demonstrated. BIA deserves more investigations concerning liver steatosis and ongoing inflammation, and it could contribute as well, synergistically with anthropometry, to monitor weight loss, body fat shifts, and metabolic risk. Am. J. Hum. Biol., 2011. © 2011 Wiley‐Liss, Inc.