Background: Malnutrition is a widespread but largely unrecognized problem in aged patients, more so as it is a contributing factor to the increased morbidity and mortality in this age group. Since direct measurements of body composition are not possible in a large number of patients, good anthropometric reference data are fundamental in assessing the nutritional state of elderly people. Objective: To examine the efficacy of calf circumference (CC) measurement for assessing the nutritional state of the elderly. Methods: The nutritional state was assessed by anthropometric and biological measurements in 911 elderly patients consecutively admitted to a geriatric unit. In the first instance, univariate analysis was performed for CC and other nutritional parameters. Linear and multiple stepwise regressions were performed to study the association between anthropometric or biological parameters and CC. At a later stage, the specificity, sensitivity and the optimal cutoff CC were established for 2 groups of patients: malnourished and controls. Finally, patients were classified according to this cutoff in order to verify the efficacy of CC in assessing their nutritional state. Results: The results of univariate analysis showed significant correlations between CC and other nutritional anthropometric markers (r = 0.706, p < 0.0001 with body mass index (BMI) and r = 0.661, p < 0.0001 with fat free mass) and biological markers (r = 0.219, p < 0.0001 with albumin and r = 0.162, p < 0.0001 with transthyretin). Multiple regression confirmed associations between CC and tricipital skinfold thickness (p < 0.0001), fat free mass (p < 0.0001), BMI (p < 0.0001), and serum albumin (p < 0.0001; r2 = 0.561). The optimal cutoff for CC was found to be 30.5 cm for both men (sensitivity 73.2%, specificity 72.8%) and women (sensitivity 78.8%, specificity 61.1%). The classification of patients according to this discriminating factor was confirmed for the optimal cutoff value. Conclusion: Calf circumference is a pertinent marker of nutritional state. The cutoff of 30.5 cm provides a good diagnostic capacity.