OBJECTIVES To study in severe patients that required parenteral nutrition the relationship between Zn and Cu amounts present in formulas, the clinical course and changes in some biochemical parameters (plasmatic and erythrocytic Zn and Cu levels) during the parenteral nutrition therapy. PATIENTS AND METHODOLOGY Five adult severe patients were studied, submitted to major abdominal surgery, and that required parenteral nutrition. Determinations were done for: 1) Zinc and copper in parenteral formulas; 2) in patients, at the beginning (T0) and at the end (Tf) of treatment, in erythrocytes: zinc (Zn-E) and cupper (Cu-E); in plasma, zinc (Zn-Pl) and in serum, cupper (Cu-S). Zinc and cupper were determined by means of atomic absorbance spectrometry. RESULTS mean +/- standard deviation values and ranges (between brackets) were: parenteral formulas (microg/g): zinc: 0.6 +/- 1.1 (2.2 - 7.0); Cupper: 2.4 +/- 0.7 (0.5-3.7). Biochemical parameters: at T0 (n = 5): Zn-E (microg/mL): 21.6 +/- 10.0 (13.6-36.1); Zn-Pl (microg/dL): 88 +/- 72 (29-205); Cu-E (microg/dL): 113 +/- 22 (60-102); Cu-E (microg/dL): 139 +/- 29 (106 +/- 156); Cu-S (microg/dL): 172 +/- 20 (158-195). Individual values compared to reference ones (normal individual from Buenos Aires with adequate nutrition) indicated that the three patients with favorable course normalized Zn-Pl and Zn-E levels. However, Cu-S increased in the three cases and Cu-E in two of them. CONCLUSIONS These results make clear that in the studied patients Zinc levels in parenteral formulas would be adequate to prevent deficiency while Copper levels could be too high.