Hepatotoxicity related to total parenteral nutrition: comparison of low-lipid and lipid-supplemented solutions

Because it is unclear whether or not the lipid or carbohydrate component of total parenteral nutrition solutions determines the development of cholestasis during total parenteral nutrition, a prospective randomized clinical trial of a predominantly carbohydrate solution (group I) versus one with isocaloric substitution of 30% nonprotein calories with lipid (group II) was performed.

Twelve patients in group I, 4 female and 8 male, and 14 patients in group II, 8 female and 6 male, were studied. Each subject received 100% to 150% of the basal energy needs, determined by the Harris-Benedict equation. Amino acids were supplied to provide a protein:kilocalorie ratio of 1:150 to 200. There were no differences between the groups in terms of mortality, nitrogen balance, and maintenance of weight. Weekly serum aspartate aminotrasferase (AST) and alkaline phosphatase (AP) levels were obtained. The change from baseline values was calculated.

The observed difference between the two groups for AST was 5.7 (higher for group II), P = .55, and for AP, 39.1 (higher for group I), P = .23. A power calculation was performed to determine the number of patients required for a statistically significant difference in AP between the two groups with an alpha of 0.05 and a power of 0.85:67 patients.

With these statistical considerations, we conclude that there was probably no statistically significant difference between the groups for an increase in AST or AP levels during total parenteral nutrition.