Qualitative manipulation of amino acid supply during total parenteral nutrition in surgical patients

Abstract

We investigated whether a qualitative manipulation of amino acid mixture could improve the nutritional status of patients undergoing surgery. Patients received total parenteral nutrition for 2 consecutive 5-day periods. Energy and nitrogen supplies were calculated using a modified Harris-Benedict equation, with an energy-to-nitrogen ratio of 125 kcal/g nitrogen (N). The mean kilocalorie lipids-to-glucose ratio was 35% to 65%. There were 6 patients in the control group and 7 patients in the experimental group. The control group received the same standard amino acid solution (Vintene, Baxter-Clintec, Maurepas, France) for the entire 10 days. The experimental group received the standard solution during the first 5 days but was switched to a more individualized solution during the last 5 days. The second solution was determined from a linear regression performed on day 3 comparing rate of infusion of each amino acid and its plasma variations after 3 hours of infusion with basal values. Amino acids were defined as oversupplied or undersupplied when not within the 95% confidence interval (above or below the curve, respectively). Daily nitrogen balance and urinary excretion of 3-methylhistidine were measured from day 1 to day 10 in all patients. For the patients in whom these measures were made, amino acid variations except those of lysine (11 of 12), ornithine (6 of 12), alanine (5 of 12), arginine (5 of 12), and glutamate (5 of 12) were within the 95% confidence interval. During the second 5-day period, imbalances persisted in the control group but were almost gone in the experimental group. Daily nitrogen balance was not significantly different between groups. However, when expressed as a mean over each period, nitrogen balance was significantly higher during the second period in the experimental group than in the control group: 4.5+/-0.8 g N/d versus 0.2+/-0.7 g N/d, p < .01. The ratio of urinary 3-methylhistidine to creatinine decreased from day 1 to day 10 in both groups. These findings suggest that the relationship between rate of infusion and plasma amino acid variation may offer a rational basis for choosing the most appropriate amino acid mixture for catabolic patients. An appropriate mixture would limit plasma amino acid imbalances and improve nitrogen retention. This therapy must now be tested in a larger population of patients.