Intraperitoneal nutritional support: initial results in a canine model.

To evaluate the capacity of the peritoneal membrane to efficiently absorb complex nutritional solutions, five adult mongrel dogs were prepared with peritoneal and intravenous catheters. In duplicate studies each dog's peritoneal catheter was rapidly infused with 1 liter of fluid containing 1.25% glucose, 5% fat emulsion, 2.5% mixed amino acids, and 20 ml of standard electrolyte additive solution. Plasma and peritoneal fluid nutrient and electrolyte concentrations and osmolarities were measured initially, 1/2 hr, and hourly for 5 hr after infusion. Mean peritoneal fluid volume increase was 56 +/- 45 ml at 5 hr (P less than 0.05). Plasma osmolarity remained unchanged as peritoneal fluid osmolarity rapidly declined to plasma level over the initial 2 hr. Peritoneal fluid concentrations of glucose, triglycerides, and amino acids all fell precipitously during the first hour (P less than 0.01) while plasma concentrations rose in reciprocal fashion (P less than 0.01). Maximal rises above basal plasma concentrations all occurred at 1/2 hr and amounted to 42.7% for glucose, 39.7% for triglycerides, and 113.2% for amino acids. Levels returned to normal or near normal by 5 hr. Net absorption of glucose was 91.7% of the total glucose load after 5 hr of intraperitoneal dwell time. Triglyceride absorption was 65.1% and amino acid absorption was 91.9%. Greater than 75% of total nutrient absorption occurred during the initial hour of dwell. It is estimated that both nonprotein energy and nitrogen requirements could be met with two to three intraperitoneal nutrient solution exchanges per day. Thus, provision of total nutritional needs may be possible with an intraperitoneal support system.