Effect of Glucose Degradation Products on the Peritoneal Membrane in a Chronic Inflammatory Infusion Model of Peritoneal Dialysis in the Rat

Abstract

Long-term use of the peritoneal membrane as a dialyzing membrane is hampered by its eventual deterioration. One of the contributing factors is glucose degradation products (GDPs) in the dialysis solution. In this study, we evaluated the effect of a low GDP solution on peritoneal permeability, the structural stability of the peritoneal membrane, and vascular endothelial growth factor (VEGF) production in a chronic inflammatory infusion model of peritoneal dialysis (PD) in the rat. Male Sprague-Dawley rats were divided into 3 groups: a conventional solution group (group C, n = 12), a test solution group (group T, n = 12), and a normal control group (group NC, n = 8). Group T rats were infused with low GDP solution (2.3% glucose solution with two compartments), and group C rats with conventional dialysis solution (2.3% glucose solution), adjusted to pH 7.0 before each exchange. Animals were infused through a permanent catheter with 25 mL of dialysis solution. In both groups, peritoneal inflammation was induced by infusing dialysis solution supplemented with lipopolysaccharide on days 8, 9, and 10 after starting dialysate infusion. Peritoneal membrane function was assessed before and 6 weeks after initiating dialysis using the 1-hour peritoneal equilibration test (PET) employing 4.25% glucose solution. Both VEGF and transforming growth factor beta1 (TGFbeta1) in the dialysate effluent were measured by ELISA. The number of vessels in the omentum was counted after staining with anti-von Willebrand factor, and the thickness of submesothelial matrix of the trichrome-stained parietal peritoneum was measured. Peritoneal tissue was analyzed for VEGF protein using immunohistochemistry. At the end of 6 weeks, the rate of glucose transport (D/D0, where D is glucose concentration in the dialysate and D0 is glucose concentration in the dialysis solution before it is infused into the peritoneal cavity) was higher in group T (p < 0.05) than in group C. Dialysate-to-plasma ratio (D/P) of protein was lower in group T (p < 0.05) than in group C; D/P(urea), D/P(sodium), and drain volumes did not differ significantly between groups C and T. Dialysate VEGF and TGFbeta levels were lower in group T (p < 0.05) than in group C. Immunohistochemical studies also revealed less VEGF in the peritoneal membranes of group T. There were significantly more peritoneal blood vessels in group C (p < 0.05) than in group T, but the thickness of submesothelial matrix of the parietal peritoneum was not different between the two groups. The VEGF levels in the dialysate effluent correlated positively with the number of blood vessels per field (r = 0.622, p < 0.005). Using a chronic inflammatory infusion model of PD in the rat, we show that dialysis with GDP-containing PD fluid is associated with increased VEGF production and peritoneal vascularization. Use of low GDP solutions may therefore be beneficial in maintaining the function and structure of the peritoneal membrane during long-term PD.