Quantification of free water transport in peritoneal dialysis

BACKGROUND: In peritoneal dialysis (PD) total net ultrafiltration (NUF) is dependent on transport through small pores and through water channels in the peritoneum. These channels are impermeable to solutes, and therefore, crystalloid osmotic-induced free water transport occurs through them. Several indirect methods to assess free water transport have been suggested. The difference in NUF between a 3.86% and a 1.36% solution gives a rough indication, but is very time consuming. The magnitude of the dip in dialysate/plasma (D/P) sodium in the initial phase of a 3.86% exchange is another way to estimate free water transport. In the present study, a method was applied to calculate free water transport by calculating sodium-associated water transport in one single 3.86% glucose dwell. METHODS: Forty PD patients underwent one standard peritoneal permeability analysis (SPA) with a 1.36% glucose solution, and another with a 3.86% glucose solution. At different time points intraperitoneal volume and sodium concentration were assessed. This made it possible to calculate total sodium transport. By subtracting this transport (which must have occurred through the small pores) from the total fluid transport, free water transport remained. These results were compared with the other methods to estimate free water transport. RESULTS: For the 1.36% glucose dwell, total transcapillary ultrafiltration in the first hour (TCUF(0-60)) was 164 mL, transport through the small pores was 129 mL, and free water transport was 35 mL (21%). For the 3.86% glucose solution, total TCUF(0-60) was 404 mL, transport through the small pores was 269 mL, and free water transport was 135 mL (34%). The contribution of free water transport in the first minute (TCUF(0-1)) was 39% of the total fluid transport. From the 40 patients, 11 patients had ultrafiltration failure (NUF <400 mL after 4 hours). For these patients the contribution of free water to TCUF(0-1) was significantly lower than for those with normal ultrafiltration (20% vs. 48%, P < 0.05). A strong correlation was present between free water transport as a percentage of total fluid transport and the maximum dip in D/P sodium (r= 0.84). The correlation was not significant with the difference in net ultrafiltration of 3.86% and 1.36% solutions (r= 0.24, P= 0.3). CONCLUSION: The method applied here is the first direct quantification of free water transport, calculated from a single standard peritoneal function test. It offers a quick possibility to evaluate patients suffering from ultrafiltration failure. In these patients free water transport was impaired, but the origin of this impairment is still to be determined