The Beneficial Influence on the Effectiveness of Automated Peritoneal Dialysis of Varying the Dwell Time (Short/Long) and Fill Volume (Small/Large): A Randomized Controlled Trial

Abstract

It is well known that the efficiency of peritoneal dialysis (PD) varies with the duration of the dwell and with the prescribed fill volume. Automated PD (APD) is classically given as a series of recurrent exchanges, each having the same dwell time and fill volume-that is, conventional APD (APD-C). We propose a new way of giving PD, using a modified version of APD-C. This method first uses a short dwell time with a small fill volume to promote ultrafiltration (UF) and subsequently uses a longer dwell time and a larger fill volume to promote removal of uremic toxins from the blood. We use the term "adapted APD" (APD-A) to describe this modified form of PD. We designed a multicenter prospective randomized crossover trial to assess the impact of APD-A in comparison with APD-C on the efficacy of dialysis. The parameters investigated were overnight UF; weekly peritoneal Kt/V(urea); weekly peritoneal creatinine clearance corrected to 1.73 m(2) body surface area (K(creat)); and phosphate (PDR) and sodium dialytic removal (SDR) in millimoles per session, corrected for glucose absorption, which provides an estimate of metabolic cost. Blood pressure was also regularly monitored. Initially, 25 patients were identified for inclusion in the study. There were 6 withdrawals in total: 2 at enrolment, 1 at day 75 (transplantation), 2 at day 30 (catheter dysfunction), and 1 for drainage alarms. All patients received the same duration of overnight APD, using the same total volume of dialysate, with the same 1.5% glucose, lactate-buffered dialysate (Balance: Fresenius Medical Care, Bad Homburg, Germany). Tolerance was good. Compared with APD-C, APD-A resulted in a significant enhancement of Kt/V(urea), K(creat), and PDR. The metabolic cost, in terms of glucose absorption, required to achieve dialytic capacity for urea, creatinine, and phosphate blood purification was significantly lower for APD-A than for APD-C, and UF increased during APD-A. With APD-A, each gram of glucose absorbed contributed to 18.25 ± 15.82 mL UF; in APD-C, each gram of glucose absorbed contributed to 15.79 ± 11.24 mL UF. However, that difference was not found to be significant (p=0.1218). The SDR was significantly higher with APD-A than with APD-C: 35.23 ± 52.00 mmol and 18.35 ± 48.68 mmol per session respectively (p<0.01). The mean blood pressure recorded at the end of each PD period (on day 45) was significantly lower when patients received APD-A than when they received APD-C. Our study provides evidence that, compared with the uniform dwell times and fill volumes used throughout an APD-C dialysis session, the varying dwell times and fill volumes as described for an APD-A dialysis session result in improved dialysis efficiency in terms of UF, Kt/V(urea), K(creat), PDR, and SDR. Those results were achieved without incurring any extra financial costs and with a reduction in the metabolic cost (assessed using glucose absorption).