Optimization of bimodal automated peritoneal dialysis prescription using the three-pore model.

Abstract

Previous studies suggested that automated peritoneal dialysis (APD) could be improved in terms of shorter treatment times and lower glucose absorption using bimodal treatment regimens, having 'ultrafiltration (UF) cycles' using a high glucose concentration and 'clearance cycles' using low or no glucose. The purpose of this study is to explore such regimes further using mathematical optimization techniques based on the three-pore model. A linear model with constraints is applied to find the shortest possible treatment time given a set of clinical treatment goals. For bimodal regimes, an exact analytical solution often exists which is herein used to construct optimal regimes giving the same Kt/V urea and/or weekly creatinine clearance and UF as a 6 × 2 L 1.36% glucose regime and an 'adapted' (2 × 1.5 L 1.36% + 3 × 3 L 1.36%) regime. Compared to the non-optimized (standard and adapted regimes), the optimized regimens demonstrated marked reductions (>40%) in glucose absorption while having an identical weekly creatinine clearance (35 L) and UF (0.5 L). Larger fill volumes of 1200 mL/m2 (UF cycles) and 1400 mL/m2 (clearance cycles) can be used to shorten the total treatment time. These theoretical results imply substantial improvements in glucose absorption using optimized APD regimens while achieving similar water and solute removal as non-optimized APD regimens. While the current results are based on a well-established theoretical model for peritoneal dialysis, experimental and clinical studies need to be performed to validate the current findings.