Determining the residual volume in peritoneal dialysis using low molecular weight markers.

Abstract

Variation in residual volume between peritoneal dialysis dwells creates uncertainty in ultrafiltration determination, dialysis efficiency, and poses a risk of overfill if the residual volume is large. Measuring the dilution of a marker molecule during fluid fill offers a convenient approach, however, estimation accuracy depends on the choice of dilution marker. We here evaluate the feasibility of creatinine and urea as dilution markers compared to albumin-based residual volumes and three-pore model estimations. This clinical, retrospective analysis comprises 56 residual volume estimations from 20 individuals, based on the dilution of pre-fill dialysate creatinine, urea and albumin concentrations during the dialysis fluid fill phase. Outcomes were compared individually. Bias induced by ultrafiltration, marker molecule mass-transfer and influence of fluid glucose contents was quantified using the three-pore model. Linear regression established conversion factors enabling conversion between the various marker molecules. Creatinine-based calculations overestimated residual volumes by 115 mL (IQR 89-149) in 1.5% dwells and 252 mL (IQR 179-313) in 4.25% glucose dwells. In hypertonic dwells, ultrafiltration was 52 mL (IQR 38-66), while intraperitoneal creatinine mass increased by 67% during fluid fill, being the leading cause of overestimation. Albumin-based volumes conformed strongly with three-pore model estimates. Correction factors effectively enabled marker molecule interchangeability. Mass-transfer of low molecular weight marker molecules is associated with residual volume overestimation. However, by applying correction factors, creatinine and urea dilution can still provide reasonable estimates, particularly when the purpose is to exclude the presence of a very large residual volume.