The variability in ultrafiltration achieved with icodextrin, possibly explained

A recent study by Jeloka et al. (Perit Dial Int 2006; 26:336-40) highlighted the high variability in maximum ultrafiltered volume (UF(max)) and the corresponding dwell time (t(max)) obtained using 7.5% icodextrin solution. We aimed to pinpoint the possible sources of this phenomenon by simulating the icodextrin ultrafiltration (UF) profiles according to the three-pore model of peritoneal transport.

The individual UF time courses observed in the study by Jeloka et al. (n = 29) were first characterized by linear and quadratic regression. We were then able to identify four main patterns. These were then adapted to UF profiles generated by the three-pore model by systematically altering the values of some model parameters, namely, the mass transfer area coefficient (MTAC or PS) for icodextrin/glucose, the peritoneal UF coefficient (LpS), the plasma colloid osmotic pressure gradient (DeltaPi), and the macromolecular clearance out of the peritoneal cavity (Cl(LF)).

Modifications in the PS values caused only marginal variations in UF(max) and t(max), while more significant changes were produced by altering LpS and Cl(LF). However, far more evident was the importance of changes in DeltaPi. In fact, lowering DeltaPi to 14 mmHg caused a steady increase in UF with 10 - 14 hour dwells. On the contrary, the UF profiles became nearly "flat" when DeltaPi was increased to 30 mmHg. The parallel shifts induced by altering icodextrin metabolite concentrations did not markedly influence UF(max) or t(max).

The UF pattern in icodextrin dwells seem to be mainly determined by the plasma colloid osmotic pressure, while only moderate changes can be seen with alterations in LpS and Cl(LF). The result is not completely unexpected considering that icodextrin acts by inducing a strong colloid osmotic gradient. A number of clinical studies would be needed, however, in order to prove this hypothesis.