Necrosis and apoptosis of polymorphonuclear cells exposed to peritoneal dialysis fluids in vitro.

Conventional peritoneal dialysis (PD) fluids are known to inhibit polymorphonuclear cells (PMN) phagocytosis, oxidative burst and enzyme release. However, the relative contributions of apoptosis and/or necrosis to this dysfunction have not been examined. We investigated the effects of osmolality, glucose concentration and heat-sterilization of PD fluids on necrosis and apoptosis of PMN. Polymorphonuclear cells were isolated from 8 healthy volunteers and exposed to different PD fluids for four hours. PMN were then double-stained with Hoechst 33342 and propidium iodide to study the proportion of viable, apoptotic and necrotic cells. Transmission electron microscopy (TEM) was performed to confirm the results obtained with flow cytometry. The fluids studied were conventionally heat-sterilized 1.5% Dianeal (1.5% D), conventionally heat-sterilized 4.25% Dianeal (4.25% D), 1.5% D in which the osmolality was increased to that of 4.25% D by adding mannitol (1.5% D + M), a filter-sterilized version of 4.25% D (4.25% D-F) and a 1.1% amino acid PD fluid (AA) (Nutrineal PD4). All PD fluids had their pH equilibrated (pH = 7.4) by the addition of sodium bicarbonate. Compared to PMN exposed to culture medium, a significantly higher proportion of necrosis was observed in PMN exposed to 1.5% D (P = 0.04). The 4.25% D induced greater necrosis than 1.5% D (P = 0.001), and the 4.25% D also induced significantly more necrosis (P = 0.002) compared to 4.25% D-F. These data suggest that the consequences of heat-sterilization, rather than high glucose concentration are responsible for the necrosis observed. Indeed, the proportion of necrotic PMN with 4.25% D-F was not significantly different from 1.5% D. The 1.5% D + M and AA induced significantly more apoptosis compared to 1.5% D (P = 0.006 and P < 0.05, respectively), suggesting that apoptosis can be induced by the high osmolality of PD fluids. However, 1.5% D +/- M also induced significantly more apoptosis (P = 0.007) compared to 4.25% D-F. This suggests that the apoptosis effect is specific for the osmolyte present in PD fluids, and that mannitol and amino acids induce more apoptosis than glucose. In summary, the different non-physiological components of conventional PD fluids evaluated in this study had a differential effect on PMN survival. Heat sterilization of high glucose-containing PD fluids was associated predominantly with necrosis of PMN, and high osmolality with apoptosis.