Phosphate-buffered saline (PBS), an isotonic solution with
a physiologic pH can be considered an example of a biocompatible dialysis fluid.
This study compared the biocompatibility of PBS with that of Dianeal 3.86%
(Baxter Healthcare Corporation, Deerfield, IL, U.S.A.), using a model of peritoneal
dialysis in the rat.
In an acute experiment, after catheter implantation, rats were infused on day 1
with PBS, on day 5 with standard dialysis solution (Dianeal 3.86%),
and on day 7 again with PBS. When rats were injected with Dianeal 3.86%,
the inflammatory reaction was suppressed as compared with PBS. The cell count
was lower with Dianeal (–85%, p < 0.001), the neutrophil:macrophage
ratio in dialysate was 80% lower (p < 0.01), total protein concentration
in the Dianeal dialysate was 73% lower (p < 0.01), and the dialysate
nitrite level was 45% lower (p < 0.01).
In a chronic experiment, after catheter implantation, rats were dialyzed for
four weeks with PBS or with Dianeal 3.86%. At the end of the study, a 1-hour
peritoneal equilibration test (PET) was performed. As evaluated on a semiquantitative
scale, macroscopic changes in the peritoneum were more severe in rats exposed
to PBS than in those exposed to Dianeal 3.86% (8.6 ± 3.2 vs
5.2 ± 2.6, p < 0.05). The thickness of the visceral peritoneum
was comparable in both groups; but, in PBS-treated rats, the peritoneal interstitium
contained more inflammatory cells and more new vessels. During the 1-hour PET,
peritoneal permeability to water and solutes was comparable in the two groups.
Despite a more physiologic composition, PBS is a less biocompatible peritoneal
dialysis solutions than is standard, acidic, hypertonic dialysis solution.
Long-term exposure of the peritoneal membrane to dialysis fluid leads to loss
of ultrafiltration and inadequate dialysis owing to alterations in membrane
permeability to solutes and water. It has been assumed that these changes, which
intensify with time on dialysis (1,2), are due to the bioincompatibility of
dialysis fluids. Dobbie (3) suggested that any agent that causes intraperitoneal
irritation may cause sterile peritonitis (“serositis”). Dialysis fluids
are bioincompatible because of their non physiologic composition: low pH, high
lactate concentration, hyperosmolality, high glucose concentration, and the
presence of glucose degradation products. Thus, phosphate-buffered saline (PBS),
an isotonic solution with a physiologic pH, can be considered a biocompatible
dialysis fluid. However, we previously showed that, in rats exposed to PBS,
the intraperitoneal inflammatory reaction is enhanced as compared with rats
treated with PBS supplemented with glucose (4).
Wang et al (5) postulated that one should not use physiologic saline (a glucose-free,
isotonic fluid) as a control solution, because it increases peritoneal lymphatic
flow. Hekking et al (6) found that rats injected with saline showed a higher
total cell count and more macrophages and neutrophils than did non dialyzed
control animals. In addition, in our previous study (7), rats given repeated
intraperitoneal injections of saline showed increased peritoneal permeability
as compared with non infused control animals. We carried out the present study
to compare the biocompatibility of PBS with that of Dianeal 3.86% (Baxter
Healthcare Corporation, Deerfield, IL, U.S.A.), using a model of peritoneal
dialysis in the rat..
In both the acute and chronic experiments, all dialysis fluids were supplemented with antibiotics [gentamicin 5 mg/L (Polfa, Tarchomin, Poland) and cefuroxime, 50 mg/L (Eli Lilly, Warsaw, Poland)] and with heparin [2500 U/L (Polfa)].
Acute experiment
In 6 male Wistar rats, weighing between 300 g and 350 g, we implanted
a peritoneal catheter according to a previously described method (8,9). Immediately
after catheter implantation, the peritoneal cavity of each rat was rinsed with
20 mL of PBS (Sigma, St. Louis, MO, U.S.A.). Afterward, 20 mL
of the same solution was introduced into the peritoneal cavity and left to be
absorbed.
On the next day [dialysis I (DI)], all rats were injected with 20 mL
PBS. Dialysate samples (5 mL) were taken after a 4-hour dwell. Using Dianeal
3.86%, the same procedure was performed in all animals on day 5 [dialysis II
(DII)]. On day 7, the procedure was repeated again, using PBS [dialysis III
(DIII)]. On days 2, 3, 4, and 6, the rats were injected with 20 mL
PBS, and the fluid was left in the peritoneal cavity for complete absorption.
For all dialysate samples, cells were counted in a hemocytometer immediately
after drainage. Afterward, a cell suspension was cytospun for a differential
count. The neutrophil:macrophage ratio in the dialysate (expressed as percentage)
was calculated for every rat. In dialysate samples, the total protein was measured
using the Lowry colorimetric method (10). Griess reagent was used to measure
dialysate nitrites (as an index of nitric oxide synthesis), after reduction
of nitrates to nitrites with nitrate reductase (Boehringer Mannheim, Mannheim,
Germany) (11).
Chronic experiment
Peritoneal catheters were implanted in 12 male Wistar rats weighing between
300 g and 350 g. Immediately after catheter implantation, the peritoneal
cavities of the rats in group 1 (n = 6) were rinsed with 20 mL
PBS, and the peritoneal cavities of the rats in group 2 (n = 6) were
rinsed with 20 mL Dianeal 3.86%. Afterward, 20 mL of the respective
solution was introduced into the peritoneal cavity and allowed to absorb. Daily,
for the next four weeks, the rats in group 1 were injected with 20 mL
PBS, and those in group 2 were injected with 20 mL Dianeal 3.86%.
At the end of the study, a 1-hour peritoneal equilibration test (PET) was performed
in each animal. Briefly, under ether anesthesia, a sample of blood was taken
from the tail vein. Thereafter, the rat was infused with 30 mL Dianeal 3.86%.
Immediately after infusion, a 2-mL sample of dialysate was drained from the
peritoneal cavity. After 60 minutes, the remaining dialysate was drained.
Peritoneal transport was determined, as was the dialysate-to-plasma (D/P) ratios
for urea nitrogen, creatinine, and total protein, and the D/D0 ratio for glucose.
In dialysate and plasma samples, urea and creatinine concentrations were measured
using an enzymatic method (Kit numbers A-371 and A-291, respectively: ANALCO,
Warsaw, Poland); glucose was measured by a colorimetric method (Sigma); and
total protein was measured using the colorimetric method described by Lowry
(10).
At the end of the PET, the rats were killed by bleeding. The abdominal cavity
was opened, and macroscopic changes were estimated according to the semiquantitative
scale. Macroscopic changes of peritoneum were classified on a scale from 0 to
12. The scale used three components:
Tissue samples from the visceral peritoneum covering the liver were then fixed in 10% formaldehyde solution in PBS and prepared for examination under a light microscope. The samples were stained using the Van Gieson method for visualization of collagen. The thickness of the peritoneal membrane was measured using interactive computer graphics analysis with the microscope Eclipse E 400 (Nikon Corporation, Tokyo, Japan).
Statistical analysis
All results are presented as mean ± standard deviation. The Mann–Whitney
test was used when two groups of results were compared, and the ANOVA test was
used when more than two groups were compared. A p value less than 0.05
was consider significant.
In the acute experiment, all parameters (dialysate cell counts, neutrophil:macrophage
ratios, and dialysate total protein and nitrite concentrations) were comparable
for dialysis days DI and DIII. These exchanges were done with PBS.
When rats were exposed to Dianeal 3.86% (DII), the dialysate cell count
was lower as compared with PBS [(DI) p < 0.001, Figure 1]. During
exchanges with Dianeal 3.86%, the low neutrophil:macrophage ratio (p <
0.01, Figure 1) was attributable to a dramatic decrease in number of neutrophils
in the dialysate, coupled with a simultaneous increase in the number of macrophages
(Figure 2). Interestingly, only rats exposed to Dianeal 3.86% showed
eosinophilia in the dialysis fluid (Figure 2). In rats exposed to Dianeal 3.86%
(DII), suppression of intraperitoneal inflammation was reflected by a lower
nitrite dialysate level (p < 0.01) and a lower dialysate concentration
of total protein (p < 0.01) as compared with PBS [(DI) Figure 1].
In the chronic experiment, the 1-hour PET results for glucose, urea, creatinine,
and total protein were comparable in rats exposed for four weeks to PBS or to
Dianeal 3.86%. Drainage of the dialysate from the peritoneal cavity was
slow in 4 rats treated with PBS. When the peritoneal cavities were opened,
we found severe adhesions of fibrous tissue involving the catheter in 4 of 6 rats
from this group. As estimated on the semiquantitative scale, macroscopic changes
in the peritoneum were more severe in rats exposed to PBS than in rats exposed
to Dianeal 3.86% (8.6 ± 3.2 vs 5.2 ± 2.6, p <
0.05). The thickness of the visceral peritoneum was comparable in the two groups
(PBS: 43.2 ± 41.2 mm; Dianeal: 32.4 ± 21.1 mm).
However, in peritoneal biopsies from the rats treated with PBS, the peritoneal
interstitium showed a much heavier infiltration by inflammatory cells, and signs
of neovascularization (Figure 3).
|
figure 1 Results of the acute experiment. Rats were exposed to phosphate-buffered saline [PBS (DI)], Dianeal 3.86% (DII), and PBS (DIII) for 4 hours. The results are expressed as a percentage of the data obtained during DI. |
|
figure 2 The results of cell differentiation in dialysate
during the acute experiment. The rats were exposed to phosphate-buffered
saline [PBS (DI)], Dianeal 3.86% (DII), and PBS (DIII) for 4 hours. |
|
figure 3 Morphological changes found in rats exposed for one month to phosphate-buffered saline (A) or Dianeal 3.86% (B). |
In this study, we showed that the intraperitoneal inflammatory reaction was
more intense in rats exposed to PBS solution (which has a physiologic pH and
osmolality) than in rats exposed to Dianeal 3.86% (Figure 1). In the
chronic experiment, the two groups showed less significant differences in peritoneal
permeability; however, PBS-treated animals showed more pathologic changes in
macroscopic and microscopic evaluations of the peritoneum. In most rats from
the PBS group, a strong reaction of the peritoneum, secondary to repeated infusions
of PBS, caused malfunction of the peritoneal catheters. It therefore seems that,
when studied in vivo, PBS is less biocompatible than standard acidic hypertonic
dialysis fluid (despite the more physiologic composition of PBS).
Hekking et al (6) reported that the dialysate cell count and the number of dialysate
macrophages were both higher in rats exposed to Dianeal 3.86% than in animals
treated with saline; however, in the data presented by these authors, the changes
were statistically nonsignificant. Based on those data, which show a higher
number of macrophages in the Dianeal 3.86% group and a similar number of
neutrophils in both groups, one can speculate that the neutrophil: macrophage
ratio was higher in the saline group, suggesting a stronger inflammatory response
after intraperitoneal infusion of saline.
One could attribute the pro-inflammatory effect of iso-osmotic–with–plasma
saline to its low pH. However, in our experiments, we found that saline solution
with normal pH (PBS) stimulates intraperitoneal inflammation. Our data also
show that the results of biocompatibility studies done in vivo do not necessarily
reflect observations from in vitro experiments: in the latter, fluids with
a physiologic pH and osmolality better preserve the function of the peritoneal
mesothelial cells and leukocytes (12). Better preservation of the viability
and function of peritoneal leukocytes permits a bioincompatible reaction in vivo,
namely a stronger inflammatory reaction. In fact, in rats chronically exposed
to PBS, we observed more morphologic changes in the peritoneum than in the peritoneum
of animals treated with Dianeal 3.86%. The changes may be secondary to
the enhanced inflammatory reaction. Lack of a significant difference in the
intensity of peritoneal thickening in the studied groups may be due to the short
duration of the study.
In the present study, we found higher dialysate eosinophilia (Figure 2)
in rats treated with Dianeal 3.86%. Previously, we demonstrated eosinophilia
in rats exposed to Dianeal 1.36% after catheter implantation (13). It has
been suggested that, because this effect diminished with time, it was due to
mechanical irritation during implantation of the catheter. However, as observed
in the present study, the relatively weak dialysate eosinophilia in rats with
implanted catheters and dialyzed with PBS suggests that the eosinophilia may
be due to other, as yet undefined, components of Dianeal 3.86% (that is,
glucose degradation products).
We believe that the results of biocompatibility tests using in vitro techniques must be verified with experiments done under in vivo conditions. Better viability and enhanced production of cytokines or free radicals by cells exposed in vitro to dialysis fluids with a physiologic composition may be reflected in vivo by the severe intraperitoneal inflammation and secondary destruction of the peritoneum. Further in vitro and in vivo studies are necessary to establish a new definition or to correct the present definitions of biocompatibility in peritoneal dialysis.
Katarzyna Wieczorowska–Tobis, md phd, Department of Pathophysiology, University Medical School of Poznan, ul. Swiecickiego 6, Poznan 60-781 Poland.