Morphology of the peritoneal membrane during continuous ambulatory peritoneal dialysis

Peritoneal sclerosis: one or two nosological entities?

The frequency, pathology, animal models, pathogenesis, clinical manifestations, diagnostic criteria, therapy and prevention of peritoneal sclerosis are reviewed. Many of these aspects have a bimodal configuration which suggests that peritoneal sclerosis, usually considered a single pathology in peritoneal dialysis, is actually two distinct nosological entities: simple sclerosis and sclerosing peritonitis. The former is very frequent, with minor anatomical alterations and low clinical impact; it is reproducible in animals by means of peritoneal dialysis, and is clearly due to the poor biocompatibility of peritoneal dialysis solutions. The latter is rare, with radical anatomical alterations and high mortality requiring valid methods of diagnosis, therapy and prevention; it can only be reproduced in animal models by means other than peritoneal dialysis and seems to be due to factors both related and unrelated to peritoneal dialysis.

Volume stress-induced peritoneal endothelin-1 release in continuous ambulatory peritoneal dialysis

In long-term peritoneal dialysis, functional deterioration of the peritoneal membrane is often associated with proliferative processes of the involved tissues leading to peritoneal fibrosis. In continuous ambulatory peritoneal dialysis (CAPD), failure to achieve target values for adequacy of dialysis is commonly corrected by increasing dwell volume; in case of ultrafiltration failure, osmolarity of the dialysate gets increased. In a prospective study, the impact of increasing dwell volume from 1500 ml to 2500 ml per dwell (volume trial) or changing the osmolarity of the dialysate from 1.36 to 3.86% glucose (hyperosmolarity trial) on the peritoneal endothelin-1 (ET-1) release was analyzed. ET-1 is known to exert significant proliferative activities on a variety of cell types leading to an accumulation of extracellular matrix. A highly significant difference in the cumulative peritoneal ET-1 synthesis was found between the low- and high-volume exchange, whereas differences in the hyperosmolarity setting were only moderate. Sixty minutes after initiating dialysis, the cumulative ET-1 synthesis was 2367 +/- 1023 fmol for the 1500 ml versus 6062 +/- 1419 fmol for the 2500 dwell (P < 0.0001) and 4572 +/- 969 fmol versus 6124 +/- 1473 fmol for the 1.36 and 3.86% glucose dwell (P < 0.05), respectively. In conclusion, increasing dwell volume leads to a strong activation of the peritoneal paracrine endothelin system. Because ET-1, apart from being a potent vasoactive peptide, contributes to fibrotic remodeling, this study indicates that volume stress-induced ET-1 release might contribute to structural alteration of the peritoneal membrane in long-term peritoneal dialysis.

Effect of pH and glucose on cultured human peritoneal mesothelial cells

OBJECTIVE

We investigated the effects of various pH and glucose concentrations on the growth of human peritoneal mesothelial cells and on coagulation and fibrinolytic factors.

MATERIALS AND METHODS

Cells were cultured at various pH values in Ham's F-12 medium containing 1.0% foetal calf serum and supplemented with D-glucose or D-mannitol at various concentrations. After 4-48 h, cell proliferation and 3H-thymidine incorporation were determined. Coagulation and fibrinolytic factors were measured after 48 h.

RESULTS

Glucose caused concentration-dependent inhibition of cell growth at all pH values, but the deleterious effect of low pH on cell proliferation was faster and stronger than that of high glucose. At a similar osmolality, mannitol caused less inhibition of cell proliferation than glucose. There was a glucose concentration-dependent increase of thrombin-antithrombin III complex production at all pH values. At pH 5.2, tissue-type plasminogen activator production was far lower than at higher pH values, and production of the plasminogen activator inhibitor showed a glucose concentration-dependent increase. At pH 6.5 or 7.3, however, the plasminogen activator inhibitor production decreased and tissue-type plasminogen activator production increased in a glucose concentration-dependent manner.

CONCLUSIONS

Low pH and/or high glucose culture medium had an inhibitory effect on peritoneal mesothelial cells, with the effect of high glucose being partially related to hyperosmolality. These cells may modulate peritoneal coagulant and fibrinolytic activity, with the balance between coagulation and fibrinolysis being disturbed by low pH and/or high glucose.

Increased production of hyaluronan by peritoneal cells and its significance in patients on CAPD

Hyaluronan (HA) is a polysaccharide that forms a critical component of extracellular matrices. HA is present in high concentrations in tissues undergoing remodeling and morphogenesis, and it appears to have an important role in the early stages of wound healing. Here, we studied the level of HA in the peritoneal dialysate effluent (PDE) from 116 stable continuous ambulatory peritoneal dialysis (CAPD) patients. Longitudinal studies over a period of 6 weeks were performed in seven of these patients who developed peritonitis. The median HA level in PDE from these patients was 154.6 microg/L (range, 29.7 to 820.2 microg/L). Dialysate level of HA increased with age of the patients, but no such correlation was shown between HA level in PDE and duration of CAPD treatment or previous episodes of peritonitis. Patients with high or average peritoneal membrane transport of small solutes had a higher HA level in the PDE than those with a low peritoneal membrane transport (P = 0.046). A significant correlation was observed between PDE level of HA and interleukin-1beta (IL-1beta) or IL-6. The plasma level of HA in these patients was significantly greater than that of healthy controls (P < 0.0001), yet the plasma concentration of HA was only 85% that of the PDE concentration. In CAPD patients with peritonitis, there was a sharp increase in the PDE levels of HA, IL-1beta, and IL-6. These values decreased progressively with resolution of peritonitis. The changes in the PDE levels of HA closely followed those of IL-1beta or IL-6. In vitro [3H]-glucosamine incorporation studies suggest that the main bulk of HA is derived from synthesis by the peritoneal mesothelial cells, whereas the amount synthesized by macrophages is trivial. We conclude that elevated levels of HA found in the PDE of stable CAPD patients originate from increased synthesis by peritoneal mesothelial cells. This event may serve as a marker of regeneration and remodeling of the peritoneal lining.

Glycosaminoglycans prevent the functional and morphological peritoneal derangement in an experimental model of peritoneal fibrosis

Chronic peritoneal dialysis results in fibrosis of the peritoneal membrane, which leads to progressive reduction in dialytic efficacy. It was recently shown that the intraperitoneal administration of glycosaminoglycans (GAGs) improves the efficiency of peritoneal dialysis in CAPD patients. To verify whether the favorable effects of GAGs are purely functional or involve a morphological amelioration of the peritoneal membrane structure, a study was carried out in an animal model of plasticizer-induced peritoneal fibrosis. Rats, in which chronic renal failure had been induced by subtotal nephrectomy, received either placebo, plasticizers (i.p.), or GAGs (s.c.), or plasticizers (i.p.) and GAGs (s.c.). Urea dialysate-to-plasma equilibrium, urea and albumin peritoneal clearance, and glucose reabsorption were determined. The peritoneal membrane was evaluated morphometrically and histologically. In plasticizer-treated animals, peritoneal function tests and morphology were dramatically deranged. On the contrary, the subcutaneous administration of GAGs in plasticizer-treated rats maintained the peritoneal physiology and normal structure. The subcutaneous administration of GAGs protects peritoneal functions by affecting the remodeling of the peritoneum, rather than by a purely functional or simple mechanical effect.

Longitudinal follow-up of CA125 in peritoneal effluent

Mesothelial changes occur during peritoneal dialysis. CA125 provides a way to study the mesothelial cells in the in vivo situation. In the present study longitudinal changes of CA125 were analyzed. In addition, the appearance of CA125 in peritoneal effluent and day-to-day variability were studied. CA125 was measured in the effluent of five stable CAPD patients during four hour dwells with 1.36% glucose, with 3.86% glucose and with 7.5% icodextrin. In addition, CA125 was determined on six consecutive days in four hour effluents of three patients and appearance rates (AR) were calculated. Longitudinal follow-up was performed in 31 patients in whom three to seven yearly observations had been made. Linear appearance of CA125 was present in all dwells. No difference was found between the appearance rates of CA125 with 3.86% glucose, compared to either 1.36% glucose or icodextrin. Mean day-to-day coefficient of variation was 6.4% for CA125 AR, but a wide variation existed in stable CA125 values among patients (mean 22.1, range 2 to 48 U/ml). A negative trend with duration of CAPD was present in the longitudinal study. A mean decrease of 2.2% per year could be calculated, but substantial interindividual differences existed. Sudden decreases of CA125 AR were found in five patients. Possible causes were found in all of them and included a severe or recurrent peritonitis, and temporary cessation of peritoneal dialysis. In one patient a sudden decrease preceded the manifestation of peritoneal sclerosis. It can be concluded that CA125 can be used for the in vivo follow-up of the mesothelium in peritoneal dialysis patients. The appearance of CA125 in effluent is linear in time and not influenced by the initial lysis of mesothelial cells. A gradual loss of mesothelial cells is likely to occur, although interindividual variability is substantial. An acceleration of the process may be caused by severe peritonitis and perhaps by temporary cessation of peritoneal dialysis. A sudden decrease in CA125 may be an alarming sign for the development or manifestation of peritoneal sclerosis.

Human peritoneal fibroblast proliferation in 3-dimensional culture: modulation by cytokines, growth factors and peritoneal dialysis effluent

Structural and functional alterations of the peritoneal membrane are a significant problem in long-term peritoneal dialysis patients. The present study has established a 3-dimensional (3D) cell culture system to study the human peritoneal fibroblast (HPFB) and to examine its proliferative responses to cytokines and growth factors as well as dialysis effluent obtained from patients during peritoneal infection. PDGF-AB, basic FGF and IL-1 beta induced a time and dose dependent increase in 3D-HPFB proliferation. At day 9 proliferation, as assessed by MTT uptake, was increased by 2.4-, 2.3- and 1.5-fold above control by PDGF-AB (50 ng/ml), bFGF (50 ng/ml) and IL-1 beta (10 ng/ml), respectively (N = 5, P = 0.04 for all). These effects could be inhibited by co-incubation with anti-PDGF-AB antibody, anti-bFGF or IL-1ra, respectively. Exposure of 3D-HPFB to TGF-beta 1 did not result in an increase in cell proliferation. Incubation of 3D-HPFB with peritoneal macrophage (PMø) or human peritoneal mesothelial cell (HPMC) conditioned medium also resulted in a time and dose dependent increase in proliferation. At day 9, proliferation was maximally increase 1.65- and 1.92-fold by peritoneal macrophage- and mesothelial cell-conditioned medium, respectively. Cell free PDE, obtained from CAPD patients during episodes of peritonitis, induced 3D-HPFB proliferation above control values (2- to 6.5-fold increases, N = 5, P < 0.05 for all). This mitogenic potential of PDE was reduced following dilution, and with time following peritonitis there was a gradual decrease in the mitogenic effect of PDE. The proliferative potential of PDE was significantly reduced following co-incubation with IL-1ra (45.7% inhibition), anti-bFGF (34.9% inhibition) and anti PDGF-AB (27.4% inhibition). These data indicate that infected PDE causes fibroblast hyperplasia which might potentially contribute to pro-fibrotic processes during CAPD.

Biocompatibility of bicarbonate buffered peritoneal dialysis fluids: influence on mesothelial cell and neutrophil function

The present study compares the effects of lactate and bicarbonate buffered PDF on human neutrophil (PMN) and human peritoneal mesothelial cell (HPMC) viability and function. Acute exposure of PMN to lactate buffered PDF at pH 5.5 (CAPD 2, 1.5% and CAPD 3, 4.25% glucose) resulted in significant reductions in cellular ATP levels, the phagocytosis of serum treated zymosan (STZ) and respiratory burst activation (CL). Exposure of PMN to bicarbonate buffered PDF (BIC 20, 1.5% glucose and BIC 30, 4.25% glucose both at pH 7.2) had no significant effect on cell viability or the CL response. Phagocytosis was, however, depressed significantly more following exposure to BIC 30 than BIC 20. PMN cellular ATP levels and phagocytosis were significantly better in cells exposed to BIC 30 than to CAPD 3 at pH 7.4 (P = 0.043 for both). Pre-exposure of HPMC to CAPD 2, CAPD 3 or BIC 30 for 30 minutes resulted in a significant reduction in cellular ATP content compared to control medium. Pre-exposure to BIC 20 did not result in a reduction in HPMC ATP levels. HPMC synthesis of IL-6 was unaffected by 15 or 30 minutes pre-exposure to BIC 20 or BIC 30, in contrast pre-exposure to CAPD 2 or CAPD 3 for 15 or 30 minutes resulted in a significant reduction in stimulated IL-6 synthesis (24.5 +/- 3.01 and 32.3 +/- 5.0 vs. 43.9 +/- 10 pg/microgram cell protein in M199, N = 6; P = 0.02). Neutralization of the pH of CAPD 2 and CAPD 3 resulted in normalization of HPMC IL-6 secretion. Analysis of IL-6 mRNA expression in control, BIC 20 and 30 pre-treated HPMC subsequently stimulated with IL-1 beta revealed no differences in the expression of the IL-6 specific 465 base pair transcripts. The improved cellular function in bicarbonate buffered PDF indicates potentially improved host defence status and preservation of the peritoneal membrane in CAPD patients.

Longitudinal evaluation of peritoneal macrophage function and activation during CAPD: maturity, cytokine synthesis and arachidonic acid metabolism

The release of cytokines and prostaglandins (PG) by peritoneal macrophages (PM luminal diameter of) may influence the cytokine network controlling peritoneal inflammation and in the long-term the function of the peritoneum as a dialysis membrane. In the present study, an evaluation of the long-term effects of peritoneal dialysis on the release of cytokines and prostaglandins, and the expression of surface markers of cellular maturation on blood and mononuclear cells has been performed in patients during their first year on CAPD. Spontaneous release of tumour necrosis factor alpha (TNF alpha) and interleukins 6 (IL-6) by PM luminal diameter of, after 4 or 24 hours in culture, increased significantly with time on CAPD, while there was a small but significant decrease in release of prostaglandin E2 (PGE2). Production of TNF alpha and IL-6 was enhanced following incubation of the cells with lipopolysaccharide (LPS), but the effect of LPS was proportionally greater on blood monocytes than on PM luminal diameter of. There was a significant increase in the concentrations of PGE2 and 6-keto-prostaglandin F1 alpha in overnight dwell peritoneal dialysis effluent with time on CAPD. The levels of TNF alpha and IL-6 in uninfected PDE were below the detection limit of the immunoassay over the whole time period studied. Expression of CD15, which correlates with immaturity, by PM luminal diameter of and blood monocytes increased with time on CAPD, while expression of CD11c, a marker of maturation, decreased on blood monocytes, but did not change significantly on PM luminal diameter of. There was also a slight increase in expression of transferrin receptor in both PM luminal diameter of and monocytes, but this did not reach statistical significance. These findings suggest that peritoneal macrophages and blood monocytes isolated from CAPD patients over a one year period become increasingly immature with time, and this is accompanied by a significant modulation of their ability to secrete inflammatory cytokines. Dysregulation of macrophage function may have important consequences with respect to inflammatory processes and the long-term function of the peritoneal membrane in CAPD patients.

Effect of lactate-buffered peritoneal dialysis fluids on human peritoneal mesothelial cell interleukin-6 and prostaglandin synthesis

The present study focused on the evaluation of constitutive and cytokine-stimulated human peritoneal mesothelial cell (HPMC) IL-6 and 6-keto-PGF1 alpha release following pre-exposure to peritoneal dialysis fluid (PDF). Exposure of HPMC to PDF pH 5.2 resulted in a time-dependent increase in cell cytotoxicity [as assessed by lactate dehydrogenase (LDH) release] and concomitant inhibition of constitutive and IL-1 beta stimulated IL-6 and 6-keto-PGF1 alpha synthesis. After 15 minutes of exposure to PDF constitutive and IL-1 beta stimulated IL-6 release were reduced by 32.0 +/- 9.7% and 76.0 +/- 7.4% (N = 6, P < 0.046 and P < 0.027, respectively). PCR amplification of reverse transcribed mRNA from HPMC pre-exposed to PDF pH 5.2 demonstrated suppression of IL-1 beta stimulated IL-6 and cyclooxygenase (Cox-1 and Cox-2) transcripts. In order to mimic the dialysis cycle in vivo, an in vitro dialysis system was established. HPMC were exposed first to control medium, PDF pH 5.2 or PDF 7.3 for 15 minutes and then sequentially to pooled spent peritoneal dialysis effluent for up to four hours. The cells were subsequently allowed to recover in control medium for 12 hours in the presence or absence of IL-1 beta or TNF-alpha (both at 1000 pg/ml). There was no evidence of significant cell toxicity as assessed by LDH release during either the 'in vitro dialysis' or 'recovery' phases. Under these conditions short term exposure to PDF pH 5.2 followed by 'in vitro dialysis' resulted in significant inhibition of cytokine stimulated IL-6 (69.6 +/- 18.2 vs. 96.7 +/- 27.9 pg/microgram, N = 13; P < 0.020 for IL-1 beta) and 6-keto-PGF1 alpha (197.5 +/- 89.2 vs. 289.6 +/- 114.5 pg/microgram, N = 13; P < 0.020 for IL-1 beta) and 6-keto-PGF1 alpha (197.5 +/- 89.2 vs. 289.6 +/- 114.5 pg/microgram, N = 13; P < 0.003) release when compared to cells incubated in control medium. Adjustment of the pH of PDF to 7.3 reversed its inhibitory effects. We conclude that short-term exposure to PDF pH 5.2 significantly inhibits HPMC cytokine and prostaglandin release, an effect which appears to be related to its initial pH. Repeated exposure to nonphysiological PDF might impair mesothelial cell function and thus modulate intraperitoneal inflammatory processes.

Mechanisms of polymorphonuclear leukocyte mediated peritoneal mesothelial cell injury

To determine the susceptibility of human peritoneal mesothelial cells to injury mediated by activated polymorphonuclear leukocytes (PMNs), we exposed cultured human peritoneal mesothelial cells to 1250, 2500, 3750, and 5000 PMNs/mm3 activated with 50 ng/ml phorbol myristate acetate (PMA) or with 10(-7) FMLP/cytochalasin B for one to five hours. PMN adhesion to mesothelial cells was determined with radiolabeled PMNs. Mesothelial cell injury was determined in five different cell lines by measuring ATP depletion and 51chromium release. In each mesothelial cell line, PMN adhesion was significantly (P < 0.001) increased when PMNs were activated; 64 +/- 1.0 to 92.5 +/- 7.0% of the activated PMNs were adherent to mesothelial cells compared to 6 +/- 1.8 to 27 +/- 2.4% of resting PMNs. Mesothelial cells responded to PMN mediated injury with a fall in ATP levels and 51chromium release that was significant (P < 0.05) by three to four hours. At five hours, ATP levels were markedly depressed to 5 to 41% of control values. Increasing concentrations of activated PMNs caused significantly (P < 0.05) greater mesothelial cell injury as determined by ATP depletion and 51chromium release. PMN adhesion, ATP depletion and 51chromium release were significantly (P < 0.01) prevented by an anti-CD18 monoclonal antibody that inhibits the CD11/CD18 adhesion molecule complex on PMNs. Similar injury and protection from injury was demonstrated when mesothelial cells were exposed to PMNs activated with FMLP/cytochalasin B.(ABSTRACT TRUNCATED AT 250 WORDS)

Human peritoneal mesothelial cell prostaglandin synthesis: induction of cyclooxygenase mRNA by peritoneal macrophage-derived cytokines

Increasing evidence suggests that the mesothelial cell contributes to the control of inflammation in both the normal and inflamed peritoneal cavity. The present study examines the regulation of prostaglandin production by human peritoneal mesothelial cells (HPMC) following stimulation with peritoneal macrophage-conditioned medium and the cytokines interleukin-1 beta (IL-1 beta) and tumor necrosis factor-alpha (TNF-alpha). IL-1 beta and TNF-alpha stimulated significant release of prostaglandin above background levels in a time and dose dependent manner. Stimulation of HPMC with IL-1 beta (500 pg/ml) or TNF-alpha (100 pg/ml) for 24 hours resulted in the release of 24.5 +/- 4.3 (N = 11) (z = 3.40, P < 0.001 vs. control) and 19.4 +/- 4.5 (N = 10; z = 3.29, P < 0.001 vs. control) pg 6-keto-PGF1 a/micrograms cellular protein, respectively. Pretreatment of HPMC with dexamethasone (10(-6) to 10(-9) M) inhibited both constitutive and cytokine stimulated prostaglandin synthesis in a dose dependent manner. Both PMø-CM and PMø-S.epiCM stimulated 6-keto-PGF1 alpha and PGE2 synthesis by HPMC in a time and dose dependent manner (PMø-S.epiCM >> PMø-CM). Co-incubation of HPMC with PMø-S.epiCM in the presence of anti-IL-1 beta and/or anti-TNF-alpha antibody, interleukin-1 receptor antagonist or soluble TNF receptor (TNF p75) significantly reduced the capacity of these supernatants to stimulate prostaglandin synthesis. Exposure of HPMC to cytokines or PMø-S.epiCM resulted in the time dependent increase in the levels of both Cox-1 and Cox-2 mRNA as assessed by RT/PCR analysis with the greatest increase being seen for Cox-2. These data demonstrate specific stimulation of eicosanoid metabolism in HPMC by peritoneal macrophage derived cytokines, indicating the possible importance of these mediators in the activation of intraperitoneal prostaglandin synthesis. HPMC prostaglandins might act as important pro/anti-inflammatory mediators contributing to a cytokine network in the peritoneal cavity during CAPD peritonitis.

Immunohistochemical studies of the peritoneal membrane and infiltrating cells in normal subjects and in patients on CAPD

We performed immunohistochemical studies on biopsies of the parietal peritoneal membrane of 33 subjects to investigate whether other cell populations, in addition to mononuclear cells free in the dialysate, might participate in the defense of the peritoneum against microbial invasion during CAPD. Leukocytes were found to concentrate in two areas: a submesothelial layer composed of elongated macrophages displaying activation and maturation markers, and perivascular, less mature macrophages closely associated with T cells and HLA-DR, ICAM-1 and VCAM-1 expressing endothelial cells. Normal mesothelial cells were found to express constitutively the transferrin receptor and the adhesion molecules ICAM-1 and VCAM-1 but not ELAM-1. There were no major differences between normal and uremic subjects, while peritoneal dialysis patients exhibited minor derangements of the submesothelial layer and slight up-regulation of the expression of HLA-DR on endothelial cells. Peritonitis was associated with increased submesothelial cellularity and, particularly, perivascular leukocyte infiltration accompanied by increased expression of HLA-DR and adhesion molecules. Besides mononuclear cells free in the dialysate, this study demonstrates the existence of two additional peritoneal membrane leukocyte populations: submesothelial macrophages, and perivascular macrophages and T cells. It also suggests the existence of a fourth population of intracavitary leukocytes adherent to mesothelial cells. Studies are now necessary to evaluate their exact role in the host defence against peritonitis during CAPD.

Risks of peritoneal membrane failure in children undergoing long-term peritoneal dialysis

Children undergoing long-term peritoneal dialysis are at risk for membrane injury, necessitating conversion to hemodialysis. We analyzed the incidence and risk factors for membrane failure (inadequate ultrafiltration with or without peritoneal adhesions and decreased peritoneal surface area) in 68 children maintained with peritoneal dialysis for more than 3 months at our institution. The overall incidence of membrane failure was 16.2% (11/68). Kaplan-Meier estimates of peritoneal membrane survival were 88% at 24 months, 72% at 36 months, 65% at 48 months, and 52% at 60 months. Logistic regression analysis demonstrated that the risk of membrane failure increased with the number of episodes of peritonitis (odds ratio 1.61). The rate of peritonitis was 1 per 7.02 patient months in children who developed membrane failure compared with 1 per 9.18 patient months in children without membrane failures but the rate of peritonitis was not predictive of membrane failure (P = 0.09). Multiple logistic regression analysis demonstrated that peritonitis caused by Pseudomonas aeruginosa or alpha streptococcal organisms were independent predictors of membrane failure. We conclude that peritoneal membrane survival declines substantially with time on peritoneal dialysis and that membrane failure is associated with peritonitis, particularly peritonitis caused by Pseudomonas aeruginosa and alpha streptococcal organisms. The mechanism(s) of membrane injury are unknown but may be related to the inflammatory response initiated during peritonitis.

Intra-peritoneal free elastase in CAPD peritonitis

Neutrophil (PMN) recruitment into the peritoneum during acute bacterial peritonitis is an important part of the host defense barrier in CAPD patients. However, the subsequent phagocytosis of bacteria may also lead to PMN degranulation and the release of lysosomal enzymes. We determined the concentration of neutrophil elastase, both in complex with its natural inhibitor alpha 1Pi (E alpha 1Pi), and in uncomplexed, free form, in infected and normal CAPD peritoneal fluid by ELISA. In addition elastase activity was estimated in a casein degradation assay. Infected fluid contained a median (range) of 1.4 nM (0 to 9.2) free elastase by ELISA and 1.2 nM (0 to 11.9) activity. There were strong correlations between the peritoneal leukocyte count and both immunoreactive elastase and activity (r = 0.816, P < 0.001, 0.687, P < 0.01, respectively). In contrast, normal fluid contained 0.0 nM (0 to 0.32) immunoreactive elastase (P < 0.01) and 0.0 nM (0 to 0.6) elastase activity (P < 0.001). E alpha 1Pi complexes were raised significantly during peritonitis at 6.2 nM (0 to 34.3) and were barely detectable in normal fluid 0.0 nM (0 to 0.17; P < 0.005). The study shows that small but significant quantities of uninhibited elastase can be detected in the peritoneal fluid of CAPD patients with acute bacterial peritonitis. This observation may have important implications for the pathogenesis of peritoneal membrane damage and the phlogistic response to infection.

Morphological response of the peritoneum and spleen to intraperitoneal biomaterials

The present study was performed to evaluate the morphological response of the peritoneum and spleen to biomaterials. Silicone elastomer, knitted dacron or rubber was implanted, respectively, into a rat's peritoneal cavity and the morphology of the peritoneum and spleen was studied at 4 hours and on the 1st, 4th, 7th and 21st day after surgery. The morphological changes were identical among groups with different implanted materials. After intraperitoneal implantation of biomaterials from 4 hours and on, an infiltration of inflammatory cells was found in the slackened edematous superficial part of the peritoneum. Also noted in the spleen were stasis, vessel dilatation and fibrin deposition. With the help of scanning electron microscopy, a marked denudation and separation of the mesothelial cells, with infiltration of inflammatory cells, were observed. Peripheral leucocytes significantly increased in number one day after intraperitoneal implantation. Three weeks after intraperitoneal implantation, the materials were completely encapsulated and the morphological aberration of the peritoneum and spleen disappeared. The findings reveal the consequence and the resolution of the host-biomaterial interaction, which could contribute to the explanation of various pathophysiological alterations, including the translocation of enteric bacteria and the development of infectious complications after intraperitoneal biomaterial implantation.

Peritoneal morphology in children treated by continuous ambulatory peritoneal dialysis

Fifty peritoneal biopsies (PB) from 35 patients with end-stage renal disease, treated by continuous ambulatory peritoneal dialysis (CAPD) and aged 2 months to 18 years, were examined by light microscopy (n = 50) and/or scanning electron microscopy. PB were performed during surgical procedures immediately before the start of, during, or after the cessation of CAPD treatment. PB from 15 children without renal disease undergoing laparatomy were examined similarly. Before the start of CAPD, a scarcity and shortening of the mesothelial microvilli was observed by scanning electron microscopy. During and after CAPD, variable alterations of mesothelium, interstitium and capillaries were found. The mesothelial layer was absent in all 5 PB obtained during episodes of active peritonitis. In patients treated by CAPD for longer than 6 months, mesothelial denudation was observed more frequently (6/11) than in children treated for shorter periods (1/7) (P < 0.08). Fibrosis of the peritoneal membrane was present in about 50% of patients during or after the cessation of CAPD without impairment of peritoneal function. No correlation was found between the presence of fibrosis and the frequency of peritonitis or the duration of CAPD treatment.

Effect of bile on peritoneal morphology in Escherichia coli peritonitis

To investigate whether bile within the abdominal cavity changes the morphologic expression of peritonitis, rats received intraperitoneal injections of saline, bile, Escherichia coli (3 x 10(8) colony-forming units), or E. coli and bile (n = 10 in each group). Specimens were taken from the parietal and visceral peritoneum for light microscopy and scanning electron microscopy, 10 h after induction of peritonitis. Bile alone increased the number of inflammatory cells, especially of the monocytic type. After E. coli alone the monocytic cell response was less pronounced, whereas the polymorphonuclear leukocyte response was comparatively large in the visceral peritoneum. Addition of bile in E. coli peritonitis caused a substantial increase in polymorphonuclear leukocytes, especially in the visceral peritoneum, and in the depth of the inflammatory cell reaction. At scanning electron microscopy mesothelial cells appeared normal after bile alone, had reduced numbers of microvilli after E. coli alone, and were denuded or separated, without microvilli, after E. coli and bile. It is concluded that the inflammatory effects of E. coli are greatly enhanced in the presence of bile.