The natural course of peritoneal membrane biology during peritoneal dialysis

Transcription

1 Kidney International, Vol. 64, Supplement 88 (2003), pp. S43 S49 The natural course of peritoneal membrane biology during peritoneal dialysis JOHN D. WILLIAMS, KATHRINE J. CRAIG, CHRIS VON RUHLAND, NICHOLAS TOPLEY, and GERAINT T WILLIAMS, FOR THE BIOPSY REGISTRY STUDY GROUP Institute of Nephrology; Medical Microscopy Sciences; and Department of Pathology, University of Wales College of Medicine, Heath Park, Cardiff, UK There is accumulating evidence to indicate that during the early years of renal replacement therapy, peritoneal dialysis (PD) provides an equivalent, if not superior, mode of dialysis to hemodialysis (HD) for a substantial number of patients [1, 2]. The benefit of PD, however, appears to be limited to the first three or four years, and the majority of patients switch therapy to HD because of technique failure. The causes of this treatment failure are multifactorial and include recurrent episodes of peritonitis, loss of residual renal function, and loss of peritoneal function [3]. The most common change in peritoneal function is a loss of ultrafiltration capacity, although a reduction in solute clearance is not an infrequent occurrence [4 6]. The causes of such functional changes are poorly understood but there is increasing evidence that they are related to changes in the structure of the membrane which correlates in most patients with the longevity of dialysis [7 10], although other factors clearly are influential. Structural changes include the loss of mesothelial cells, an increase in the thickness of the submesothelial compact zone, and a plethora of vascular changes, which range from classical small vessel atherosclerosis to venular changes. The etiology of these structural changes, however, remains speculative and includes the uremic process itself, recurrent infections, and the continuous exposure of the membrane to bioincompatible dialysis fluids [11]. Dialysis fluid has long been recognized as bioincompatible with the homeostasis of the peritoneal cavity [12 16]. A low ph, hyperosmolar fluid containing lactate as a buffering agent, and glucose degradation products (GDP) as a byproduct of production will inevitably have long-term negative pathophysiologic consequences [17, 18, 19]. By definition however, the fluid cannot be phys- Key words: peritoneal dialysis, VEGF, homeostasis. C 2003 by the International Society of Nephrology iologic [16]. The challenge in the design of new fluids is to reduce the components with the greatest potential to cause long-term pathologic events while maintaining the efficacy of the fluid as a tool for dialysis. The article will examine current data from the biopsy registry and will revisit those conclusions drawn from the original publication. It will also take the opportunity to examine, in paired specimens, the comparative changes in visceral and parietal peritoneum. Finally, it will examine in greater detail the evidence for an increase in blood vessel number in specimens taken from PD patients. METHODS All samples were collected and fixed in accordance with a standardized protocol and shipped to a central laboratory for processing, staining, and analysis [7]. Normal samples were obtained from living kidney donors with no history of previous abdominal pathology. Uremic samples were from predialysis patients or from HD patients at catheter insertion (in no case had any of these patients undergone PD). Biopsies from PD patients were categorized into four groups: (1) Samples obtained at transplantation (considered to be random samples); (2) samples obtained during incidental surgery (hernia repair or cholecystectomy); (3) samples obtained at surgery related to PD problems (catheter malposition, catheter replacement); and (4) samples obtained from patients with membrane failure. Mesothelial cell analysis was carried out using images from scanning and transmission electron micrographs (SEM and TEM). Submesothelial compact zone thickness and percentage vasculopathy [7] were measured from semithin sections (0.35 lm) stained with Toluidine blue and Periodic Acid Thiocarbohydrazide Silver Proteinate Silver Enhancement (PATCH-SP-SE), respectively. Visceral and parietal peritoneal morphology were compared in paired samples from the same subject. A histopathologist blinded to the clinical status of the patient carried out the analysis. S-43

2 S-44 Williams et al: Natural course of peritoneal membrane biology during PD Fig. 1. (A) Normal parietal peritoneum. There is a surface layer of flattened, cohesive mesothelial cells bearing microvilli and with indistinct cytoplasm containing occasional small vesicles. Nuclei are relatively inconspicuous and nucleoli are not seen. There is a submesothelial compact zone of mature collagen fibers. Toluidine blue stain. (B) Parietal peritoneum from a patient treated only by hemodialysis. The mesothelial cells show reactive changes. While cohesive, they are enlarged and appear rounded or cuboidal. The cytoplasm is prominent and finely granular, and the nuclei are irregular with prominent nucleoli and marginated chromatin. There is thickening of the submesothelial compact zone. Toluidine blue stain. (C) Parietal peritoneum from a patient treated with peritoneal dialysis. The mesothelial cells show reactive changes similar to those in B, but they are beginning to separate, both from each other and from the underlying stroma. Toluidine blue stain. (D) Parietal peritoneum from a patient treated with peritoneal dialysis. There is thickening of the submesothelial compact zone by coarse collagen bundles and the overlying mesothelium shows degenerative changes superimposed on the reactive changes seen in B and C. The mesothelial cells are pale, rounded, and discohesive and some are shedding from the surface. Toluidine blue stain. RESULTS Peritoneal membrane morphology Normal individuals. Fifteen normal parietal specimens were examined. All had intact mesothelial cell surface covering. Mesothelial cells seen en face by SEM were of polygonal appearance with a dense covering of surface microvilli. In section, they were of a flattened appearance (Fig. 1A), and TEM revealed cell-cell contact maintained by tight junctions. The median thickness of the submesothelial compact zone was 40 lm (interquartile range [IQR], 30 to 70 lm). A detailed analysis of the vascular structures in the submesothelial compact zone failed to identify any vasculopathy changes in either arteries or venules.

3 Williams et al: Natural course of peritoneal membrane biology during PD S-45 Thickness of submesothelial compact zone, microns N = Donor Uremic HD Fig. 2. Thickness of the submesothelial compact zone (microns) in patients donating a kidney for transplantation, patients with uremia undergoing catheter insertion and patients with end-stage renal disease who have only had hemodialysis therapy. For each box plot, median values are represented by the line within the box. The box represents 50% of the values (the 25th and 75th centiles), with the bars representing the highest and lowest values, excluding outliers ( ) and extremes ( ). Pre-dialysis and hemodialysis patients. In 27 parietal samples recovered from this group of patients, surface mesothelium was absent in 18.1%. In those patients with intact mesothelium, the cells sometimes assumed a reactive state, with increased apical-basolateral diameter (Fig. 1B), changes in surface microvilli, and an increase in intracytoplasmic organelles. In other samples mesothelial cells appeared rounded and degenerate. Parietal specimens taken from these patients showed significant thickening of the submesothelial compact zone (median thickness 150 lm (90 to 230 lm) (Fig. 2). Examination of the submesothelial vasculature showed a characteristic subendothelial hyalinization in 29%, which appeared to be mostly in venules. Grading of this vasculopathy indicated that in the majority of cases the vasculopathy was of Grade 1 or 2 [7]. Occasional arterioles showed evidence of intimal atherosclerotic changes. Peritoneal dialysis patients. In samples obtained from patients on PD, surface mesothelial cells showed a change in morphology of a similar nature to that seen in uremic patients (Fig. 1C). In addition, a significant proportion of the cells appeared degenerate with loss of nuclear density and a paucity of intracytoplasmic organelles. In a small percentage of patients, cells became columnar in appearance, vacuolated, and discohesive (Fig. 1D). As previously described [7], parietal submesothelial compact zone thickness increased with time on PD. Analysis by origin of sample, however, demonstrated that samples obtained at random from patients undergoing renal transplantation had least thickening of their submesothelial compact zone at 180 lm (IQR, 99 to 255 lm; N = 59); patients biopsies at the time of incidental surgery had a median thickness of 280 lm (IQR, 190 to 360 lm; N = 19). Patients undergoing surgery for PD-related problems had a further and significant increase in compact zone thickness with a median of 420 lm (IQR, 240 to 750 lm; N = 57). The greatest increase in submesothelial compact zone thickening was found in that group of patients with membrane dysfunction 700 lm (IQR, 460 to 1100 lm; N = 27). The majority of this group consisted of patients with loss of ultrafiltration and solute clearance. Changes in blood vessel structure were morphologically similar in biopsies from PD patients to those changes identified in uremic patients, although the prevalence of changes increased on PD. The lowest proportion of vasculopathic changes in biopsies from PD patients were seen in the patients biopsied at transplantation. Almost 50% of this group of patients showed no evidence of vasculopathy. In the remainder most of the changes were of Grades 1 and 2 (Table 1). In the membrane dysfunction group, however, almost 70% of the patients had Grade 4 changes with obliteration of vessel lumina. Further analysis of vasculopathy and fibrosis were made in respect of patient age. There was no relationship between the age of the patient and the thickness of the submesothelial compact zone or with respect to the presence or grade of vasculopathy. Comparison of visceral and parietal peritoneum specimens Thirty-eight PD patients had samples taken from both visceral and parietal membranes. Compact zone thickness as well as degree of vasculopathy was measured in each sample. The median thickness of the visceral samples was 20 lm (IQR, 10 to 80), while the median thickness of the parietal samples was 505 lm (IQR, 180 to 850) (P = , Wilcoxon signed-rank test) (Fig. 3). The median thickness of normal parietal peritoneum was 50 lm (IQR, 25 to 135) for comparison. Blood vessel changes were graded 1 to 4 as previously defined [7]. Vasculopathy was generally more prevalent and of greater severity in parietal than in visceral samples (chi-square test, P = 0.002) (Table 2). DISCUSSION For the purpose of understanding the way in which changes in the peritoneal membrane impact on the homeostasis of the peritoneal cavity and on the dialytic function of the membrane, it is important to define those elements that make up the membrane. The interface between the peritoneal cavity and the membrane begins at the mesothelial monolayer and its basement membrane. The submesothelial compact zone bridges the space between the mesothelial cells and the underlying

4 S-46 Williams et al: Natural course of peritoneal membrane biology during PD Table 1. Grade of vasculopathy and origin of sample No vasculopathy Grade 1 Grade 2 Grade 3 Grade 4 Donor N = (100%) 0 (0%) 0 (0%) 0 (0%) 0 (0)% Catheter insertion N = (55%) 3 (15%) 3 (15%) 3 (15%) 0 (0%) Transplant N = (63.6%) 21 (17.8%) 16 (13.6%) 2 (1.7%) 4 (3.4%) Incidental surgery N = (46.4%) 3 (10.7%) 10 (35.7%) 0 (0%) 2 (7.1%) Catheter change N = (27.1%) 6 (10.2%) 9 (15.3%) 10 (16.9%) 18 (30.5%) Membrane failure N = 27 3 (11.1%) 0 (0%) 3 (11.1%) 3 (11.1%) 18 (66.7%) Grade 1 = Thickening of lumen < 7 microns. Grade 2 = Thickening of lumen > 7 microns, no luminal distortion. Grade 3 = Thickening with luminal distortion. Grade 4 = Obliteration of lumen. Thickness of submesothelial compact zone, microns N = Random Problems Fig. 3. Thickness of the submesothelial compact zone (microns) in patients whose samples were obtained in a random manner (during transplantation or incidental surgery), and those obtained when patients were having problems with peritoneal dialysis therapy (those undergoing catheter replacement or having surgery due to membrane dysfunction). For each box plot, median values are represented by the line within the box. The box represents 50% of the values (the 25th and 75th centiles), with the bars representing the highest and lowest values. Outliers and extreme values are not charted. Table 2. Grade of vasculopathy: paired visceral and parietal samples No Grade Grade Grade Grade Parietal vasculopathy Visceral No vasculopathy Grade Grade Grade Grade vascular plexus. Dialysis takes place between the mesothelial cell surface facing the peritoneal cavity and the inner surface of the peritoneal capillary endothelial cell. Thus, any change in the structure of this membrane, either quantitative or qualitative, may have a significant impact on dialysis. The relative importance of visceral as opposed to parietal peritoneal membrane in terms of dialysis function remains speculative. Such evidence as exists dictates that dialysis takes place mainly across the parietal peritoneal membrane [20, 21]. Most biopsies of human peritoneum are taken from the parietal membrane. Most animal studies (in particular vascular dynamics) are carried out on the visceral component. We present evidence from paired human biopsies which indicates that changes in the visceral membrane are significantly less pronounced than those from the corresponding parietal membrane (Fig. 4) [22]. The parietal membrane demonstrates a significant greater tendency to increase the thickness of its submesothelial compact zone and exhibits a significantly greater prevalence of vasculopathy than does the corresponding visceral membrane. In view of this, great caution should be taken in translating the findings of experimental studies on visceral peritoneum to clinical implications for human parietal peritoneum. Interestingly, and in contrast, encapsulating peritoneal sclerosis (EPS), when it occurs, is almost exclusively a feature of the visceral peritoneal membrane [10, 23], suggesting that a different pathogenetic mechanism may be involved. Numerous publications have highlighted the loss of mesothelial cells from the peritoneal surface as an almost universal feature in patients on continuous ambulatory peritoneal dialysis (CAPD). Most of these publications, however, are based on unquantified evidence and a small sample number [24 27]. The degree of care applied when taking biopsies is also unclear. We found that the loss of mesothelium correlated with the extent of vasculopathy and with an increased thickness of the submesothelial compact zone. In addition, the remaining mesothelium was, in most biopsies, of abnormal appearance, showing morphologic changes of activation and/or degeneration [27] (Fig. 1). Degenerate mesothelial cells were discohesive and more likely to shed into the peritoneal cavity. A recent study [28] has examined shed mesothelial cells isolated from PD effluent and compared them with others grown from omental biopsies. With time on dialysis there appeared to be a possible transdifferentiation of mesothelial cells from an epithelioid to a mesenchymal/fibroblastic phenotype. Accompanying this change was a decrease in cytokeratin expression and an increase in vimentin expression. According to earlier studies,

5 Williams et al: Natural course of peritoneal membrane biology during PD S-47 Thickness of partietal membrane, microns Thickness of visceral membrane, microns Fig. 4. Comparison of the thickness (in microns) of paired samples of visceral and parietal peritoneum taken simultaneously. however, an increased vimentin expression was frequently seen in cultured mesothelial cells when compared to in situ cells [29]. In addition, the switch between keratin and vimentin expression appears to be reversible according to the culture conditions and the presence of growth factors [30]. A switch in phenotype from epithelioid to fibroblastic appeared to confer on these cells a migratory phenotype, allowing them to move into the submesothelial stroma (compact zone), where they may contribute directly to the fibrotic process [28]. This proposed direction of cell movement is in contrast to the previous hypothesis that mesothelial cells are replaced from stromal cells that migrate in an upward direction [31, 32]. This has been challenged recently, however, by labeling studies suggesting that the replacement of mesothelial cells lost from the surface occurs by the proliferation of remaining surface cells [33]. We were unable to identify in vivo similar fibroblastic phenotypic changes in the mesothelial cells seen in over 60 biopsy specimens. The morphologic appearance of samples from the present study suggests that the most common phenotypic change seen in mesothelial cells from biopsies of patients on PD was an increase in intracellular organelles, or that the cells took on a degenerate appearance with loss of cytoplasmic organization. It is therefore difficult to explain the differences between these studies and raises the question of whether effluent cells are truly representative of the remaining surface mesothelial cells. The loss of cells from the surface of the peritoneal membrane is likely to be due to a change in their phenotype, which allows them to detach and be washed out in the PD effluent, as opposed to migration and invasion of the stroma. Numerous studies have identified changes to the compact zone that lies beneath the mesothelial monolayer [7, 10, 23 27, 33]. It is accepted that an increase in fibrosis occurs within this region and that it is broadly related to time on dialysis. The Biopsy Registry has identified that this increased thickness precedes PD therapy (i.e., it is present as the result of uremia in both predialysis and patients treated only with HD) [7]. In addition, the degree of fibrosis varies between individuals: in some cases, individuals exhibiting a 10-fold increase in thickness after eight years on dialysis, and others barely a two-fold change over the same period of time. We were able to demonstrate that those patients with a history of problems with PD (multiple infections, malfunctioning catheters, or declining membrane function) were most likely to have significant thickening of the submesothelial compact zone [7]. The change in membrane thickness is therefore likely to be multifactorial and the result of the variety of pathologic insults endured by the membrane during time on PD. The third, and arguably the most important component of the peritoneal membrane is the vascular bed. Anatomically, the bulk of this lies at the junction of the submesothelial compact zone and its underlying adipose tissue or serosa. Some vessels, however, lie within the compact zone. In the visceral membrane where there is little or no compact zone, and the vessels may be adjacent to the basement membrane of the mesothelial monolayer. Changes to the vessels were first carefully categorized in a small study by Honda et al [9]. They noted a subendothelial hyalinization in small vessels thought to be venules. At its most extreme this resulted in obliteration of vessels. In addition, those vessels with structural change showed an accumulation of advanced glycation end products (AGE). Finally, they demonstrated in a small number of patients a correlation between this vasculopathy and a decrease in ultrafiltration across the peritoneal membrane. We were able to confirm these findings in a larger study [7]. The subendothelial hyalinization appears to be a process that begins during the progression of uremia and is seen in its mild and moderate forms before PD begins. The imposition of PD then accelerates the process. But, as with thickening of the compact zone, the changes are not universal across the PD population and affect those individuals who experience recurrent problems with PD. Analysis of the data from the biopsy registry demonstrates a highly significant correlation between the vasculopathy and thickening of the compact zone. The nature of the correlation suggests that the vasculopathy may be a major factor in influencing fibrosis. It may be that a feature of the obliterative vasculopathy is ischemia, which is known from other studies to result in fibrosis [34]. An increase in vessel numbers has also been identified in human biopsy studies [35]. This change appears to correlate with an increase in fibrous tissue. Using nitric oxide synthase (NOS) activity as a surrogate marker of vessel number, in addition to direct counting, Combet

6 S-48 Williams et al: Natural course of peritoneal membrane biology during PD et al [36] have confirmed these findings, demonstrating a quantitative increase in NOS activity with time on PD. In the same study, the presence of AGE in the peritoneal membrane colocalized in blood vessels with vascular endothelial growth factor (VEGF), a growth factor associated with increased blood vessel permeability previously demonstrated to be locally produced by the peritoneal membrane of patients on PD [37]. Parallel studies in animal models of PD (with or without uremia) have demonstrated the ability of glucosebased dialysis solutions to induce new vessel formation in the visceral peritoneum. A role for transforming growth factor b1 (TGF-b1) and VEGF in the induction of fibrosis and neovascularization has been suggested by recent gene transfer studies, although whether these short-term studies can be used to explain the long-term changes of the peritoneal membrane in humans has yet to be established [38 41]. CONCLUSION The data emerging from the Biopsy Registry indicate that changes to the morphology of the peritoneal membrane begin during the period of uremia that precedes renal replacement therapy. These changes persist but do not appear to progress appreciably during hemodialysis. With the introduction of PD, however, there is, in some patients, a progressive worsening both of compact zone fibrosis and of vasculopathy. The acquisition of random samples from patients undergoing renal transplantation represents a true reflection of the changes during uncomplicated dialysis. These changes contrast sharply with the appearance of the membrane in those patients who develop problems with PD and/or develop membrane failure. Thus, not all patients on PD inevitably develop progressive membrane changes. The availability of paired visceral and parietal samples also gives insight into the lack of correlation between the two sites in terms of morphologic changes. This should lead to caution when interpreting membrane changes in animal models (most of which focus on the visceral peritoneum), as well as indicating that the development of EPS (a visceral membrane phenomenon) may not be a direct progression of simple fibrosis. ACKNOWLEDGMENT We are indebted to Baxter Healthcare Corporation for the provision of an educational grant to support this project. Reprint requests to John D. Williams, University of Wales College of Medicine, Heath Park, Cardiff CF14 4XN, UK. williamsjd4@cf.ac.uk REFERENCES 1. COLES GA, WILLIAMS JD: What is the place of peritoneal dialysis in the integrated treatment of renal failure? Kidney Int 54: , FENTON SS, SCHAUBEL DE, DESMEULES M,et al: Hemodialysis versus peritoneal dialysis: a comparison of adjusted mortality rates. Am J Kidney Dis 30: , VAN BIESEN W, DEQUIDT C, VIJT D, et al: Analysis of the reasons for transfers between haemodialysis and peritoneal dialysis and their effect on survivals. Adv Perit Dial 14:90 94, Adequacy of dialysis and nutrition in continuous peritoneal dialysis: Association with clinical outcomes. Canada-USA (CANUSA) Peritoneal Dialysis Study Group. J Am Soc Nephrol 7: , DAVIES SJ, PHILLIPS L, GRIFfiTHS AM, et al: What really happens to people on long-term peritoneal dialysis? Kidney Int 54: , DAVIES SJ, BRYAN J, PHILLIPS L, RUSSELL GI: Longitudinal changes in peritoneal kinetics: the effects of peritoneal dialysis and peritonitis. Nephrol Dial Transplant 11: , WILLIAMS JD, CRAIG KJ, TOPLEY N, et al: Morphologic changes in the peritoneal membrane of patients with renal disease. J Am Soc Nephrol 13: , HONDA K, NITTA K, HORITA S, et al: Accumulation of advanced glycation end products in the peritoneal vasculature of continuous ambulatory peritoneal dialysis patients with low ultra-filtration. Nephrol Dial Transplant 14: , HONDA K, NITTA K, HORITA H, et al: Morphological changes in the peritoneal vasculature of patients on CAPD with ultrafiltration failure. Nephron 72: , DI PAOLO N, SACCHI G, DE MIA M, et al: Morphology of the peritoneal membrane during continuous ambulatory peritoneal dialysis. Nephron 44: , TOPLEY N: Membrane longevity in peritoneal dialysis: impact of infection and bio-incompatible solutions. Adv Ren Replace Ther 5: , JORRES A, WILLIAMS JD, TOPLEY N: Peritoneal dialysis solution biocompatibility: inhibitory mechanisms and recent studies with bicarbonate-buffered solutions. Perit Dial Int 17:S42 46, TOPLEY N, ALOBAIDI HM, DAVIES M, et al: The effect of dialysate on peritoneal phagocyte oxidative metabolism. Kidney Int 34: , DUWE AK, VAS SI, WEATHERHEAD JW: Effects of the composition of peritoneal dialysis fluid on chemiluminescence, phagocytosis and bactericidal activity in vitro. Infect Immun 33: , TOPLEY N: Biocompatibility of peritoneal dialysis solutions and host defense. Adv Ren Replace Ther 3: , HOLMES CJ: Biocompatibility of peritoneal dialysis solutions. Perit Dial Int 13:88 94, WIESLANDER AP: Cytotoxicity of peritoneal dialysis fluid-is it related to glucose breakdown products. Nephrol Dial Transplant 11: , THORNALLEY PJ, LANGBORG A, MINHAS HS: Formation of glyoxal, methylglyoxal and 3-deoxyglucosone in the glycation of proteins by glucose. Biochem J 344 Pt 1: , NAKAYAMA M, KAWAGUCHI Y, YAMADA K, et al: Immunohistochemical detection of advanced glycosylation end-products in the peritoneum and its possible pathophysiological role in CAPD. Kidney Int 51: , FLESSNER MF: Small-solute transport across specific peritoneal tissue surfaces in the rat. J Am Soc Nephrol 7: , FLESSNER MF: Osmotic barrier of the parietal peritoneum. Am J Physiol 267:F , CRAIG KJ, TOPLEY N, WILLIAMS GT, WILLIAMS JD: Morphological changes in the peritoneal membrane of patients on peritoneal dialysis (PD) correlate with total glucose exposure and peritonitis rates. J Am Soc Nephrol 12:425A, GAROSI G, DI PAOLO N: Peritoneal sclerosis: One or two nosological entities. Semin Dial 15: , DOBBIE J, ZAKI M, WILSON L: Ultrastructural studies on the peritoneum with special reference to chronic ambulatory peritoneal dialysis. Scott Med J 26: DOBBIE JW, LLOYD JK, GALL CA: Categorization of ultrastructural changes in peritoneal mesothelium, stroma and blood vessels in uremia and CAPD patients. Adv Perit Dial 6:3 12, DOBBIE JW: Pathogenesis of peritoneal fibrosing syndromes (sclerosing peritonitis) in peritoneal dialysis. Perit Dial Int 12:14 27, DOBBIE JW: Peritoneal ultrastructure and changes with continuous ambulatory peritoneal dialysis. Perit Dial Int 13:S , 1993

7 Williams et al: Natural course of peritoneal membrane biology during PD S YANEZ-MO M, LARA-PEZZI E, SELGAS R, et al: Peritoneal dialysis and epithelial-to-mesenchymal transition of mesothelial cells. N Engl J Med 348: , LAROCCA PJ, RHEINWALD JG: Co expression of simple epithelial keratins and vimentin by human mesothelium and mesothelioma in vivo and in culture. Cancer Res 44: , CONNELL ND, RHEINWALD JG: Regulation of cytoskeleton in mesothelial cells: reversible loss of keratin and increase in vimentin during rapid growth in culture. Cell 34: , BOLEN JW, HAMMAR SP, MCNUTT MA: Reactive and neoplastic serosal tissue. A light microscopic, ultrastructural and immunochemical study. Am J Surg Pathol 10:34 47, BOLEN JW, HAMMAR SP, MCNUTT MA: Serosal tissue: Reactive tissue as a model for understanding mesotheliomas. Ultrastruct Pathol 11: , MUTSAERS SE, WHITAKER D, PAPADIMITRIOU JM: Mesothelial regeneration is not dependent on subserosal cells. J Pathol 190:86 92, GOTLOIB L, SHOSTAK A: The functional anatomy of the peritoneum as a dialysing membrane, in Peritoneal Dialysis, edited by Twardowski ZJ, Nolph KD, Khanna R, New York, Churchill Livingstone, 1990, pp MATEIJSEN MA, VAN DER WAL AC, HENDRIKS PM, et al: Vascular and interstitial changes in the peritoneum of CAPD patients with peritoneal sclerosis. Perit Dial Int 19: , COMBET S, FERRIER ML, VAN LANDSCHOOT M,et al: Chronic uremia induces permeability changes, increased nitric oxide synthase expression, and structural modifications in the peritoneum. J Am Soc Nephrol 12: , ZWEERS MM, DE WAART DR, SMIT W, et al: Growth factors VEGF and TGF-beta1 in peritoneal dialysis [see comments]. J Lab Clin Med 134: , KOLB M, BONNIAUD P, GALT T, et al: Differences in the fibrogenic response after transfer of active transforming growth factor-beta1 gene to lungs of fibrosis-prone and fibrosis-resistant mouse strains. Am J Respir Cell Mol Biol 27: , MARGETTS PJ, KOLB M, YU L, et al: Inflammatory cytokines, angiogenesis, and fibrosis in the rat peritoneum. Am J Pathol 160: , MARGETTS PJ, GYORFFY S, KOLB M, et al: Antiangiogenic and antifibrotic gene therapy in a chronic infusion model of peritoneal dialysis in rats. J Am Soc Nephrol 13: , MARGETTS PJ, KOLB M, GALT T, et al: Gene transfer of transforming growth factor-beta1 to the rat peritoneum: effects on membrane function. J Am Soc Nephrol 12: , 2001 BIOPSY REGISTRY STUDY GROUP Mr. A. Jurewicz, Mr. C. Darby, Miss R. Lord, Mr. R. Chavez, University of Wales (Cardiff, Wales) Dr. A. Davenport, Royal Free Hospital School of Medicine (London, UK) Dr. A. Slingeneyer, Montpellier Hospital, (Montpellier, France) Dr. R. Selgas, Dr. A. Aguilera, Hospital de la Princesa (Madrid, Spain) Dr. M. Kim, Dr. Y. Kurioka, Shirasagi Hospital, (Osaka-shi, Dr. I. Iwamato, Kinki University Hospital the Third, (Osaka, Dr. T. Ono, Kinki University Sakai Hospital, (Osaka, Dr. M. Kubota, Kiyukai Oji Hospital, (Tokyo, Dr. A. Yasui, Matsushita Memorial Hospital, (Osaka, Dr. H. Kawanishi, Akanekai Tsuchiya General Hospital, (Hiroshima-shi, Dr. M. Nakayama, Jikei University School of Medicine, (Tokyo, Dr. Y. Hiriata, Hiroshima Prefectural Hiroshima Hospital, (Hiroshima-shi, Dr. M. Miyazaki, Nagasaki University, (Nagasaki-shi, Dr. Y. Nakamura, National Hiroshima Hospital, (Hiroshima-shi, Dr. K. Nagai, Dr. N. Mori, Shizuoka General Hospital, (Shizuoka, Dr. T. Maeba, Saint Mariana University, (Kawasakishi, Dr. M. Gotoh, Dr. C. Tomita, Hitachi General Hospital, (Hitachi-shi, Dr. M. Nozawa, Kameda General Hospital, (Kamogawa-shi, Dr. T. Kitahara, Dr. H. Kawai, Maebashi Saiseikai Hospital, (Maebashi, Dr. K. Hiraoka, Hyogo Medical College,(Hyogo-ken, Dr. Y. Kawaguchi, Jikei University, (Tokyo, Kaoru Yamabe, Baxter Healthcare, (Tokyo,