Hypothesis: Gender and Encapsulating Peritoneal Sclerosis

Encapsulating peritoneal sclerosis (EPS) is a serious complication of peritoneal dialysis (PD) and is characterized by the onset of intermittent nausea, vomiting, and other symptoms of a small bowel ileus (1). Radiographic imaging of the abdomen in these patients reveals a thickened peritoneal membrane involving the visceral peritoneal surface to the parietal peritoneal surface, with encasement of small bowel that limits normal motility. Also referred to as abdominal “cocooning,” the thickened fibrotic membrane can create cystic fluid collections that have been pathognomonic for the diagnosis of EPS.

Reports suggest that many cases of EPS have shown improvement with tamoxifen therapy, an antiestrogen compound. As females have higher rates of other fibrotic diseases, such as scleroderma, this paper will more fully report on the possible role of gender in the development of EPS, stating a hypothesis of possible female predominance, a hypothesis that cannot be proven or countered until there is fuller reporting of gender in the EPS literature.

The etiology of EPS is unclear. Patients exposed to peritoneal dialysate have many peritoneal membrane changes, characterized by loss of normal mesothelial cell morphology, expansion of the submesothelial subcompact zone, and neovascularization of the peritoneal membrane (2). It is unclear if, in some patients, EPS represents a serious final progression of this underlying peritoneal reaction to dialysate or is due to a “second hit” insult to peritoneum already damaged by the dialysis process. The inflammatory/fibrotic peritoneal capsule that characterizes EPS is likely due in part to aberrant fibroblast activity, with abnormal fibroblast expression of growth factors such as transforming growth factor-beta (TGF-β) stimulating the production of collagen and other extracellular matrix components. The population of peritoneal fibroblasts may be heterogeneous, with some fibroblasts originally existing as peritoneal mesothelial cells that, upon exposure to bio-incompatible peritoneal dialysate, react by changing their morphology and function to become fibroblasts, a phenomenon termed epithelial-to-mesenchymal transition (3). How these changes of epithelial-to-mesenchymal transition relate to EPS is still unclear.

Encapsulating peritoneal sclerosis has been described worldwide and can occur in children and adults of both genders. Due to incomplete reporting of gender in the available literature, it is not possible to determine if EPS has an overall gender predominance. The most extensive experience with EPS has come from Japan, where patients often remain on PD for over a decade. The medical/social environment in Japan has resulted in low rates of renal transplantation and therefore patients must rely on long-term dialysis treatments. The incidence of EPS in patients on PD for over 15 years has been reported to be as high as 17% (4). Many descriptions of Japanese EPS cases fail to mention the gender of the patients. For example, Kawanishi et al. described their experience with 100 surgical procedures in 86 patients with EPS and gender was not reported (5). Another large Japanese report described a sampling of 300 PD centers across Japan. A questionnaire survey noted 256 cases of EPS from those centers but the gender of the patients was not mentioned (6). In Japanese reports where gender is mentioned, there would appear to be a male predominance to EPS. For example, a report in 2004 of 48 Japanese cases noted that 75% of the patients were male and that the overall prevalence for males was increased compared to females (7).

Outside Japan, however, it is intriguing that so many small case series describe EPS in females, and often young premenopausal females (8–11). Many, but not all, larger reports of EPS also suggest a possible female predominance. For example, Summers and colleagues reported on 16 patients with severe EPS: 11 of the 16 were female and 9 of those 11 were females less than 50 years of age (12). Similarly, Campbell et al. described 15 cases of suspected EPS. Thirteen of the 15 cases were in females and 10 of those 13 occurred in females less than 55 years of age (13). This group later expanded their reporting to 54 Australian patients and noted a persistent female predominance (1.7:1.0), with mean age of all patients at 48 years (14). Oules and colleagues described 55 cases of cocooning peritoneal sclerosis collected from the Registry of the European Dialysis and Transplant Association–European Renal Association (EDTA Registry) and stated the cases revealed a “predominance of elderly and female patients” (15). However, a fuller understanding of EPS incidence by gender is not possible in these reports due to lack of reporting overall on genders of the PD population from which these EPS cases were abstracted. Nevertheless, it is interesting to note the possible female predominance to EPS outside Japan.

To speculate that EPS may occur more frequently in females, especially in premenopausal females, is a relevant concern as there is a female predominance, as mentioned, in other fibrotic/sclerotic disease processes, such as systemic lupus erythematosus, rheumatoid arthritis, and scleroderma, and young females have been described as having more robust fibrotic scar formation after cutaneous injuries such as burns. Fibroblasts have been demonstrated to express estrogen receptors and, once stimulated, produce several growth factors such as TGF-β (16–18). As mentioned, overexpression of TGF-β in the peritoneum results in peritoneal membrane changes notable for expansion and thickening of the submesothelial tissue. It is intriguing to speculate that estrogen stimulation of these peritoneal fibroblasts may lead to overexpression of TGF-β and allow estrogen to be permissive in the development of peritoneal fibrosis syndromes such as EPS.

Disruption of the peritoneal membrane by surgery or other insults often leads to peritoneal adhesion formation. In animal models, adhesion formation has been found to be influenced by sex hormone therapy, with estradiol compounds increasing adhesion formation and estrogen antagonism with GnRH analogs reducing adhesion formation (19–21). The adhesion tissue in this model is noted to be very vascular and stains heavily for vascular endothelial growth factor, suggesting that peritoneal production of vascular endothelial growth factor may, in part, be under estrogen regulation. Specimens with EPS have been noted to demonstrate angiogenesis and the degree of neovascularization predicts recurrence of EPS after decortication surgery (4).

Various treatments for EPS have been described, including medical and surgical options. Medically, success has been described with a variety of anti-inflammatory and antifibrotic agents such as glucocorticoids, mycophenolate mofetil, methotrexate, and colchicine (4,11). An increasing amount of literature has described treatment of EPS with tamoxifen (22). Tamoxifen is a selective estrogen receptor modulator (SERM), and therefore a nonsteroidal estrogen antagonist, that has been used in the treatment of other fibrotic reactions such as retroperitoneal fibrosis. Tamoxifen therapy decreases fibroblast proliferation and reduces fibroblast TGF-β expression (23,24). Success with this agent as an estrogen antagonist again suggests that estrogen itself may somehow be permissive in the development of EPS. The role of estrogen in EPS could even be contributory in males. Obese males have been demonstrated to exhibit increased serum levels and production of estrogens, putatively due to reduced peripheral estrogen receptors, reduced estrogen–testosterone-binding globulins, and a resultant increase in estrogen production rates. Other explanations for estrogen excess in obese males include enhanced peripheral conversion of testosterone to estradiol in adipocytes, as adipose tissue is the site of aromatization of androgens to estrogens (25,26). It is unclear whether the males described with EPS are more obese than their non-EPS counterparts.

At the present level of understanding, no firm conclusions can be made about the contribution of gender to the epidemiology of EPS. Japanese reporting seems to suggest a male predominance; other centers in Australia and Europe may show a female predominance. A better understanding of any gender predisposition to EPS or whether a higher estrogen state in premenopausal females or obese males may contribute to peritoneal fibrosis needs further clarification. All future reporting of EPS should be careful to document patient age, gender, and, if male, whether obesity was a notable factor. A fuller understanding of the role of gender in the development of EPS may be relevant to treatment options. For example, as tamoxifen therapy for EPS has been successful, further investigation is needed to determine if estrogen antagonism with tamoxifen may confer a particular benefit in female cases of EPS. Finally, as fibroblasts have been demonstrated to display estrogen receptors, further research is needed to understand whether estrogen stimulation of fibroblasts within damaged peritoneal tissue may contribute to the development of encapsulating peritoneal sclerosis.