| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||
COMMENTARIES |
Division of Nephrology, St. Joseph's Hospital Department of Medicine, McMaster University Hamilton, Ontario, Canada
e-mail: margetts{at}mcmaster.ca
An original article in this issue of Peritoneal Dialysis International describes a rodent model of peritoneal dialysate exposure in which the effect of chronic heparin administration is evaluated (1). Schilte and colleagues use a daily infusion of 3.86% glucose conventional peritoneal dialysis (PD) fluid through a tunneled catheter over 5 weeks. They identify a number of changes that developed in the peritoneal tissue in response to chronic exposure to peritoneal dialysate. Peritoneal fluid was supplemented with unfractionated heparin (UFH) or low molecular weight heparin (LMWH) and a control group was treated with dialysis fluid only. Of note, the addition of heparin did not appear to ameliorate the damage induced by peritoneal fluid exposure.
Heparin is a variable mixture of sulfated disaccharide polymers and is a member of the glycosaminoglycan family, which includes keratan sulfate, dermatan sulfate, chondroitin sulfate, heparan sulfate, and hyaluronic acid. Heparin binds to antithrombin and induces a conformational change that allows for the binding and inactivation of enzymes involved in the coagulation pathway, such as thrombin and factor Xa (2). Heparin was first discovered in 1916 and was purified for clinical use by the Connaught Laboratories, Toronto, which is famous for production of insulin (3).
LMWH is manufactured through the depolymerization of UFH. The longer polysaccharide chains in UFH are important for providing a scaffold to support the interaction between thrombin and antithrombin. This scaffold is not required for the interaction between factor Xa and antithrombin (3). Aside from greater bioavailability, LMWH has greater anti-factor Xa activity than antithrombin effect.
There are numerous reports of heparins having effects beyond anticoagulation. Anticoagulation has been widely used in patients with cancer, with some unexpected improvements in overall mortality rates reported in some studies (4). Heparin, through inhibition of thrombin and fibrin deposition, may inhibit the necessary environment for tumor cells and their associated blood vessels to develop (4). Heparin may have direct antiangiogenic properties through binding of endostatin (2). Heparin also blocks endothelial P-selectin and may thus impair metastatic seeding (5). There is some evidence that heparin, especially LMWH, may interfere with growth factors binding to their receptors, thus inhibiting cellular proliferation (4). Specifically, LMWH can inhibit fibroblast growth factor and vascular endothelial growth factor activity (5).
Heparin has been shown to have anti-inflammatory properties. Several mechanisms have been demonstrated, including inhibition of production of tumor necrosis factor-alpha (6), binding to P-selectin to inhibit leukocyte migration, and systemic release of tissue factor pathway inhibitor (7). Heparin inhibits the production of reactive oxygen species (8) and has a vasodilatory effect through increased production of nitric oxide (9).
All this provides ample rationale for the study of heparin in PD. Specifically, the peritoneum is subjected to inflammation, angiogenesis, and fibrosis during the course of PD therapy (10). These changes are associated with increased solute transport, which in turn is associated with ultrafiltration dysfunction (11) and increased mortality risk (12). Therefore, the anti-inflammatory and antiangiogenic properties of heparin are of possible value in preventing the complications of long-term peritoneal membrane injury. Furthermore, heparin is already used intermittently in PD patients during episodes of peritonitis and when fibrin is observed in the PD effluent.
The fibrin inhibitory effect of heparin may have benefits for long-term PD patients beyond the usual use in maintenance of catheter patency. It is known that the first steps in wound healing involve platelet aggregation and subsequent blood clotting. This fibrin-rich clot provides a matrix to support angiogenesis and migration of fibroblasts (13). Honda and Oda have demonstrated that fibrin deposition occurs at the surface of the fibrous adhesions seen in encapsulating peritoneal sclerosis (EPS) (14). From these observations it was concluded that fibrin deposition is a "key event in the pathogenesis of EPS" (14). Ongoing exudation of fibrinogen from these vessels and subsequent fibrin deposition without dilution and washing with PD fluid may explain the increased incidence of EPS after cessation of PD therapy.
There are few in vitro studies of heparin in PD. One worrisome observation was made by Manalaysay and colleagues: they found that heparin had inhibitory effects on mesothelial cell growth and protein production (15). This was confirmed by a second study (16).
As outlined by Schilte and colleagues (1), other animal studies have been carried out using chronic exposure to glucose-based dialysis fluids supplemented with heparin. Pawlaczyk and colleagues treated rats daily with dialysis fluid (17). They found that heparin increased ultrafiltration at an early time point in the experiment (day 10) but had minimal effect by day 30. Solute transport, measured by glucose absorption, was not different between groups.
Bazargani and colleagues carried out an interesting study with a single intraperitoneal (IP) infusion of dialysis fluid supplemented with LMWH (18). They used both direct IP injection and injection via an indwelling catheter. LMWH reduced inflammation and increased ultrafiltration in this experiment. The presence of the catheter induced complement activation, which was not inhibited by heparin.
De Vriese and colleagues compared the effects of IP heparin administration with a heparin-coated catheter and found that the heparin-coated catheter optimally preserved catheter function (19). This observation was supported by another study by Zareie and colleagues (20). Opposed to these observations, Kim and colleagues studied the effects of heparinized PD catheters in a rat model of daily dialysis exposure (21). This study was essentially negative for an effect on biofilm formation, exit-site changes, and peritonitis rates.
Human studies are similarly few and with mixed results. The LMWH tinzaparin was administered IP and membrane transport was assessed in a double-blind crossover study of 21 PD patients (22). In that 3-month study, solute transport was significantly reduced with LMWH [16% reduction in dialysate-to-plasma ratio (D/P) of creatinine], with a significant increase in ultrafiltration. Only 11 patients completed the study. The authors did not observe any bleeding complications. Peritonitis rates were very high in this study, related possibly to the need to add heparin to the dialysate fluids. In a second publication from this same patient sample, Sjoland and colleagues observed a decrease in systemic and IP inflammatory markers in patients treated with LMWH (23).
Mizuiri and colleagues treated 11 stable chronic PD patients with 5000 U heparin in the long dwell over 60 days (24). They were interested mainly in the effect of heparin on the formation of advanced glycation end products (AGEs). Interestingly, they found an increase in AGEs in the peritoneal effluent with IP heparin treatment. This was associated with increased D/P urea and creatinine. This was not a randomized trial and the changes in peritoneal AGEs and solute transport may not have been related to administration of heparin. Ponce and colleagues also observed an increase in solute transport in a single-dwell study using UFH (25).
The rationale therefore for routine IP heparin use in PD patients is strong. In vitro, animal, and human studies all demonstrate mixed results. Some variability in the observed results may be related to dose and systemic bioavailability. Schilte and colleagues observed limited effects of IP heparin on parameters of systemic coagulation (1). This has also been observed in patients on continuous ambulatory PD being treated with IP heparin (26). De Vriese and colleagues observed local accumulation of heparin in the peritoneal tissues (19). The purported anti-inflammatory effects of heparin, such as binding to P-selectin, may require systemic distribution to be effective. The one positive clinical trial used tinzaparin, a LMWH that may have better systemic bioavailability when delivered IP (22).
What does the Schilte study add to our understanding of chronic IP heparin use in PD patients? Despite certain limitations, that careful study clearly demonstrates that UFH or LMWH administration does not ameliorate the peritoneal membrane damage induced by exposure to high-glucose bioincompatible PD solution. We study animal models for two main reasons: to answer basic biological questions and as a prelude to human clinical studies. The study by Schilte and colleagues was carefully carried out and the results analyzed appropriately. Does this study provide insight into the activity and mechanism of heparin in the peritoneum? Clearly, UFH and LMWH did not have a beneficial effect on the peritoneal membrane. With evidence for antithrombin, anti-inflammatory, and antiangiogenic activity of heparin, the remaining question is, Why was heparin not effective in this model? Is the nature of the inflammatory response in the rodent model with an indwelling catheter different to what would be observed in the peritoneum of PD patients (18)? Was the dose of heparin used too low? Is a greater systemic exposure required for an anti-inflammatory action? Would different results be found if the model were optimized [perhaps with heparinized catheters and antibiotics (19)] to limit catheter malfunction? Antithrombin levels in the rodent peritoneum were not assessed in this experiment. In the noninflamed human peritoneum, antithrombin levels are very low (27), which may limit the effect of heparin as an anti-fibrin agent. Finally, heparin was not used during the terminal dwell, so an acute effect of heparin on peritoneal solute transport was not assessed. Future carefully conducted animal studies may help answer some of these questions and provide further mechanistic insights into the role of IP heparin in PD patients.
The second issue is whether this animal study will influence future clinical investigation. Animal models of chronic PD fluid exposure have been refined over time and have yielded important insights into peritoneal membrane injury. Can these concepts be directly translated to PD patients? The one positive clinical study by Sjoland and colleagues was very small (11 patients completed the study) and of limited duration (3 months of exposure) (22). However, the findings were strikingly positive, with a decrease in D/P creatinine of about 0.1, which, according to a recent meta-analysis (12), could potentially translate into a 15% reduced mortality risk. Clearly, there are risks with heparin therapy. LMWH has been associated with increased bleeding risk in patients with end-stage kidney disease (28). Heparin binds platelet-activating factor 4. This complex is antigenic in some people and may cause heparin-induced thrombocytopenia, a rare and serious side effect of heparin administration. More recently, heparin contaminated with over-sulfated chondroitin sulfate made its way into clinical use, leading to several well publicized deaths (29). Finally, the Sjoland study demonstrated a very high risk of peritonitis, perhaps from repeatedly adding heparin into the dialysis fluid (22). With potential risks and benefits weighed, it would still be reasonable to repeat the Sjoland study to confirm or refute these benefits.
Despite the additional insight provided by the Schilte study, the potential long-term benefits of heparin, especially LMWH, as a dialysate additive in PD patients remains to be elucidated.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
