Brenner and Rector's The Kidney, 8th ed.

CHAPTER 59. Peritoneal Dialysis

Ajay Sharma   Peter G. Blake

  

 

Peritoneal Membrane Anatomy, 2007

  

 

Peritoneal Transport, 2008

  

 

Diffusion, 2009

  

 

Fluid Removal, 2009

  

 

Fluid Absorption, 2009

  

 

Peritoneal Equilibration Test, 2010

  

 

Changes in Transport with Time on Peritoneal Dialysis, 2010

  

 

The Peritoneal Catheter, 2011

  

 

Catheter Insertion, 2013

  

 

Catheter Complications, 2015

  

 

Peritoneal Dialysis Solutions, 2015

  

 

Osmotic Agent, 2016

  

 

pH and Glucose Degradation Products, 2016

  

 

Buffers, 2017

  

 

Calcium, 2017

  

 

Sodium, 2018

  

 

The Peritoneal Modalities, 2018

  

 

The Peritoneal Prescription, 2018

  

 

Clearances and Adequacy, 2018

  

 

Measurement of Clearance, 2019

  

 

Clearance Targets, 2020

  

 

Factors Affecting Clearance, 2020

  

 

Volume Status, 2021

  

 

Ultrafiltration Failure, 2022

  

 

Nutrition, Inflammation, Residual Renal Function, and Cardiovascular Disease, 2022

  

 

Nutrition, 2022

  

 

Inflammation, 2023

  

 

Residual Renal Function, 2023

  

 

Cardiovascular Disease, 2024

  

 

Peritonitis, 2024

  

 

Diagnosis of Peritonitis, 2024

  

 

Management of Peritonitis, 2025

  

 

Prevention of Peritonitis, 2027

  

 

Noninfectious Complications, 2027

  

 

Mechanical, 2027

  

 

Encapsulating Peritoneal Sclerosis, 2028

  

 

Metabolic Complications, 2028

  

 

Acute Peritoneal Dialysis, 2029

  

 

Cost-Effectiveness of Peritoneal Dialysis, 2029

  

 

Patient Outcomes with Peritoneal Dialysis, 2029

  

 

Economics of Peritoneal Dialysis, 2030

Peritoneal dialysis (PD) was first performed on humans by Ganter, in Germany, in the 1920s.[1] However, successful treatment of small numbers of patients with acute renal failure was not reported until the late 1940s, when Palmer developed the first permanent peritoneal catheter made of silicone rubber. [2] [3] [4] [5] Gradually, experience grew until the early 1960s, when PD was first used for chronic treatment of patients with end-stage renal disease (ESRD).[6] [7] Pioneers such as Boen, McDonald, and Tenckhoff developed cuffed catheters and automated cycling equipment. [6] [7] [8] By the mid 1970s, the use of intermittent PD to treat ESRD patients had become relatively common. The major breakthrough, however, in the evolution of PD into a modality that could be used to treat thousands of patients was the development of continuous ambulatory peritoneal dialysis (CAPD) by Moncrief and Popovich in 1977.[9] Subsequent modifications by Oreopoulos in Toronto, and by Nolph and colleagues in Missouri, facilitated the widespread use of CAPD. [10] [11] The relative simplicity of the technique, its low cost, and the facility with which it could be performed by the patient at home all contributed to its popularity. By the late 1980s, it had become the preferred initial modality for patients in many countries, including Canada, Australia and the United Kingdom. Worldwide, it was estimated that about 14% of chronic dialysis patients were being maintained on CAPD.[12] In the 1990s, absolute numbers of patients on PD continued to grow, but with the aging of the ESRD population and the more widespread availability of hemodialysis, the growth in PD leveled out.[13] In the first decade of the 21st century, it is estimated that more than 120,000 patients worldwide are being kept alive by PD, and this accounts for approximately 8% of the world's chronic dialysis population.[14]

The enormous success of PD over the past 30 years has been made possible by better understanding of the anatomy and physiology underlying the modality and by greater appreciation of the complications of the therapy. This has led to the development of better technology, new delivery systems and solutions, and better strategies for maintaining patients more successfully on PD.

This chapter begins by focusing on anatomy and physiology that underlie the therapy. It then reviews in some detail the present status of peritoneal access, PD solutions, contemporary PD prescription, and management of infectious and mechanical complications. It also focuses on the complex and still incompletely understood interrelationships between cardiovascular disease, inflammation, residual renal function, nutrition, volume status, and peritoneal transport. Finally, some information about comparative patient outcomes and economics internationally will be provided.

PERITONEAL MEMBRANE ANATOMY

The peritoneal membrane is the largest serosal surface in the body with an approximate area of 1 to 2 M2 in adults ( Fig. 59-1 ).[16] The parietal peritoneum covers the abdominal wall and diaphragm, and receives its blood supply from the abdominal wall vasculature. The visceral peritoneum represents 80% to 90% of the total peritoneal surface area and receives it blood supply from the portal circulation.

FIGURE 59-1  Peritoneal cavity.

 

Histologically, the peritoneal membrane comprises the mesothelium, the interstitium, and the peritoneal capillaries ( Fig. 59-2 ). [17] [18] [19] The mesothelium is a monolayer of cells bearing microvilli and is fixed to a continuous basement membrane; it produces a thin layer of phospholipids, which is a rich surfactant fluid that allows free movement of the visceral organs during respiration and peristalsis. The interstitium is an amorphous matrix-like substance containing connective tissue fibers, fibroblasts, adipocytes, capillaries, and lymphatics. It is made up of colloid- and water-rich phases. [18] [19] The colloid-rich phase contains glycosaminoglycans, which are covalently bound to proteins, forming proteoglycans such as chondroitin, dermatan, keratin, and heparin sulfate.[20] The peritoneal microvasculature comprises true capillaries and postcapillary venules. The peritoneal endothelium lies on a basement membrane of negatively charged glycocalyx.

FIGURE 59-2  Peritoneal membrane histology.  (With permission from Williams JD, Cardiff University.)



The lymphatic drainage of the peritoneal cavity is mainly through stomata, which are present in the diaphragmatic peritoneum that drains into lymphatic plexuses and eventually through the lymphatic collecting ducts into the venous circulation. [19] [21] Other lymphatic drainage occurs through subserosal lymphatic channels on the mesothelial surface. Drainage is promoted by periodic respiratory compression and release of lymphatic vessels.

PERITONEAL TRANSPORT

With regard to PD, the peritoneal membrane was initially thought of as a dialyzer with six resistances in series.[22] These comprised the fluid film within the peritoneal capillary lumen, the peritoneal endothelium, the peritoneal capillary basement membrane, the interstitium, the mesothelium, and the fluid film within the peritoneal cavity. It is now recognized, however, that the major site of resistance to peritoneal transport is provided by the peritoneal capillary. [17] [18] [20] The interstitium also plays a significant role, especially in pa-tients undergoing long-term PD.[17] The mesothelium is no longer considered to be a significant barrier to peritoneal transport.

Understanding of peritoneal transport has been greatly facilitated by the development of two concepts over the past two decades. These are the three-pore model for peritoneal capillary transport and the distributed model, which particularly helps explain differences in transport characteristics among PD patients. The three-pore model was derived from general vascular physiology and was successfully applied to the unique situation of PD by Rippe.[23] It states that solute and water transport across the peritoneal capillary occurs through three different sets of pores. The smallest of these are known as ultrapores and have diameters of 4 to 6 Angstroms. They correspond to aquaporin I channels and transport only water and not solute.[24] Small pores are abundant, have a diameter of 40 to 60 Angstroms, and correspond to clefts in the endothelium. They transport small solutes and water in proportion to their concentrations in serum. Large pores have a diameter of 100 to 200 Angstroms and also correspond to clefts in the endothelium. These pores are much fewer in number and are responsible for the transport of macromolecules such as albumin. In standard glucose-based PD, approximately half of peritoneal water transport is accounted for by small pores and the other half by ultrapores.[25] Because the proportion of water transported by ultrapores is solute free, the total ultrafiltrate in PD has a lower solute concentration than serum. This is the phenomenon of sieving, which is referred to later.[26]

The “distributed model” also emphasizes the importance of capillaries in peritoneal transport.[18] It states that the distribution of peritoneal capillaries within the submesothelial matrix of the membrane is a major determinant of differ-ences between patients in their peritoneal transport characteristics. The theory emphasizes that the contribution of a given capillary to peritoneal transport depends on its proximity to the mesothelium and that patients with a greater density of blood vessels or vascularity of the peritoneal membrane will transport solute more rapidly.[27] All of this emphasizes the point that peritoneal transport is determined, not so much by the peritoneal membrane surface area but rather by the “effective peritoneal surface area.” In other words, the vascularity of the membrane is more important than its area.

A further concept needs to be appreciated to understand peritoneal transport during dialysis. This is the issue of peritoneal blood flow which is notoriously difficult to measure, but is estimated to be approximately 50 to 100 mL/min.[28] The maximum clearance of small solutes that can be achieved by PD is thought to be less than 30 mL/min, and so the statement is often made that peritoneal transport or clearance of small solutes is not limited by peritoneal blood flow.[29] Although this is true, it should be pointed out that vasoactive agents and episodes of inflammation, both of which augment peritoneal blood flow, also increase peritoneal transport. [30] [31] The point, however, is that this increase in transport is not due to the increased blood flow per se, but rather to the consequent increase in effective peritoneal surface area. Although this distinction may appear pedantic, it is important for an understanding of peritoneal transport.

With an appreciation of these basic concepts of peritoneal transport, the events during a PD dwell will now be analyzed.

First, it should be remembered that the two cardinal requirements for dialysis are that solute be cleared and fluid be removed from the uremic patient. Three distinct peritoneal transport processes occur and determine the success of fluid removal and solute clearance.[32] These three processes are diffusion, convective ultrafiltration, and peritoneal fluid absorption. It needs to be appreciated that all three processes occur simultaneously, but for simplicity, they will be each be reviewed separately.

Diffusion

The key factor determining diffusion for a given solute is of course the concentration gradient.[32] In renal failure, we are looking at the plasma to dialysate concentration gradient for solutes such as urea or creatinine. The other major factor determining solute diffusion is the ability of the membrane to transport the solute concerned. This is expressed as the mass transfer area coefficient (MTAC). This coefficient is expressed in milliliters per minute and is equal to the theoretical clearance of the solute that would be achieved if the concentration gradient was always infinitely high. The MTAC for a given solute depends on the effective peritoneal surface area, which, in turn, is determined by the size and vascularity of the patient's peritoneal membrane. [32] [33] There may also be a smaller contribution determined by the thickness of the submesothelial matrix.[19] At the start of a PD dwell, the concentration gradient is maximal and solute removal is fastest. As the dwell proceeds and the gradient decreases, transport slows down. More frequent drainage and replenishment of the cavity with fresh solution tends to keep the gradient greater and maximizes peritoneal transport. The MTAC can also be modified in the clinical setting but only to a modest degree. Increased dwell volumes increase the surface area of peritoneal membrane available for diffusion and so raise the MTAC.[34] As previously mentioned, agents or insults that increase vascularity will also increase the effective peritoneal surface area and, therefore, the MTAC. [30] [31] It should be pointed out that the MTAC is analogous to the term “KOA” as used for dialyzers in hemodialysis. In general, smaller solutes have higher MTAC values than larger ones. Typical values for adults in PD are shown in Table 59-1 .


TABLE 59-1   -- Mass Transfer Area Coefficients for Various Solutes

Solute

Molecular Weight (Daltons)

MTAC (mL/min)

Urea

60

17

Creatinine

113

10

Uric Acid

158

10

Glucose

180

11

Inulin

5500

4

Data from Krediet RT, Zuyderhoudt MJ, Boeschoeten EW, Arisz L: Alterations in the transport of water and solutes during peritonitis in CAPD patients. Eur J Clin Invest 17:43–52, 1987; Heimburger O, Waniewski J, Werynski A, Lindholm B: A quantitative description of solute and fluid transport during peritoneal dialysis. Kidney Int 41:1320–1332, 1992; and Waniewski J, Heimburger O, Werynski A, Lindholm B: Aqueous solute concentrations and evaluation of mass transfer area coefficients in peritoneal dialysis. Nephrol Dial Transplant 7:50–56, 1992.

MTAC, mass transfer area coefficient.

 

 

 

 

Fluid Removal

Ultrafiltration, or fluid removal, on PD in achieved by osmotic or oncotic forces, as distinct from the hydrostatic pressure gradients that are applied in hemodialysis. In standard PD, the osmotic gradient for glucose is the key determinant of fluid removal.[32] It should be noted that any osmotic agent used in PD has a reflection coefficient.[35] This expresses the degree to which that agent is retained in the peritoneal cavity and so can continue to apply its osmotic or oncotic pressure across the membrane. The closer the value is to 1.0, the better the agent is retained and the more sustained is the ultrafiltration. Although glucose has long been the mainstay of PD, it actually has a very low reflection coefficient of about 0.02.[36] This is because of its low molecular weight, which allows it to diffuse easily from the peritoneal cavity into the peritoneal capillary blood. This remarkably low reflection coefficient highlights the theoretical nonsuitability of glucose as an osmotic agent for PD. In the same way that the MTAC expresses the ability of the membrane to transport solute, the peritoneal ultrafiltration coefficient (KUF) expresses the ability of the membrane to transport water. This coefficient is determined by the effective peritoneal surface area and also by the inherent hydraulic conductance of the membrane. [32] [35] This conductance may be influenced by factors such as the thickness of the submesothelial matrix and perhaps by the density of pores in the peritoneal capillaries of a given patient. The KUF is analogous to the same term used to characterize fluid removal capacity in hemodialysis membranes.

Of course, the standard Starling forces of hydrostatic and oncotic pressure also influence fluid removal during PD. [32] [34] [37] The hydrostatic pressure gradient is determined by the difference between peritoneal capillary pressure and intraperitoneal pressure. The latter is, in turn, influenced by the volume of dialysate in place relative to the size and compliance of the patient's peritoneal cavity. It is also influenced by the patient's posture in that intraperitoneal pressure, for a given dialysate volume, is lower when the patient is supine and higher when the patient is standing. [34] [37] Just as in other capillary beds, oncotic pressure is also a determinant of water movement. Peritoneal capillary albumin concentrations tend to hold fluid within the vasculature. In recent years, peritoneal solutions based on polyglucose agent, icodextrin, have become popular.[38] These solutions have approximately the same osmolarity as serum and so do not remove fluid by osmotic forces. However, the large icodextrin molecules have an oncotic effect analogous to that of albumin and remove fluid in this manner. Table 59-2 demonstrates the typical values for the osmotic and hydrostatic pressures determining fluid removal during PD.[37]


TABLE 59-2   -- Typical Pressure Gradients Across the Peritoneal Membrane in Peritoneal Dialysis

 

Peritoneal Capillary Pressure

Intraperitoneal Pressure (2.0L Solution)

Pressure Gradient Favoring Ultrafiltration

Hydrostatic pressure (mm Hg)

17

8

9

Colloid osmotic pressure (mm Hg)

21

<0.1

-21

Osmolality (mOsm/kg H2O)

305

347 (1.5% dextrose)

 

 

 

486 (4.25% dextrose)

 

Maximum crystalloid osmotic pressure gradient (mm Hg)

 

 

+24[*] (1.5% dextrose)

 

 

 

+105[*] (4.25% dextrose)

Net maximum pressure gradient (mm Hg)

 

 

+12 (1.5% dextrose)

 

 

 

+93 (4.25% dextrose)

Adapted from Krediet RT, Lindholm B, Rippe B: Pathophysiology of peritoneal membrane failure. Perit Dial Int 20(suppl 4):S22–S42, 2000.

*

Based on osmolality gradient × solute reflection co-efficient × Van't Hoff constant. i.e., 42 × 0.03 × 19.3 = 24 for 1.5% dextrose 181 × 0.03 × 19.3 = 105 for 4.25% dextrose

 

 

Fluid Absorption

During the course of a PD dwell, there is constant removal of fluid from the peritoneal cavity through the lymphatics back into the peritoneal circulation.[21] This peritoneal lymphatic flow can be demonstrated by infusing a macromolecule such as dextran into the peritoneal cavity and measuring both its disappearance from the cavity and its appearance in the systemic circulation.[39] Interestingly, the disappearance rates of peritoneal macromolecules are notably greater than their appearance rates in the circulation. This had led to significant controversy about the nature of peritoneal fluid absorption. [40] [41] It is now appreciated that much of the absorption occurs not directly into the lymphatics but rather into the tissues of the abdominal wall in a process that is driven by peritoneal hydrostatic pressure. It is presumed that much of this anterior abdominal wall fluid is subsequently removed by subserosal lymphatics and also by peritoneal capillaries.[42] It has been argued that this fluid absorption should be included in estimates of peritoneal lymphatic flow because ultimately much of the fluid concerned ends up in the lymphatics and back in the systemic circulation. Others have questioned this and have raised the possibility that such fluid absorption may not be an ongoing one-way process but rather that an equilibrium may be reached so that it may not be valid to count this fluid removal as part of lymphatic flow. The end result, however, is that peritoneal fluid absorption occurs at a rate that is somewhere in the range of 1 to 2 mL/min or 250 to 500 mL during a 4-hour peritoneal dwell. This fluid absorption is a “bulk flow” process, and so results in both water and solute being restored to the systemic circulation rather than being removed in the peritoneal effluent. It thus detracts from the ability of PD to clear solute and remove fluid.

It should be noted that overall solute removal depends on the combined contributions of diffusion and convection minus the effect of peritoneal fluid absorption.[32] The extent to which ultrafiltration contributes to solute removal depends not only on the net ultrafiltration but also on the sieving coefficient of the solute concerned. [32] [36] This is the fraction of the solute that passes through the membrane in association with transported water. Because of the sieving effect of the ultrapores in glucose-driven peritoneal ultrafiltration, the sieving coefficients for most small solutes, including urea and sodium, are in the range of 0.6, meaning that the concentration of these solutes in pure ultrafiltrate is about 60% of what it is in the serum. [25] [26] Overall, however, in can be said that convection is relatively more important in solute removal in PD as compared with hemodialysis, accounting typically for 10% to 20% of total solute removal. A more detailed mathematical analysis of solute and fluid removal in PD is provided by Heimburger and co-workers.[32]

Peritoneal Equilibration Test

The peritoneal equilibration test (PET), first described by Twardowski in 1987, is the standard test used in clinical practice to assess peritoneal transport, although alternatives have frequently been advocated. [43] [44] [45] The advantage of the PET is its simplicity in that it can be relatively easily performed and interpreted. Essentially, it measures over 4 hours the processes of equilibration between plasma and dialysate for urea, creatinine, and sodium as well as the diffusion out of the cavity of glucose (see Fig. 59-1 ). The technical aspects of performing a PET are reviewed in detail elsewhere, but key points are to ensure maximal emptying of the long dwell before the test, good mixing of dialysate in the abdomen, and careful labeling and measurement of samples.[46] The typical PET curves with separation of patients into the 4 transport categories are displayed in Figure 59-3 . As can be seen, the bands are relatively tight for urea and glucose, and so the degree of equilibaration between dialysate and plasma for creatinine after 4 hours (D/P Cr) is used to determine a given patient's transport status.[43] It is now appreciated that the PET is primarily a measure of the effective surface area or vascularity of the patient's membrane, and so D/P Cr values differ between patients and increase during peritonitis. [27] [30] [31]

FIGURE 59-3  A-D, Peritoneal equilibration curves for creatinine, urea, glucose, and sodium.  (Adapted from Twardowski ZJ, Nolph KD, Khanna R, et al: Peritoneal Equilibration Test. Perit Dial Bull 7:138–144, 1987.)

 

 

Initially, the PET results were thought to be useful in designing the PD prescription, with the idea being that high transporters would benefit from the short dwells of cycler PD, whereas low transporters would do better with the long-duration dwells of CAPD.[43] However, in practice, the type of PD done is more often driven by cost and lifestyle issue, and most patients can be managed on either cycler or CAPD. However, it has become apparent that high transport status is associated with worse patient outcomes, at least on CAPD. [46] [47] [48] The explanation for this is unclear.[49] It may simply be that high transporters have more difficulty with fluid removal due to more rapid dissipation of the glucose osmotic gradient.[50] Alternatively, it may be that high transport status is a marker of comorbidity or of systemic inflammation and that these factors, rather than the transport status per se account for the adverse outcomes. [51] [52] [53] Regardless, the PET is now as much a guide to patient prognosis and potential complications than a determinant of the type of PD used.

Changes in Transport with Time on Peritoneal Dialysis

It has long been recognized that there is a tendency for peritoneal transport to alter with time on PD. [54] [55] On average, there is a gradual increase in D/P Cr and in MTAC values over the years. This does not occur in every patient and is more marked in some than in others.[54] It is associated with a decrease in ultrafiltration capacity, which leads to ultrafiltration failure in the worst affected patients ( Fig. 59-4 ). [56] [57] The associated pathology has been well elucidated by the Cardiff Peritoneal Biopsy Registry and includes loss of the mesothelium, gross thickening of the subserosal matrix, hyalinosis and obliterative changes in postcapillary venules, and most significantly, neoangiogenesis within the peritoneal membrane.[58] The latter change makes the membrane more vascular and explains why transport characteristics should increase and, conversely, why glucose driven ultrafiltration might decrease.

FIGURE 59-4  Changes in peritoneal transport characteristics and ultrafiltration with time on peritoneal dialysis.  (Adapted from Davies SJ, Bryan J, Phillips L, Russell GI: Longitudinal changes in peritoneal kinetics: The effects of peritoneal dialysis and peritonitis. Nephrol Dial Transplant 11:498–506, 1996.)

 

 

The etiology of these changes has long been debated. They may be accelerated by peritonitis episodes, but they can clearly occur in the absence of this complication. There is some evidence that the changes are associated with the systemic inflammatory changes reported in many patients with ESRD.[53] The most popular theory, however, is that the changes are related to cumulative exposure to bioincompatible PD solutions and, in particular, to glucose and glucose degradation products (GDPs). De Vriese and colleagues [59] [60] have described a plausible pathway, involving vascular endothelial growth factor, through which glucose exposure causes proliferation of peritoneal blood vessels. Davies [61] [62] has recently presented impressive clinical evidence that greater cumulative glucose exposure is associated with worse membrane deterioration and that nonglucose fluids such as icodextrin may be protective.

There is also evidence that the fall off in ultrafiltration with time in patients treated with standard glucose PD fluids is disproportionately greater than can be accounted for by the increase in D/P Cr.[62] This raises the question of whether an additional mechanism is involved, such as a fall in hydraulic conductivity related to aquaporin damage or to thickening of the subserosal matrix.

THE PERITONEAL CATHETER

The catheter used to provide access to the peritoneal cavity is critical to the success of PD. A good catheter should achieve consistently brisk dialysate inflow and outflow as well as low rates of infection. There are two main problems with catheters that bioengineers have attempted to resolve. The first is the predisposition to exit site and tunnel infection and to biofilm formation on the intra-peritoneal segment, all of which may ultimately lead to peritonitis and catheter loss. The second is impaired function that usually manifests as poor or absent outflow and that is predisposed to by migration of the catheter tip out of the pelvis and by wrapping of omentum and bowel around the tip. [63] [64]

Catheters are broadly categorized as acute and chronic.[63] Acute catheters are straight and rigid, and inserted at the bedside for short-term dialysis. Any prolongation of use is associated with a high incidence of infection and malfunction. Chronic catheters are designed for long-term use and are commonly made of silicon rubber because it is flexible, induces minimal foreign body reaction, and stimulates squamous epithelium proliferation around the catheter.[65]

Palmer and Quinton[65] were the first to use silicone rubber catheters. Tenckhoff[66] improved upon this design in 1968 by introducing a Dacron internal cuff, a shortened subcutaneous tunnel, and a second cuff placed near the exit site. The standard catheter nowadays is silicone rubber with one or two cuffs. The cuffs act to anchor the catheter and to provide a barrier to ascending infection.

A basic chronic catheter design constitutes four components: the extra-abdominal, subcutaneous, transmural, and intraperitoneal segments ( Fig. 59-5 ). All chronic catheters now have a Dacron cuff at the point of entry into the cavity between the transmural and intraperitoneal segments. A double-cuff catheter has an additional cuff placed about 2 to 3 cm internal to the exit site, in the subcutaneous tunnel. The extra-abdominal segment protrudes out of the exit site. It enables convenient safe handling of the catheter. The subcutaneous segment sits in the tunnel and provides a protection against infection through in-growth of epithelial cells. The intramural segment provides mechanical anchorage and a watertight seal. The Dacron cuff stimulates fibrosis to provide an antibacterial seal. Swan-neck catheters have a permanent bend in the intramural segment between the two cuffs so that the intraperitoneal segment enters the pelvis facing downward, in order to minimize the risk of migration due to so-called catheter memory, while the extraperitoneal segment exits the skin also facing downward.[67]

FIGURE 59-5  Standard peritoneal catheters.  (From Gokal R, Alexander S, Ash S, et al: Peritoneal catheters and exit-site practices toward optimum peritoneal access: 1998 Update. Perit Dial Int 18:11–33, 1998, with permission.)

 

 

The intraperitoneal segment has multiple small sidehole openings to improve fluid and solute movement; it typically has barium-impregnated strips to facilitate radiologic localization. The tip of the intraperitoneal segment can be modified to improve function. Coiled catheter tips are designed to reduce omental obstruction of the catheter and to minimize inflow pain due to a jet effect caused by the dialysate flow.[68] Other catheter modifications have been devised to decrease catheter migration out of the pelvis and to improve catheter function by keeping apart the two layers of peritoneum, the omentum and bowel, all of which can interfere with outflow. Swan-neck, Toronto-Western, Pail-Handle (Cruz), and Ash (Advantage) are some of the commonly used modified catheters ( Fig. 59-6 ). [63] [64] [67] [68] [69] [70] [71]

FIGURE 59-6  Adaptations of the peritoneal catheter.  (From Gokal R, Alexander S, Ash S, et al: Peritoneal catheters and exit-site practices toward optimum peritoneal access: 1998 Update. Perit Dial Int 18:11–33, 1998, with permission.)

 

 

For these purposes, the Swan-neck Missouri catheter has a flange and bead surrounding the catheter distal to the internal cuff and comes with both a straight and coiled tip.[67] Similarly, the Toronto-Western catheter has two Silastic discs 5 cm apart at the tip of its intra-abdominal part to stabilize it and prevent visceral wrapping.[71] The Ash (Advantage) catheter has a T-shaped configuration with two separate intraperitoneal limbs in order to prevent migration and to distribute inflow more widely.[69] The self-locating catheter has 12 g of tungsten inserted into the tip of a conventional Tenckhoff catheter to keep it firmly in the pelvis.[72]

Other modifications exist. The Moncrief-Popovich catheter has a long external cuff and is designed for prolonged subcutaneous burial of the extraperitoneal segment before use in order to promote tissue in-growth in a sterile environment.[73] The Cruz catheter is made of polyurethane, which is stronger and smoother than silicone.[70] This allows for a thinner wall and a larger internal diameter, which was expected to allow faster outflow. It was also hoped to be less favorable to growth of biofilm. Unlike silicon, polyurethane is degraded by alcohol and iodine, which can lead to catheter cracks on repeated exposure to these agents.

Clinical trials comparing catheter designs have been somewhat disappointing with varying results. [74] [75] [76] [77] [78] Regardless of type, overall catheter survival should be 85% to 90% at 1 year.[63] No particular catheter is definitively superior to the standard silicon Tenckhoff in terms of outcomes. However, most authorities prefer double-cuff catheters and some studies have shown these to be associated with longer time to first peritonitis episode, better technique survival, and less resistant exit site infection. [75] [79] [80] However, in a systematic meta- analysis, no difference was reported in peritonitis rates with respect to straight versus coiled or single versus doubled cuffed catheters.[76]

There is no consistent evidence that the other novel catheters described earlier are superior to standard designs, but individual centers may report good experiences with them. [67] [68] [69] [70] [71] [72] [73] [78] In conclusion, the choice of a catheter in a PD program depends on the preference and experience of the physician who will be inserting it.

Catheter Insertion

Patients should be evaluated for suitability for PD before catheter placement. Apart from cognitive and social problems, issues such as multiple previous complicated abdominal surgeries with likelihood of adhesions or the presence of bowel or urinary ostomies or a convincing history of complicated diverticular disease or recent abdominal aneurysm repair are all relative, although not absolute, contraindications.[63] The abdominal wall needs to be assessed looking for defects or weaknesses and hernias, and these may also be contraindications or may at least need to be repaired at the time of catheter insertion.

Preparation for catheter insertion should include emptying of bowel and bladder as well as usual surgical precautions. The exit site should be located away from belt lines and other pressure points, and marked in advance. Prophylactic antibiotics are generally recommended. [81] [82]

Insertion may be performed with the patient under general or local anesthesia. It may involve full surgical dissection and placement, which is best in complicated cases in which adhesions may be present but is associated with higher incidence of significant postoperative ileus and hospitalization. The incision can be midline or paramedian.[83] Blind placement using trocar and guidewire is an alternative and does not require a formal operating room but has a significant failure rate and a small risk of visceral perforation. Peritoneoscopic insertion is increasingly popular and allows visualization without the disadvantages of conventional open procedures. One study shows better technique survival and less early leaks and infections with peritoneoscopic insertion.[84] However, other studies are less convincing and the choice for a given center is again critically dependent on the preferences and expertise of the operator.[63] [85]

Perioperative complications include hemorrhage, which is relatively common, and visceral perforation, which occurs in less than 0.5% of cases.[86] To avoid the catheter clotting up with fibrin or blood, repeated flushing with heparin is done postoperatively until the effluent is clear. Ideally, the catheter dressing should be undisturbed for 1 to 2 weeks to allow good healing and to reduce the rate of pericatheter leaks. For the same reason, it is best not to use the catheter for at least 2 weeks. If PD has to be initiated, low dwell volumes in a supine position should be used.[63]

Catheter Complications

The principal catheter-related complications are exit site and tunnel infection, pericatheter leaks, and poor catheter function. Less frequent are catheter perforations and fractures and trauma to abdominal viscera, usually occurring on insertion. Infections are dealt with under the heading of “peritonitis” and leaks under “non-infectious complications.”

Catheter dysfunction is a relatively common and often frustrating complication of PD. It occurs most frequently at the initiation of PD but can happen at any time. It may be complete two-way, or just one way, outflow obstruction. Or it may just be slow catheter drainage or drainage that is excessively positional. Causes of catheter obstruction include excess fibrin formation occluding the catheter lumen, constipation, catheter migration, and omental wrapping. Obstruction with fibrin can occur in the context of peritonitis or for no obvious predisposing reason. The type most often encountered is a two-way obstruction. It is managed by irrigation of the catheter with heparin or with a fibrinolytic agent such as tissue plasminogen activator.[87] Protocols typically involve the fibrinolytic agent dwelling within the catheter for a period of time. If there is any question that a hypoactive bowel may be contributing to the problem, aggressive prescription of laxatives and subsequent dietary measures are indicated. If the problem does not rapidly resolve, a plain x-ray study of the abdomen is required. This will show if the catheter is in an appropriate position in the pelvis. Catheters with their tips in the upper or middle of the abdomen are less likely to function. Usually, the obstruction in these cases is outflow related. If the catheter appears to be in a good position and does not improve with fibrinolytics or laxatives, omental wrapping is presumed. Sometimes a few days of laxative therapy may resolve the issue. However, if nonfunction persists, catheter manipulation is required. This can be done in the radiologic suite using a guidewire or by a physician or surgeon using a laparoscope, or even an open procedure to reposition the catheter. [88] [89] The latter approach may include an omentopexy or omentectomy.[90] As with catheter insertions, the choice of procedure tends to depend on the available experience and expertise. Of course, any prolonged catheter nonfunction will require a switch to hemodialysis unless there is reasonable residual function, as is often the case, if the problem occurs at or shortly after initiation of dialysis.

If the catheter drainage is unduly positional and remains so despite laxatives and fibrinolytics, a clinical judgment has to be made as to whether it is interfering enough with either the adequacy of the dialysis or with the patient's quality of life to justify a manipulation. Some cases resolve with time or the patient learns to manage the situation but others persist or worsen. One strategy that may help in cycler patients is to perform tidal PD. Here, refilling with solution is initiated after only 50% or 80% or some other preprogrammed percentage of the dialysate has been drained. This works in cases iin which the poor catheter function is apparent only late in the drain cycle.[91] It is also a useful approach in patients with so-called drain pain. It does not detract from the adequacy of clearances and may even improve them in situations in which drain times have been prolonged for only modest increases in effluent volume.

PERITONEAL DIALYSIS SOLUTIONS

The constituents of PD solutions are best considered under the headings of the osmotic agent used; the pH and GDP content, which are closely linked; the buffer used; and the levels of calcium and sodium. Solutions generally do not contain potassium unless this is added before infusion in order to treat hypokalemia. Typical PD fluid composition is shown in Table 59-3 .


TABLE 59-3   -- Typical Composition of Common Peritoneal Dialysis Solutions

 

Conventional Glucose Based

Icodextrin

Low GDP Lactate Buffered

Low GDP Bicarbonate Buffered

 

1.5%

2.5%

4.25%

 

 

 

Na (mEq/L)

132

132

13

132

132

132

Cl (mEq/L)

96

96

96

96

96

96

Ca (mEq/L)

3.5[*]

3.5[*]

3.5[*]

3.5[*]

3.5[*]

3.5[*]

Mg (mEq/L)

0.5

0.5

0.5

0.5

0.5

0.5

Lactate (meq/L)

40

40

40

40

40

40

Bicarbonate (mEq/L)

-

-

-

-

-

25

Glucose (mgs%)

1360

2270

3860

-

1360–3860

1360–3860

pH

5.2

5.2

5.2

5.2

7.3

7.3

Osmolality (osmol/kg)

345

395

484

282

345–484

345–484

GDPs

+

++

+++

+

very low

very low

 

GDP, glucose degradation product.

 

*

Low calcium solutions have 2.5 mEq (1.25 mmol).

 

Osmotic Agent

Standard peritoneal solutions have, since the inception of the modality, been based on the use of glucose as the osmotic agent. Typically, solutions are provided in three different concentrations of glucose so that ultrafiltration can be appropriately varied. In North America, these are presented as 1.5%, 2.5%, and 4.25% dextrose solutions. In Europe, the same concentrations are marketed but are described in terms of their concentration of anhydrous dextrose so that they are 1.36%, 2.27%, and 3.86%, respectively. The osmolarities of these 3 solution concentrations are 345, 395, and 484 mOsm/kg, respectively.

Although glucose has been effective as the osmotic agent in PD for decades, there have been increasing concerns in recent times about its use. These have focused on two broad areas. The first is the toxic effects of glucose on the peritoneal membrane, and there is increasing evidence both from cell culture and animal models and from clinical observations that this concern is justified. [58] [59] [60] [92] Clear pathways through which damage occurs have been described in the experimental setting. Also, recent longitudinal clinical studies have suggested that patients who use more hypertonic glucose have more rapid deterioration in peritoneal membrane function. [60] [61] The second significant concern relates to the metabolic and potentially adverse cardiovascular effects of constant systemic absorption of glucose. These include the induction or exacerbation of hyperglycemia, hyperlipidemia, hyperinsulinemia, and obesity, all potentially important cardiovascular risk factors in a population with an already high rate of cardiac disease.[92] Against this background, interest in alternative osmotic agents to glucose has grown.

The first such agent to be used clinically was amino acids.[93] These are relatively weak osmotic agents when used in concentrations that are clinically tolerated, and so the main purpose has been to supplement nutrition rather than to augment ultrafiltration. A solution of 1.1% amino acids, as is typically used in clinical practice, has an osmotic strength similar to that of 1.5% glucose. This preparation can be used for only one dwell daily because of its potential to exacerbate uremia and acidosis.[94] For the same reasons, stronger amino acid solutions are not used. The composition of the commercially available solution has been designed specifically for the uremic patient in that amino acids such as methionine and cysteine that accumulate in renal failure are not included. A number of clinical trials of intraperitoneal amino acids have been carried out but most have been small and of short duration, and none has been statistically powered to look at the effect of the solution on patient survival. [95] [96] [97] Improvements in nitrogen intake and balance have been identified, and there has been some evidence of an anabolic effect. The best and longest duration study has shown some anthropometric and biochemical benefits with less decline in these indices seen in the treatment versus the control group.[97] However, these are classic intermediate outcomes of uncertain significance and have not led to widespread use of the solution. Indeed in the large U.S. market, the product is not even approved for use. At present, intraperitoneal amino acids are seen as a largely unproven intervention requiring further evaluation.

A more successful alternative to glucose has been icodextrin.[38] This is a mixture of glucose polymers with a mean molecular weight of about 20,000 kD. Icodextrin induces ultrafiltration by oncotic rather osmotic pressure, and the solution is actually isosmotic relative to normal plasma.[98] The icodextrin molecule is too large to diffuse across the peritoneal membrane, so systemic absorption can occur only through lymphatic flow. [99] [100] This does occur but not to a degree that impairs the ability of icodextrin to induce ultrafiltration in a sustained manner during long dwells. Accordingly, the solution is most useful for the long nocturnal dwell of CAPD and the long diurnal dwell of automated peritoneal dialysis (APD), where it will typically remove between 300 and 600 mL of fluid. The main theoretical concern about icodextrin is that it leads to unphysiologically high levels of maltose and of other oligosacccharides in serum.[100] No toxic effect of these has been apparent despite over a decade of clinical use. Side effects include skin rashes and occasional aseptic peritonitis, which has been related to contamination in the manufacturing process.[101] Icodextrin is a significant advance in the practice of PD. It has been shown in randomized controlled trials to improve volume status as measured by bioimpedance and by other techniques assessing body composition. [102] [103] It has also been shown to improve echocardiographic indices, but surprisingly, no consistent beneficial effect on blood pressure has been demonstrated. [102] [103] [104] It is particularly useful in high transporter patients who tend to resorb fluid during conventional long dwells. It has been shown to prevent ultrafiltration failure in such patients and to prolong technique survival on PD.[105] Recently, evidence has appeared that peritoneal membrane function in long-term PD patients is better preserved in those using icodextrin as compared with standard glucose regimens, but these observations are not from randomized trials.[62]

More recently, icodextrin has been evaluated for its use as a glucose-sparing agent. There is some evidence that it may lead to a less adverse lipid profile, easier better glycemic control, less hyperinsulinemia, and less obesity and, theoretically, all this may be of benefit in reducing cardiovascular risk. [106] [107] [108] [109] Icodextrin is now being used for its glucose-sparing metabolic advantages as much as for its benefits in maintaining ultrafiltration.

The use of icodextrin, however, is typically limited to one dwell a day, because of its time course of action, its relatively high cost, and persisting theoretical concerns about possible toxic effects of a high serum maltose.

Other osmotic agents, such as fructose, xylitol, sorbitol, and mannitol, have been investigated, but none has been appropriate for clinical practice. Glycerol, combined with amino acids, has been the subject of clinical trials and appeared safe and effective.[110] There is a clear need for a nonglucose agent that is not absorbed and that can be used in short duration dwells.

pH and Glucose Degradation Products

One of the great concerns about using glucose as an osmotic agent is that it is converted to GDPs such as formaldehyde and glyoxal during the heat sterilization process.[111] If sterilization is carried out at a pH of 7.2, massive GDP generation occurs and there is caramelization of the glucose. GDPs are toxic to the peritoneal membrane in a manner similar to glucose and lead to deposition of advanced glycosylation end products (AGEs). These activate mesothelial cell surface AGE receptors, which induce mesothelial cells to differentiate into myofibroblasts and promote peritoneal fibrosis.[60] Indeed, it may be that most glucose damage to the membrane is mediated by GDPs.

In order to decrease GDP generation and to avoid glucose caramelization, heat sterilization is carried out at a pH of about 5.2. Lower values would be better but such acidic solutions could not be infused into the patient's abdomen because they would cause pain and because low pH of itself may be toxic to the membrane and to host defenses. The present pH of 5.2 for glucose-based fluids, and indeed for icodextrin, is thus a compromise between the harm that may be done by GDPs and that which may be done by low pH. In an effort to break out of these constraints, multipouch solution bags have been developed by all the major companies who manufacture PD solutions. These involve separation of glucose from the other components of the solutions ( Fig. 59-7 ). The glucose-containing pouch can then be sterilized at a much lower pH of 3.0, thus minimizing GDP generation. The pouch containing the constituents other than glucose can be safely sterilized at an alkaline pH of about 8.0. The two pouches are then mixed when the patient breaks a frangible partition just before infusion, resulting in a normal pH, low GDP solution. Examples of such a product are the Fresenius solution, Balance, and the Gambro solution, Gambrosol Trio/Unica. Although these are normal pH solutions, the buffer used is lactate. Although theoretically attractive, it is not yet clear what the clinical benefits of these solutions are. They do avoid the relatively uncommon problem of infusion pain, which is pH related and which can still occur with standard pH 5.2 solutions. The more critical question is whether these normal pH low GDP solutions are less damaging to the peritoneal membrane. Evidence for this comes from cell culture and animal models and also from the use of surrogate end points in clinical studies. [112] [113] Thus, in randomized trials, these solutions led to significantly higher PD effluent levels of CA125, which is considered to be a marker of mesothelial mass and so of membrane health. They also may have systemic benefits with lower C-reactive protein levels.[114] There is even some suggestion that they are associated with better preservation of residual renal function, perhaps because systemically absorbed GDPs may be nephrotoxic.[115] There is, however, no consistent clinical evidence that they lead to better preservation of membrane transport or of ultrafiltration capacity. With regard to more critical end points such as technique or patient survival, there is observational data suggesting a survival benefit.[116] However, there is a high likelihood that this finding may be confounded by the greater likelihood of healthier patients receiving these more expensive solutions.[117] Large randomized trials looking at technique and patient survival are required and are being initiated. For the present, use of low GDP solutions is limited to a minority of patients in Western Europe and East Asia, where enthusiasm for this approach has been greatest.

FIGURE 59-7  Double-pouch solution bags with different pH in each pouch during heat sterilization and before use.

 

Buffers

It is important to distinguish between the issue of the pH and the buffer used when considering PD solutions. The buffer has traditionally been lactate, which is then converted in the liver to bicarbonate.[112] Acetate was briefly used 2 decades ago but has no advantages over lactate and may even predispose the patient to encapsulating peritoneal sclerosis. Lactate is generally an effective buffer, and most PD patients have well-controlled acid-base status, mainly because of the continuous nature of the modality.[112] However, concerns have been raised about the biocompatibility of lactate, and there is some bench science data to support the notion that lactate may be bad for the peritoneal membrane and for host defenses. [118] [119] The use of bicarbonate has been limited by concerns about precipitation of calcium and bicarbonate if both are constituents of the same solution. However, the same multipouch technology that has allowed the generation of low GDP solutions can be used to solve this problem. Thus, the calcium and magnesium can be stored in the same low-pH pouch as the glucose, while the bicarbonate buffer can be in the high-pH pouch with the other constituents of the solution. When mixing occurs before infusion, it results in a normal-pH, low-GDP, bicarbonate-buffered solution. In fact, such solutions may still contain some lactate because of concerns that high bicarbonate solutions may lead to high carbon dioxide levels that may not be desirable. [118] [119] [120] For example, the Baxter solution, Physioneal, contains 25 mmol/L of bicarbonate and 15 mmol/L of lactate. [118] [119] [120]The Fresenius BicaVera product, however, contains only bicarbonate at a concentration of 34 mmol/L.[121] A key question is whether bicarbonate-buffered, low-GDP, normal-pH solutions have any advantage over lactate-buffered ones. So far, there is no clinical evidence that they do. Compared with standard lactate-based, low-pH, high-GDP solutions, bicarbonate solutions appear to show the same advantages as have been demonstrated for lactate-based, low-GDP solutions. [118] [119] [120] [121] They decrease infusion pain and raise effluent CA210 levels, but again convincing data on membrane or host defense benefits or on technique or patient survival are lacking. [118] [119] [120] [121] [122] Large studies are required to demonstrate these benefits if the higher cost of such products is to be justified.

Calcium

There has been a trend over the past 2 decades to reduce the calcium content of PD solutions. Traditionally this was 1.75 mmol/L (3.5 mEq/L) which is notably higher than the typical serum ionic calcium level of 1.1 to 1.3 mmol/L in patients with renal failure. This results in diffusion of calcium into the patient resulting in positive calcium balance. Although this was previously considered desirable in that it helped suppress hyperparathyroidism, it became less attractive with the widespread use of calcium-based phosphate binders and 1-hydroxylated vitamin D products. Hypercalcemia, excessive parathyroid suppression, and adynamic bone diseases became increasing concerns, and there was a switch to lower cacium dialysis solutions in both PD and hemodialysis. The most common calcium concentration now used is 1.25 mmol/L (2.5 mEq/L), and this may lead to neutral or even negative calcium balance in PD patients taking calcium-based binders and vitamin D. [123] [124] More recent concerns about the effect of positive calcium balance on vascular and other soft tissue calcification have heightened the popularity of low-calcium solutions. Their use has likely led to more hyperparathyroidism, but this has been considered an acceptable trade off. Concerns that low calcium solutions might be associated with more peritonitis have not materialized.[125]

Sodium

Interest in variations in PD solution sodium concentrations has not been so great but has recently revived. Standard solutions typically contain 132 mmol/L, so there is only a modest concentration gradient to promote sodium removal at the start of a dwell. [126] [127] Consequently, most sodium removal occurs by ultrafiltration. However, because of sodium sieving, the ultrafiltrate dilutes the dialysis solution sodium level early in the dwell and it may drop to 120 to 125 mmol/L, resulting in some diffusive loss of sodium as the dwell proceeds (see Fig. 59-3 ). [25] [26] It is sometimes said, therefore, that sodium removal early in a dwell is mainly convective and later in a dwell is mainly diffusive. Increased awareness of the high prevalence of hypertension in dialysis patients has led to concerns that volume status is not being adequately normalized and has increased interest in low-sodium solutions, which might facilitate greater sodium removal. [128] [129] One problem is that if sodium levels are reduced, glucose levels would need to be raised to maintain osmolarity and this may not be considered a desirable trade off. Again, clinical trials are pending.

THE PERITONEAL MODALITIES

PD for ESRD is nowadays delivered either as CAPD or as APD. The old modality of intermittent PD whereby patients attended a PD unit two or three times a week for 10 to 24 hours of treatment at a time is now considered an inadequate therapy and has become rare, except as a temporary measure pending initiation of training or after an abdominal operation.[6] It will not be further considered here.

CAPD was introduced in 1977 and rapidly became popular because it allowed patients to have a simple form of dialysis under their own control in their own homes.[9] To this day, these features remain the great appeal of PD. CAPD usually involves four exchanges of 1.5 to 2.5 L of solution daily, with the night dwell being 8 to 9 hours and the day dwells being 4 to 6 hours each ( Fig. 59-8 ). Some centers commonly prescribe only three dwells daily for cost or convenience, introducing a fourth only if problems arise or if residual renal function is lost.

FIGURE 59-8  Continuous ambulatory peritoneal dialysis and automated peritoneal dialysis regimens.

 

Most centers now use a double-bag format, with both fill and drain bags preattached to a Y-shaped transfer set of tubing ( Fig. 59-9 ).[130] This method removes the requirement for the patient to spike or connect the tubing to the fill bag, and this procedure has been repeatedly shown in randomized controlled trials to reduce the risk of peritonitis substantially.[130] It does still require the patient to make a connection between the transfer set tubing and the extension tubing, which is, in turn, attached to the catheter usually through a titanium connector. In different CAPD giving sets, this particular connection step is modified in different ways to reduce the risk of accidental contamination, but there is no conclusive evidence which is superior.[131]

FIGURE 59-9  Double-bag system with transfer set and extension tubing showing three phases of the exchange procedure: (A) drain, (B) flush, and (C) fill.

 

APD has greatly increased in popularity in the past decade and is now used more than CAPD in the United States.[132] Its appeal is that it frees up the daytime from PD procedures for patients and their caregivers. It also has the capacity to deliver more clearance and remove more fluid. Its recent popularity also reflects the improved technology and design of modern cyclers, which are more compact, light, portable, reliable, and easier to operate than previous models.[133] Typically, patients on APD have three to seven cycles of 1.5 to 2.5 L delivered over 9 hours at nighttime (see Fig. 59-8 ). Dwell times may range from 45 minutes to 3 hours, depending on prescription requirements. Most patients leave a dwell in at the end of the cycling period and drain this dwell out again before the next cycling period about 15 hours later. This is called continuous cycled peritoneal dialysis. Others have no day dwell, usually because of good residual renal function or mechanical contraindications and this is called nocturnal intermittent peritoneal dialysis or day dry APD. Still other patients have more than one day dwell, requiring them to do an exchange sometime during the day. This is sometimes called high-dose APD or PD plus, but these are proprietary names and the simple description of APD with 2 day dwells is preferred (see Fig. 59-9 ).[134]

Other APD prescriptions involve one or more short day dwells, which leave some of the day time dry either to facilitate ultrafiltration or for comfort or mechanical reasons.

Most cyclers now use hydraulic pumps rather than gravity to deliver and drain the PD solution.[133] They have the facility to do tidal PD which involves deliberately allowing an incomplete drain of a proportion of the infused fluid (e.g., 50% or 80%) before refilling with the next cycle.[135] This can be used to minimize down time with a poorly draining catheter or to avoid drain pain. Cyclers also allow the daytime solution to be delivered before disconnection and can be used for additional daytime exchanges in an approach that decreases cost and may increase convenience, compared with doing those exchanges with CAPD tubing.[134]

THE PERITONEAL PRESCRIPTION

Determination of the PD prescription requires consideration of the clearance and volume removal requirements of the patient. However, it also needs to take into account nutritional, cardiovascular and metabolic status because all of these factors may be influenced by systemic glucose absorption. The social situation is important in that it will influence capacity and willingness to carry out simple versus more complicated prescriptions. Finally, cost is inevitably a consideration.[136]

Clearances and Adequacy

The term adequacy is generally used to refer to clearances achieved on dialysis. It was the major focus of clinical discussion and investigation in PD throughout the 1990s and a variety of guidelines were generated. [137] [138] [139] [140] However, two well-conducted major randomized clinical trials published in the early 2000s failed to show benefit for augmented prescriptions as compared with standard ones. [141] [142] An analogous trial in hemodialysis was also negative, and so there has been a move to de-emphasize high clearance targets and to return to recommendations more easily achieved with standard prescriptions.[143]

The fundamental misunderstanding with regard to clearances was the notion that residual renal small solute clearance could be replaced in terms of its survival implications with equivalent amounts of peritoneal small solute clearance, but it is now manifestly clear that this is not so.[144] The literature is totally consistent and convincing on the association between residual function and multiple clinical outcomes on PD, including patient survival, but no such association has been shown for peritoneal clearances within the commonly prescribed range. [138] [142] [144] Therefore, the impression is that once a minimum level of clearance sufficient to maintain life and avoid acute uremic complications has been achieved, further increments within this prescribed range do not add further benefit for the patient, and because such increments are often onerous and expensive and increase exposure to glucose, they may have an overall negative effect.[144] The question now is to define what this minimum level is and the approach has been to play safe and recommend no less than the lower doses used in the randomized trials. [145] [146] It is of course possible that substantially higher clearances than were used in the trials, both in PD and in hemodialysis, might confer incremental benefit but there is no proof of this and no easy way to deliver such therapy in PD anyway.

These conclusions may seem somewhat pessimistic, but the findings are positive in that they have moved PD, and dialysis generally, away from what now appears to have been a therapeutic cul-de-sac and allowed investigators to concentrate on issues much more relevant to patient outcomes. This section therefore is relatively brief and focuses on the knowledge required to deliver dialysis doses sufficient to meet the newer more modest targets.

Measurement of Clearance

Clearance on PD is a combination of peritoneal clearance and that due to residual renal function. [138] [139] The latter is especially important in PD because it is maintained for longer and is a larger proportion of the total clearance than in hemodialysis. It is typically measured using either or both of two indices—fractional urea clearance, imported from hemodialysis and best known as Kt/V, and creatinine clearance normalized for body surface area (CrCl). The calculation of these two indices requires 24-hour collections of PD effluent and urine with measurements of urea and creatinine content, a simultaneous blood sample to measure serum urea and creatinine, and then a simple standard calculation of clearance. For the calculation of the residual component of CrCl, a mean of renal urea and creatinine clearance is used because residual renal creatinine clearance is known to substantially overestimate the true glomerular filtration rate. The clearance is then normalized to a measure of body size, which for urea is, by convention, the total body water or “V” to give Kt/V and for creatinine is the body surface area (BSA) to give CrCl. Both BSA and V are estimated from formulas based on height, weight, and sex.[139] It is recommended that desirable body weight be used in these formulas because excess fat tissue is not believed to require more clearance. The resulting daily values are multiplied by 7 and expressed per week. This was originally done to facilitate comparison with hemodialysis though there is now widespread skepticism about the validity of such comparisons between intermittently delivered and continuously delivered dialytic modalities. [147] [148]

Clearance Targets

The ADEMEX study was the largest ever randomized controlled trial done in PD, involving almost 1000 patients followed for 2 years at multiple centers in Mexico.[141] It was a very well done study and showed convincingly that the augmented clearance regimens recommended by the NKF-DOQI and other bodies in the 1990s did not confer any benefit on patients as compared with standard 4 × 2 L CAPD. This finding was confirmed by another randomized trial from Hong Kong comparing three levels of Kt/V.[142] These studies led to a reduction in weekly Kt/V targets by KDOQI from 2.0 to 1.7 with the same target for both APD and CAPD and for all transport types. [145] [146] CrCl was considered to add no extra useful information, although some groups still support its use in high transporters.

Some caveats need to be stated and some issues remain to be resolved. It is still reasonable practice if a patient is doing poorly and is suspected of being uremic to give a trial of increasing dialysis dose, even if Kt/V already exceeds 1.7 weekly. Conversely, if a patient cannot tolerate the prescriptions required to reach 1.7 weekly but is clinically well or has a limited prognosis for other reasons, a clinical judgment to continue PD may be very appropriate. There is also some concern that the populations studied in the important trials had relatively low rates of cardiovascular disease and so again a higher dose might be tried if such patients appeared uremic.

The existing studies did not include APD patients, and so it is unknown if this literature applies to them. However, in the absence of evidence that it does not, the assumption that it does seems reasonable. The importance of a day dwell in APD patients who can achieve Kt/V 1.7 with cycling alone is also controversial. The essence of the concern expressed by some is that middle molecule clearance will not be adequate in such situations. If the patient still has residual function of greater than 2 to 4 mL/min, this concern is not justified because even the failing kidneys will provide markedly more middle molecule clearance than one PD day dwell. In the anuric patient who would typically have to be small or a high transporter to achieve a Kt/V of 1.7 from cycling alone, the concern is more real, however, and although there is no convincing clinical evidence, addition of a day dwell would seem reasonable.

Factors Affecting Clearance

It will be apparent that residual renal clearance and body size are two of the three major determinants of Kt/V and CrCl.[149] The third is of course the peritoneal prescription, and this is the one that can be altered ( Table 59-4 ). The membrane transport characteristics influence the delivered clearance for a given prescription. This is particularly so for CrCl because, as can be seen in the standard PET curves, creatinine transport exhibits a wider range of variation than urea transport (see Fig. 59-1 ). Similarly, membrane transport has bigger effect in APD than CAPD because the short cycles used in the APD are associated with greater variations in equilibration than the longer duration dwells of the CAPD.[149]


TABLE 59-4   -- Peritoneal Dialysis Prescription—Variables Affecting Clearance

CAPD

APD

 

Cycler Component

Day Dwell Component

Number of dwells

Time on cycler

Number of day dwells

Dwell volumes

Frequency of cycles

Duration of day dwells

Type and tonicity of solution

Cycler dwell volumes

Day dwell volumes

 

Type and tonicity of solution

Type and tonicity of solution

 

APD, automated peritoneal dialysis; CAPD, continuous ambulatory peritoneal dialysis.

 

 

 

CAPD prescription is essentially simple, and clearance can be increased either by raising the dwell volume from 1.5 or 2.0 to 2.5 or 3.0L or by going from three or four exchanges daily to four or five. Both are effective, but increasing dwell volume is usually preferred because it allows better spacing of dwells, is less intrusive on patient time, is associated with better adherence, and is less expensive. [149] [150] However, increasing dwell volume may be limited by patient tolerance related to mechanical symptoms such as bloating and back pain, and to histories of hernias and leaks. Clearances can also be raised modestly by increasing ultrafiltration, but this is generally not favored because it entails greater exposure to hypertonic glucose with consequent metabolic and other side effects.

Generally, the Kt/V target of 1.7 per week can be achieved, even in the anuric state, using four dwells of 2 or 2.5L daily in all but the largest of patients, especially when correction for body size is based on appropriate weight rather than actual weight, which is often higher due to obesity.

The effect of the prescription on clearance in APD is a little more complex (see Table 59-4 ). It is helpful to think in terms of the component due to the cycling and that due to the day dwell. Modifications of the latter are more effective in raising clearance. Thus, the introduction of a day dwell in a previously “day dry” patient raises Kt/V and CrCl by 30% to 40%, and a second day dwell will add a further 15% to 25%.[149] This reflects the time dependence of Kt/V and particularly of CrCl. The volume of the day dwell can also be raised to modestly improve clearance.

The cycler component of clearance is affected by the duration of cycling, but in practice, there is rarely scope to go beyond 9 hours and sometimes patients request shorter times. Larger cycler dwell volumes can increase clearance a little but the effect may be modified if intraperitoneal pressure rises sufficiently to compromise ultrafiltration. [151] [152] [153] [154] Durand and colleagues[151] recommend measurements of intraperitoneal pressure to identify a patient's optimal fill volume but this is not routine practice. They find that it is typically in the 2- to 2.5-L range, depending on patient size and other unclear factors. Increasing the frequency of cycling also raises clearance. Any increase in down time due to more time being spent draining and refilling is offset by the benefits of replenishing the dialysate to plasma gradient for diffusion and ultrafiltration. It has been known for decades from the old acute PD literature that hourly cycles give better clearances than less frequent regimens. and this has been reconfirmed in the modern era. [155] [156] [157] [158] However, this is not a cost-effective strategy to raise clearance and, in practice, seven cycles over 9 hours is as high as most centers go, unless there are no other options.[136] Tidal PD was conceived as a method to reduce down time with a view to raising clearance.[135] However, repeated studies have shown no benefit for 50% tidal regimens. [157] [159] [160] [161] Lower degrees of tidal PD are helpful for relieving drain pain and may actually improve clearance if catheter drainage is suboptimal because it is well recognized that the large majority of drainage occurs in the first 8 to 10 minutes of the drain cycle.

In conclusion, a reasonable approach to APD prescription is to start with four or five 2-L cycles over 9 hours at night with a single day dwell, or even no day dwell at all if residual function is substantial. In many centers, the single day dwell will nowadays be icodextrin, but if glucose is used, an early drain may be required in some patients to avoid excess fluid absorption. If this single-day-dwell approach does not achieve the clearance target, either from the start or subsequently as residual function declines, then there are two common approaches. Either the number of cycles can be increased to six or seven nightly or a second day dwell can be added. Adding a second day dwell is less expensive but adds an additional procedure and so is a little more onerous for the patient or caregiver. Increasing the number of cycles may raise clearance a little less but is generally well tolerated. The choice should, if possible, take into account the patient's lifestyle circumstances.

Volume Status

Although the high prevalence of cardiovascular disease and its adverse effect on survival is well recognized in ESRD, no study has clearly shown that a measure of volume status predicts outcome on PD. The relationship between blood pressure and survival in ESRD generally is confounded by the tendency of the least well patients, including those with cardiac failure, to have low blood pressure. The association of high transport status with adverse outcome has been interpreted as an indicator of the importance of volume status in PD but it can also be argued that high transport status reflects comorbidity and inflammation. [48] [49] [50]

Other investigators have noted that higher salt and water removal on PD are associated with better outcomes, and this has been interpreted as a measure of the adequacy and importance of volume control. [162] [163] However, it can be convincingly argued, extrapolating from hemodialysis, that daily salt and water removal reflects salt and water intake and not the degree of volume control.[164] Ultimately, prospective studies of volume status using body composition techniques such as bioimpedance or isotope dilution are required to answer these questions. In the meantime, the recommended approach is to presume that fluid overload is undesirable and to optimize volume control. This means attempting to normalize blood pressure and remove edema while avoiding hypotension and symptoms of dehydration and, at the same time, trying to limit excess exposure to hypertonic dialysate glucose.[57]

Strategies to control volume status should always include consideration of salt intake.[57] Although PD has been considered to allow liberal salt and water intake, this is emphatically not the case when high blood pressure or other signs of volume overload are present. Salt removal through the urine should also be promoted using high-dose loop diuretics as required and either angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs) as well as other validated strategies to preserve residual renal function. [165] [166] [167] Peritoneal strategies should include a focus on the long-duration nocturnal dwell in CAPD and day dwell in APD. If fluid resorption or minimal ultrafiltration is being achieved in these settings and if there is evidence of volume overload, a switch from glucose to icodextrin should be considered. [57] [98] Alternatively, the dwell can be shortened by doing an early drain, followed by either a dry period or an additional dwell. Ultimately, more hypertonic glucose will often be required to normalize volume and should be used, if necessary. However, other strategies should be used first in the light of real concerns about the toxic effects of excess glucose absorption on the metabolic and cardiovascular risk profile and on membrane longevity.[92]

Controversy has arisen recently about the efficacy of APD in managing volume status. It has previously been believed that the short cycles of APD would improve fluid removal. However, it has been argued that during short cycles, the effect of sodium sieving becomes greater as the dwell is not long enough to allow diffusion to contribute to sodium removal, and so a slightly lower effluent sodium results. Rodriguez-Carmona and colleagues have presented data showing less effective sodium and fluid removal on APD relative to CAPD, in part due to this sodium-sieving effect and in part due to fluid resorption during the long day dwell.[168] This is not randomized data, and the literature in general does not show worse control of volume and blood pressure on APD, but the issue needs further investigation.[132]

Management of hypertension in PD patients is generally considered to be part of the management of volume status. There are, however, patients whose blood pressure remains high with any degree of fluid removal that can be tolerated. In this situation, antihypertensives are required. Preference can be given to converting enzyme inhibitors, angiotensin receptor blockers, and loop diuretics because of their benefits in maintaining renal clearance and urine output, but additional agents may be needed.

Ultrafiltration Failure

Ultrafiltration failure is defined as clinical evidence of fluid overload occurring in association with less than 400 mL of ultrafiltrate after a standardized 4-hour duration dwell with a 4.25% glucose 2-L dwell. [44] [57] Before addressing this topic further, it should be noted that there are many other causes of fluid overload in PD patients, and these should be excluded before ultrafiltration failure is diagnosed. These include excess salt and water intake, loss of residual urine output, incorrect choice of dialysis solution tonicity, noncompliance with exchanges, large postdrain residual volume owing to leaks or catheter dysfunction and impaired osmotic gradient due to hyperglycemia.[57]Often, a combination of such factors coexists.

True ultrafiltration failure is rare in the first 1 to 2 years on PD but becomes progressively more common after that owing, as already mentioned, to increases in peritoneal transport characteristics (see Fig. 59-4 ). [54] [55] [56] [169] The more vascular membrane leads to more rapid dissipation of the glucose osmotic gradient and so to less fluid removal. In more marked cases, this causes what is called type I ultrafiltration failure. Less common is type II ultrafiltration failure, which is associated with decreased peritoneal transport due to reduced peritoneal surface area. This is most often due to adhesions after peritonitis and is not usually consistent with ongoing effective PD.[57] Type III ultrafiltration failure is due to excessive lymphatic absorption from the peritoneal cavity and is usually a diagnosis of exclusion as transport status may be high, low, or normal. [57] [170] Other variants of ultrafiltration failure are those due to aquaporin dysfunction or to deceased hydraulic conductance of unclear etiology. [57] [171] Again, it should be emphasized that different causes of fluid overload may coexist. For example a patient with high lymphatic flow may lose urine output and become a more rapid transporter with time on PD.

Management of all cases of ultrafiltration failure include general measures such as review of salt and water intake and restriction of these as indicated, consideration of high-dose loop diuretics if urine output is still significant and review of a recent PET result to help make the exact diagnosis.[57] Type I cases need to avoid long glucose-based dwells and should switch to either icodextrin for such dwells or to APD or to both.[57] If icodextrin is not available, APD with short day dwells may be effective. Type II cases need to switch to hemodialysis, whereas other types can be tried on APD or icodextrin or both.

NUTRITION, INFLAMMATION, RESIDUAL RENAL FUNCTION, AND CARDIOVASCULAR DISEASE

These four topics are now recognized to be closely interconnected and so are dealt with in a common section.

Nutrition

The prevalence of malnutrition in ESRD is high, and multiple studies using a variety of measures have confirmed this finding in PD patients. [172] [173] [174] Importantly, nutritional indices have consistently been shown to predict clinical outcomes in PD, just as they have in ESRD generally. Some of these nutritional measures include serum albumin and prealbumin, lean body mass, total body nitrogen, creatinine excretion, anthropometric indices, subjective global assessment, and a variety of composite nutritional scores. [47] [137] [172] [173] [174] [175] [176] [177]

It is not clear, however, that these indices reflect only nutrition in the sense of adequate dietary intake. Many are more affected by factors such as inflammation and comorbid illness.[178] For example, the major determinants of the serum albumin are dialysate protein losses, the presence of systemic inflammation as indicated by serum C-reactive protein levels, and volume status and do not include dietary protein intake. [47] [179] [180] This point is important because it raises the question of whether purely nutritional supplementation will be effective in preventing the predicted adverse outcomes.

Unfortunately, evidence for the efficacy of nutritional supplements is very slight. The few controlled studies looking at oral nitrogen supplements in PD are negative or inconclusive, although this approach continues to be widely used when protein intake is low. [181] [182] [183] Intraperitoneal amino acids have been studied more extensively and do increase nitrogen intake but have less effect on other endpoints. [93] [94] [95] [96] The best and longest randomized trial done to date shows some better preservation of anthropometric and biochemical measures in treated patients, but these end points would be considered to be intermediate ones of uncertain clinical significance. Also, there is a lack of a large well-done trial looking at truly important outcomes such as patient and technique survival, hospitalization, or quality of life.[97]

Other investigators have looked at measures to improve nutrition that go beyond supplementation and involve altering the uremic milieu. One such approach includes correction of acidosis or even induction of mild alkalosis using oral alkali. This strategy is based on the extensive literature showing that acidosis induces catabolism in ESRD. [186] [187] Two randomized studies in PD patients have confirmed that this approach improves nitrogen balance and may increase muscle circumference and decrease hospitalization rates. [184] [185] Acidosis is uncommon in patients with PD owing to the continuous nature of the dialysis, and it is interesting that these studies actually induce mild alkalosis rather than correct acidosis. Another anabolic strategy is to administer recombinant growth hormone or insulin-like growth factor I. [189] [190] Both of these agents were strikingly effective in improving nutritional indices in short-term studies, but the use of each is limited by cost and particularly by evidence of serious side effects. An analagous, less expensive approach is to administer anabolic steroids. A randomized trial of this strategy done in a mixed cohort of PD and hemodialysis patients showed improvements in nutritional indices including such practical measures as speed of walking and grip strength.[191] However, the strategy is uncommonly used in clinical practice.

Other approaches to improve nutrition such as raising dialysis dose are often attempted but have not been effective in randomized trials. [141] [192] [193] More novel treatments such as administration of ghrelin are being assessed, and short-term benefits have been demonstrated.[194]

The conclusion has to be that malnutrition in uremia is not generally correctable by nutritional supplements alone and that an effective strategy has to deal with the underlying cause, which frequently involves inflammation, comorbidity and an imbalance between anabolism and catabolism, which may be inherently part of uremia and acidosis or which may itself reflect underlying inflammation.

Inflammation

If the 1990s were the era of adequacy in the dialysis literature and at academic meetings, then inflammation has been the focus of the present decade. The observations of Kaysen, Bergstrom and Stenwinkel, in particular, led to a recognition that nutritional abnormalities in uremia were related to activation on a chronic basis of the so-called acute phase response. [195] [196] [197] The description of the high prevalence, often higher than 50%, of elevated serum CRP levels in patients on PD and hemodialysis both showed the widespread nature of the disorder and the identification of the association with atherosclerotic heart disease led to the coining of the term malnutrition inflammation syndrome (MIA). [197] [198] [199] Recent work has extended this syndrome in patients with PD to include other features such as more rapid loss of residual renal function, faster rises in peritoneal transport status, and cardiac disease generally. [53] [200] [201] [202] [203]

In essence, a high proportion of ESRD patients appear to be consistently in a state of acute inflammation despite the absence of detectable foci of infection.[199] This process is driven by cytokines such as interleukin-1, interleukin-6, and tumor necrosis factor, which induce catabolism and lead the liver to switch from albumin production to that of acute phase reactants such as ferritin, ceruloplasmin, fibrinogen, and alpha macroglobulin. [195] [205] This effectively involves the sacrifice of nutritional status in favor of protection of the host from a systemic inflammatory insult. Some have speculated that the insult is atherosclerosis, which is a form of chronic inflammation that may even have an infective etiology. [206] [207] More commonly, the vascular disease of MIA is thought of as another consequence rather than as a cause of the inflammation. In this general area of uremia, malnutrition, cardiac disease, and loss of residual renal and peritoneal membrane function, an understanding of the direction of causation continues to be elusive.

Therapeutic strategies to deal with MIA have been even slower to appear than an understanding of the condition's pathogenesis. Small studies indicate that agents such as thiazolidinediones, statins, and ACE inhibitors may have relevant anti-inflammatory effects but the impact is likely modest. [208] [209] [210] [211] [212] More targeted interventions aiming at key cytokines such as interleukin-6 and tumor necrosis factor may be more promising.[211]

Residual Renal Function

It has long been recognized that residual renal function is better preserved on PD than on hemodialysis. This observation, first made in the early 1980s, has been confirmed in multiple contemporary studies. [213] [214] [215] [216] [217]Likely explanations include greater hemodynamic stability and less episodes of volume depletion, lack of exposure to an artificial extracorporeal membrane, and perhaps lower dietary protein intake. Interestingly, one study showed that hemodialysis patients treated with high-flux polysulfone membranes and ultrapure dialysate kept their residual function as long as a cohort of PD patients.[218]

A recurrent theme in PD-related research over the past decade has been the critical importance of residual renal function in the determination of patient outcomes. This observation was initially made with regard to patient survival in cohort studies such as CANUSA and from retrospective analyses of large databases. [219] [220] Thus CANUSA found that for every extra 5 L/week of residual clearance the risk of mortality fell by 12%.[219]

Subsequent studies have shown a consistent association between residual function and cardiac end points such as volume overload, left ventricular hypertrophy, and congestive heart failure. [221] [222] [223] A major association between preserved renal function and control of hyperphosphatemia has also been reported.[224] Multiple nutritional indices have consistently been found to be worse in those without residual function. [225] [226] More recently, an association between residual function and systemic inflammation has been reported, and Chung and associates [53] [227] have identified an adverse prognostic triad of more rapid loss of residual function, increased peritoneal transport status, and systemic inflammation as manifested by high serum C reactive protein levels and found that these are all significantly correlated.

There is thus a developing concept of the whole MIA syndrome being associated with impaired cardiac status, increased membrane transport, and the development of anuria. What is unclear from all these studies is the direction of causation. Does the loss of residual renal function lead to volume overload and uremic inflammation, which, in turn, promotes malnutrition and membrane deterioration and further volume overload with adverse cardiac consequences and death? Alternatively, is the main driver systemic inflammation, which induces not only malnutrition and membrane deterioration but also more rapid loss of renal function? Or is the truth a more complex interaction of both explanations? The question is central because it raises the issue of how important are strategies to preserve renal function for as long as possible in PD patients. Are such approaches actually going to prolong survival or is a strategy targeting uremic inflammation more likely to reduce mortality?

These questions particularly arise with the findings in two recent small but impressive randomized controlled trials that ACE inhibitors and angiotensin receptor blockers promote preservation of renal function in PD patients. [166] [167] High-dose furosemide has also been shown to preserve urinary volume better than placebo, but there was no effect on clearance.[165] Other studies have varyingly shown associations between a variety of other factors and more rapid loss of renal function on PD. [215] [216] [228] [229] [230] [231] These have included a number of peritonitis episodes, diabetes mellitus, proteinuria, cardiovascular disease, volume of PD fluid used daily, and episodes of dehydration. It is less clear how important aminoglycoside and contrast exposure are to renal function on PD, and although studies are small with inconsistent results, it seems prudent to minimize use of these. [230] [232] Others have suggested that systemically absorbed GDPs from sterilization of dialysate glucose may be nephrotoxic, and there is preliminary evidence suggesting that the use of low GDP solutions may help preserve residual function.[233]

More controversial is whether APD is less protective of residual function than CAPD. This observation has been made in four small studies but was not detected in five larger ones and so cannot be considered proven. [215] [216] [228] [229] [230] [231] [234] [235] [236] Theoretically, APD might be more nephrotoxic owing to exposure to more PD solution and so more systemically absorbed GDPs, but this is speculative.

Cardiovascular Disease

Just as is the case for ESRD generally, cardiovascular disease is the most common cause of death in PD patients. There is a suggestion in the recent literature that PD patients with underlying diagnoses of cardiac failure or ischemic heart disease may be at greater risk on PD, as compared with hemodialysis. [237] [238] Although this finding is controversial and based on observational retrospective data from one country only, it does raise concerns, and there are reasons why PD patients might be predisposed to cardiovascular disease. The main concern relates to the possible adverse effects of the systemic glucose absorption that occurs with PD and how this might exacerbate hyperglycemia, hyperlipidemia, hyperinsulinemia, and the metabolic syndrome generally. These concepts have led to interest in glucose-sparing approaches to PD, such as the use of newer solutions. [107] [108] Aggressive use of lipid-lowering agents has been advocated in PD patients, and it is known that statins are effective and safe.[210] However, there is no trial showing that the beneficial effects of these agents in the general medical population can be extrapolated to those on PD. Indeed, the negative results of a lipid-lowering trial in diabetic hemodialysis patients have raised doubts about this approach.[239]

A favorable feature of PD with regard to cardiac disease is the already discussed better preservation of residual renal function. The association between residual function and cardiac disease has been well described recently. [221] [222] [223] Less clear is the situation with hypertension, in which PD was initially thought to be associated with better control compared with hemodialysis. More recent studies have cast doubt this idea but are always difficult to interpret due to their nonrandomized design. [240] [241] It is frequently noted that, once residual function is lost, control of blood pressure becomes more challenging.

PERITONITIS

Peritonitis remains a leading complication of PD. It contributes to patient morbidity, technique failure, hospitalizations, and occasionally, mortality.[242]

In most major centers, over the past 2 decades, peritonitis rates have fallen from one every 12 months to one every 2 to 4 years. [244] [245] This decrease is mainly due to a fall in cases caused by coagulase-negative staphylococci, coinciding with less risk of touch contamination due to a move away from the use of spike PD giving sets to integrated double-bag systems (see Fig. 59-9 ).[130] The proportion of peritonitis due to gram-negative organisms and methicillin-resistant staphylococci has increased, however. [244] [245]

Diagnosis of Peritonitis

The three cardinal features of peritonitis are cloudy effluent, abdominal pain and a positive effluent culture. Not all cases are typical, but at least two of these three features are required to diagnose peritonitis convincingly.[242]

Nevertheless, a patient with PD presenting with cloudy effluent should be presumed to have peritonitis until proven otherwise. [242] [246] Other causes of turbid dialysate such as fibrin or other proteins, lipids (chylous ascites), and prolonged dwell time should be considered if the effluent white cell (WBC) count is normal and the culture negative. [247] [248]

Abdominal pain may precede cloudy effluent in some cases. The intensity of symptoms varies by etiologic microorganism and with the severity of the infection. Staphylococcus epidermidis and diphtheroids are usually associated with mild pain, whereas with Staphylococcus aureus, Pseudomonas aeroginosa, and fungi, the pain is often more severe. The presence of high fever is not typical of PD peritonitis and suggests systemic sepsis.[249]

The diagnostic approach involves confirmation of the diagnosis and exclusion of peritonitis mimicking conditions, which may require a different management plan.

Peritonitis is confirmed by peritoneal leukocytosis and documentation of the causative microorganism. PD effluent should be sent for total and differential WBC counts, Gram stain, and culture before starting empiric antibiotics for the suspected peritonitis.[242] A culture should be sent from the exit site, if drainage is present. Blood cultures should be obtained only if there are severe systemic symptoms and are seldom positive.[249] Effluent WBC counts should exceed 100 cells/mL (0.1× 109 cells/L), with more than 50% neutrophils. Occasionally, fungal and mycobacterial infections may have initial lymphocytic predominance. In approximately 10% of cases, effluent WBC count may be less than 100 cells/mL at presentation owing to a delayed host response.[242] Cell counts should be repeated if initially low in those with suggestive symptoms.

A shorter dwell time before the PD effluent collection, as can occur in APD patients, may lead to an underestimate of the cell count. In this situation, greater than 50% neutrophils suggests peritonitis, independent of the total WBC counts.[242] In APD patients with peritonitis, day dwell effluent white cell counts are similar to those on CAPD. In those on day-dry APD, if in doubt, an effluent sample can be collected after a 1-L dwell, drained after 1 to 2 hours. A relatively low dialysate WBC count may be a manifestation of tunnel infection, rather than peritonitis.

Gram stain of the peritoneal fluid in peritonitis is often negative. However, the identification of any organism is a helpful guide to therapy, being predictive of the culture results.[242] A Gram stain may be particularly useful in the early diagnosis of fungal peritonitis.

If proper culture technique is followed, the effluent culture should be positive in approximately 80% to 90% of peritonitis cases. A negative culture, despite suggestive symptoms and peritoneal leukocytosis, may be due to fastidious organisms, inadequate samples or collection techniques, laboratory problems, pre-existing antibiotic treatment. or nonbacterial peritonitis.[246] Repeat cultures may be positive. Collecting 5 to 10 mL of effluent dialysate in each of two blood culture bottles and culturing the sediment after centrifuging 50 mL of effluent enhances the recovery rate.[250] One study reported better results with the use of “BacT/Alert” blood culture bottles as compared with standard peptone broth and agar media.[251]

In most series, the majority of cases of peritonitis continue to be due to gram-positive organisms ( Table 59-5 ). [244] [245] About a third of all cases are still accounted for by coagulase-negative staphylococci, which are likely acquired through touch contamination. They cause a relatively mild peritonitis with a low rate of complications. Diphtheroid peritonitis represents about 5% of cases and has a similar pathogenesis, presentation, and outcome.[252]Staphylococcus aureus peritonitis causes 10% to 20% of cases in most series. It is often associated with exit site and tunnel infection and is a more severe infection with a higher probability of hospitalization, catheter loss, and technique failure.[253] Enterococcus is another well-characterized cause of gram-positive peritonitis and responds quite well to treatment. [254] [255]


TABLE 59-5   -- Organisms Causing PD-Associated Peritonitis—Contemporary Data from a Single Center

Gram Positive Organisms

71.2%

 Coagulase-negative staphylococci

39.9%

 Staphylococcus aureus

21.6%

 Streptococcus species

7.9%

 Enterococcus species

1.9%

Gram-Negative Organisms

23.3%

 Escherichia coli

8.6%

 Pseudomonas aeruginosa

4.6%

 Other Pseudomonas species

0.7%

 Acinetobacter species

2.6%

 Klebsiella species

1.9%

 Serratia species

1.3%

 Enterobacter

1.1%

Other Gram-Negative Organisms

5.4%

 Multiorganism

1.9%

 Fungi

3.4%

 Mycobacteria

0.1%

Adapted from Kim DK, Yoo TH, Ryu DR, et al: Changes in causative organisms and their antimicrobial susceptibilities in CAPD peritonitis: a single center's experience over one decade. Perit Dial Int 24:424–432, 2004.

 

 

 

Pseudomonas organisms cause peritonitis in about 10% of cases and can be acquired from the exit site or from enteric sources.[256] It is also associated with exit site and tunnel infection. It is typically difficult to treat with a high rate of relapse and catheter loss. So-called nonpseudomonal gram-negative (NPGN) peritonitis is increasingly important with the decline in coagulase negative staphylococcal cases. [257] [258] It accounts for 20% to 30% of peritonitis. It is acquired either by touch contamination or by migration across the bowel wall and is associated with a rate of relapse and catheter loss similar to that for S. aureus. Typical etiologic organisms are Escherichia coli, klebsiella, serratia, and proteus. The most dreaded peritonitis is fungal, which represents less than 5% of cases in most series and is most often due to Candida albicans. [259] [260] [261] It typically occurs in patients who have received multiple antibiotic courses, often in the context of relapsing bacterial peritonitis. Almost invariably, catheter removal is required and should be done promptly. Rates of technique failure are high, and the mortality risk is significant. Much less common are tuberculous and other mycobacterial peritonitis. [262] [263]

With spreading antibiotic resistance, peritonitis due to entities such as methicillin-resistant S. aureus, vancomycin-resistant enterococcci, and extended-spectrum β-lactamase-producing organisms have become more important. [254] [257] [264] [265] [266]

In gram-negative peritonitis, the concern often arises that the infection is secondary to intra-abdominal pathologies, such as diverticulitis, appendicitis, cholecystitis, or abscess formation. [267] [268] [269] [270] [271] Secondary bacterial peritonitis occurs in less than 6% of cases of peritonitis in CAPD patients.[267] Its recognition is very important, however, owing to its worse prognosis and requirement for different management. Clues to its identification include symptoms related to the disease, fecal matter in the effluent, and cultures positive for multiple enteric organisms or anaerobic bacteria. [242] [267] Outcome is worse, with a 42% death rate in one report. [268] [271] Mortality correlates not only with the disease process causing the peritonitis, but also with the time to diagnosis and definitive surgical intervention.[268] Computed tomographic (CT) scan or radionuclide scan can be used to document underlying causative pathology, but an early exploratory laparotomy is indicated in suspected secondary peritonitis.

It should be noted that the large majority of cases of polymicrobial peritonitis are not secondary to an underlying catastrophe and resolve with either antibiotics or catheter removal. Older patients and those whose cultures show anaerobes are at highest risk of an underlying surgical cause for their presentation.[271]

Cloudy fluid, with a differential increase in cell lines other than neutrophils, suggests alternative etiologies. Eosinophilic peritonitis is defined as greater than 10% eosinophils in the effluent at presentation and has been reported with allergic reactions, exposure to drugs such as vancomycin, fungal and viral infections, and early after catheter placement. [247] [248] [272] [273] Icodextrin-associated sterile peritonitis can occur at any time after exposure to the solution and manifests with mild abdominal pain, no systemic symptoms, cloudy effluent with a predominance of macrophages and eosinophils, and sterile cultures.[274] Icodextrin peritonitis is of particular interest after a significant outbreak in Europe in the early 2000s. This was attributed to peptidoglycan contamination in the manufacturing process.[101]

Management of Peritonitis

The key considerations are the initial choice of empiric antibiotics before identification of the organism, the subsequent choice and duration of treatment based on the culture results, and the critical question of whether and when to remove the catheter in nonresponsive cases. Guidance in these materials has long been provided by the International Society of Peritoneal Dialysis (ISPD) recommendations, which are revised every 3 to 4 years.[242] Usual dosage regimens are shown in Table 59-6 .


TABLE 59-6   -- Intraperitoneal Antibiotic Dosing Recommendations for CAPD Patients[*]

 

Intermittent (Once Daily)

Continuous (mg/L, all exchanges)

Aminoglycosides:

 

 

 Amikacin

2 mg/kg

LD 25, MD 12

 Gentamicin

0.6 mg/kg

LD 8, MD 4

 Netilmicin

0.6 mg/kg

LD 8, MD 4

 Tobramycin

0.6 mg/kg

LD 8, MD 4

Cephalosporins:

 

 

 Cefazolin

15 mg/kg

LD 500, MD 125

 Cefepime

1000 mg

LD 500, MD 125

 Cephalothin

15 mg/kg

LD 500, MD 125

 Ceftazidine

1000–1500 mg

LD 500, MD 125

Penicillins:

 

 

 Ampicillin

no data

MD 125

 Amoxicillin

no data

LD 250–500, MD 50

 Penicillin G

no data

LD 50,000 units, MD 25,000 units

Others:

 

 

 Ciprofloxacin

no data

LD 50, MD 25

 Vancomycin

15–30 mg/kg every 5–7 days

LD 1000, MD 25

 Aztreonam

no data

LD 1000, MD 250

 Amphotericin

not applicable

1.5

 Impenem/Cilistatin

1000 mg bid

LD 500, MD 200

Adapted from Piraino B, Bailie GR, Bernardini J, et al: ISPD guidelines/recommendations. Peritoneal dialysis—related infections recommendations: 2005 update. Perit Dial Int 25:107–131, 2005.

*

In patients with >100 mL/day urine output dose should be empirically increased by 25%. CAPD, continuous ambulatory peritoneal dialysis; LD, loading dose; MD, maintenance dose.

 

 

Initial Empiric Therapy

This must account for all the common etiologies of peritonitis. The 2005 ISPD guidelines recommended that gram-positive organisms be covered by either a cephalosporin or vancomycin and gram-negative organisms by a third-generation cephalosporin or an aminoglycoside.[242]

In some centers, with a high rate of methicillin-resistant organisms, vancomycin is preferred for gram-positive coverage, but this has raised concerns about promoting colonization with vancomycin-resistant enterococci and other centers use a first-generation cephalosporin (such as cephalothin or cefazolin). For gram-negative coverage, the guidelines recommend an aminoglycoside, ceftazidime, cefepime, or carbapenem.[242] Aminoglycosides are effective and inexpensive, but there are concerns about nephrotoxicity in patients with residual renal function, although this may have been overstated and short duration, once-daily courses appear safe.[275] Quinolones can also be used for empiric coverage of gram-negative organisms, if local sensitivities permit.[276] For cephalosporin-allergic patients, aztreonam is an alternative to ceftazidime or cefepime for gram-negative coverage, if aminoglycosides are not used.

Other antibiotic regimens have been proposed. In one study, initial treatment with imipenem/cilastatin had a similar peritonitis cure rate compared with a cefazolin plus ceftazidime regime.[277] Cefepime alone was as effective as vancomycin plus netilimicin in another randomized trial.[278] Quinolones given orally are acceptable alternatives for gram-negative coverage[276] and do reach adequate levels within the peritoneum even with APD.[279] In another study, oral ofloxacin alone (400 mg, followed by 300 mg daily) was as effective as intraperitoneal cephalothin with tobramycin in CAPD patients.[280] However, ciprofloxacin alone is not an ideal choice for S. aureus peritonitis, owing to reports of slow resolution with this approach. Oral cephalosporin therapy is an alternative in mild peritonitis with a relatively asymptomatic patient but should not be considered in severe cases.[281]

Intermittent or Continuous Intraperitoneal Dosing

In patients on CAPD, intraperitoneal dosing of antibiotics is preferred to intravenous dosing, given the increased local concentration with intraperitoneal administration.[242]

Intraperitoneal administration can be either continuous (with antibiotics given in each exchange) or intermittent (given once daily). With intermittent dosing, the antibiotic must dwell for at least 6 hours. There is no conclusive evidence as to which is more effective but intermittent is more convenient and widely practiced for cephalosporins, aminoglycosides, and of course, vancomycin. In the case of the latter two, antibiotic levels help to ensure efficacy and avoid toxicity. [242] [275]

On APD, given this lack of data, recommendations are that it is safe to add first-generation cephalosporins to each exchange. [242] [282] [283] Vancomycin can be administered intermittently with a dosing interval every 4 to 5 days, but the optimal interval should be determined by monitoring levels. [282] [283]

Subsequent Antibiotic Management

The initial empiric regimen can be adjusted once the culture results are known and a specific organism identified. The ongoing response should be monitored both clinically and by repeated cell counts and effluent cultures.

Coagulase-Negative Staphylococcus

First-generation cephalosporins are preferable for 2 weeks of therapy. Units with a high rate of methicillin resistance, may prefer to use vancomycin. Most peritonitis episodes due to a coagulase-negative staphylococcus respond rapidly to therapy.

Streptococcus and Enterococcus

Streptococcal and enterococcal peritonitis should be treated with intraperitoneal ampicillin. An aminoglycoside, given intraperitoneally once daily, may be added for synergy in enterococcal cases. For vancomycin-resistant enterococcus peritonitis, linezolid or quinupristin/dalfopristin is proposed, but these agents can be toxic. [284] [285]

Staphylococcus Aureus

If the organism is methicillin sensitive, cephalosporin can be continued, and if it is methicillin resistant, vancomycin is indicated. Three weeks of treatment are required, and if needed, rifampin, 600 mg/day orally, can be added, but only for 1 week to avoid resistance and not in regions where tuberculosis is endemic. Teicoplanin can be used as an alternative to vancomycin. In vancomycin-resistant S. aureus is found, linezolid, daptomycin, or quinupristin/dalfopristin are the antibiotics of choice.[284]

If there is concomitant exit site infection, due to the same organism, or tunnel infection, S. aureus peritonitis is unlikely to be cured without removal of the catheter. [242] [253]

Pseudomonas Aeruginosa

P. aeruginosa peritonitis requires administration of at least two antibiotics with differing mechanisms of action.[242] One antibiotic can be an oral quinolone, with the other being intraperitoneal ceftazidime, cefepime, tobramycin, or intravenous piperacillin.[256]

With this peritonitis, a catheter-related infection is frequently present and so catheter removal is required, with appropriate antibiotics administered for 2 weeks, while the patient is on hemodialysis. [242] [256] A new catheter can be placed after a period of time. With no catheter infection, antibiotics alone for 3 weeks can be tried. A similar approach is taken for Stenotrophomonas maltophilia peritonitis.

Nonpseudomonal Gram-Negative Peritonitis

Here, the recommendation is a single antibiotic based on sensitivity, safety, and convenience and given for 2 to 3 weeks.[242] Catheter removal is often required because the organisms embed in catheter biofilm. Recent data focusing on the poor outcome with this peritonitis support routinely adding a second antibiotic based on culture sensitivities.[257]

Polymicrobial Peritonitis

With multiple enteric organisms, intra-abdominal pathologies such as ischemic bowel or diverticular disease should be considered, and a surgical consultation should be obtained and laparotomy performed if response to treatment is not rapid. [267] [268] [269] [270] [271] The optimal therapy is metronidazole, in combination with ampicillin and ceftazidime or an aminoglycoside. [242] [271] With multiple gram-positive organisms, contamination or catheter infection is most likely. Appropriate antibiotics should be administered, and peritonitis usually resolves without catheter removal, unless the catheter is infected.[271]

Culture-Negative Peritonitis

With culture-negative peritonitis, repeat cell and differential counts, along with culture, should be obtained after day three. If the repeat cell count points toward an unresolved infection, special culture techniques should be used to isolate unusual causes of peritonitis, such as fungi and mycobacteria and catheter removal should be considered. [246] [247] [248] If there is prompt clinical improvement with empiric therapy, the gram-negative cover should be stopped and the gram-positive cover be continued for 2 weeks.

Fungal Peritonitis

Fungal peritonitis is difficult to eradicate and generally requires early catheter removal. [242] [259] [260] [261] Prolonged treatment with antifungal agents to attempt clearance is not encouraged. Fungal peritonitis is serious, leading to death in approximately 25% of episodes. Evidence suggests that prompt catheter removal poses a lesser risk of death.

Initial therapy may be a combination of amphotericin B and flucytosine until the culture results are available with susceptibilities.[242] Caspofungin, fluconazole, or voriconazole may replace amphotericin B, based on culture and sensitivity.[260] Intraperitoneal amphotericin causes chemical peritonitis and pain, but intravenous use leads to poor peritoneal penetration. Voriconazole is an alternative when filamentous fungi have been cultured and can be used for candida peritonitis.[260] If flucytosine is used, regular monitoring of serum concentrations is necessary to avoid bone marrow toxicity. Emergence of resistance to the imidazoles has occurred, thus indicating the importance of sensitivities, where available. Therapy should be continued after catheter removal, typically with flucytosine 1000 mg and fluconazole 100 to 200 mg daily by mouth for an additional 10 days. [242] [260]

Mycobacterial Peritonitis

This should be treated with rifampin, isoniazid, pyrazinamide, and ofloxacin. [262] [263] Given that rifampin penetration into dialysis fluid may be poor, this agent may need to be given intraperitoneally. Pyrazinamide and ofloxacin should be withdrawn after 3 months, whereas rifampin and isoniazid should be continued for 12 months. Pyridoxine should be provided. Catheter removal is generally required for atypical mycobacteria but not always for tuberculous cases. [262] [263]

Indications for Catheter Removal in Peritonitis

This is the most critical decision in the management of peritonitis. If delayed too long, the patient may become unnecessarily ill and the cavity may be damaged permanently by adhesions; however, unneccessary removal is also undesirable for the patient. It is recommended that the catheter should be removed in following conditions[242]:

  

   

Relapsing peritonitis—This is defined as an episode with the same organism that caused the preceding episode of peritonitis or one sterile episode within 4 weeks after completing the course of antibiotics.

  

   

Refractory peritonitis—This is defined as the failure to respond within 5 days of appropriate antibiotics.

  

   

Refractory catheter infections (exit site and tunnel infections).

  

   

Fungal peritonitis.

  

   

Fecal peritonitis or other cause of significant intra-abdominal pathology.

  

   

Peritonitis occurring in association with intra-abdominal pathology.

Simultaneous catheter removal and new catheter replacement can be performed for refractory exit site and tunnel infection and in relapsing peritonitis cases provided there has been temporary clearance of florid peritoneal infection.[286] [287] [288] [289] It should not be performed in nonresponding peritonitis. [242] [289] In this setting, a minimum period of 2 to 3 weeks is recommended between the time of catheter removal and placement of a new catheter.

Prevention of Peritonitis

Key points in prevention of peritonitis include the use of integrated double-bag systems in CAPD and in ensuring a flush before fill step after connection of the tubing to the solution bags in APD. [130] [290] [291] Good patient training practices are also important with home visits and re-education after episodes of peritonitis and after any major illness. At a unit level, regular review of peritonitis and exit site infection rates and of unit practices also help.

A major proportion of the more difficult peritonitis infections occur as a complication of infection of the exit site, tunnel, or catheter. Therefore, prevention and treatment of exit site infection is important to reduce peritonitis rates.[242] Mupirocin applied either to the nares or directly to the exit site is a widely used strategy to prevent staphylococcus aureus exit site infection, which occurs almost exclusively in those who have nasal carriage of the organism.[292] [293] [294] A large randomized controlled trial showed that intranasal application 5 days a month is effective in reducing S. aureus exit site infection rates, and other studies suggest that daily local exit site application is equally effective.[292] More recently, Bernardini and co-workers[295] have compared daily exit site application of gentamicin versus mupirocin and found a significant reduction in both pseudomonas and nonpseudomonas gram-negative exit site infection and peritonitis rates. This encouraging finding needs to be confirmed but is a big step forward in that nonpseudomonas infections had been previously considered to be mainly related to transmural migration of organisms from the bowel and so not to be easily amenable to prophylaxis.

NONINFECTIOUS COMPLICATIONS

Mechanical

Mechanical complications of PD are relatively common, as a consequence of the raised pressure inside the peritoneal cavity.[296] Most frequent is the problem of hernias, which occur in 10% to 20% of patients during their time on PD. [296] [297] Most common are indirect inguinal hernias related to a patent processus vaginalis through which the diaysate migrates to cause genital swelling. Also common are umbilical and abdominal incisional hernias. If hernias are painful or large enough to cause discomfort, they should be surgically repaired. With asymptomatic hernias, a judgment needs to be made as to whether the risks of bowel incarceration are sufficient to justify the inconvenience and associated risks of surgery. Frequently, for older sicker patients with limited life expectancy, a conservative approach will be preferable, as long as the patient is aware of the need to seek urgent medical attention in the event of acute pain occurring. Repair does not require a routine switch to hemodialysis, and patients can be managed with a 24- to 48-hour break from PD, followed by 2- to 3-week period of day dry APD or low-volume CAPD.[298]

Pericatheter leaks are also common, especially in the early weeks after initiation of PD. [297] [298] [299] [300] They usually occur around the exit site and present as frank dialysate leakage or as abdominal wall swelling or as dependent genital edema. Again, the management is to either discontinue PD for a number of weeks or to switch to day dry APD, and these options are often feasible because the leaks usually occur at a time when residual renal function is still good. Many leaks do not recur when PD is resumed, but some do and catheter reinsertion may eventually be required.[300] Obese patients and those on steroids are more likely to have leaks.[299] Prolongation of the period between catheter insertion and initial use to 4 weeks or longer may reduce the risk of leaks. [296] [299]

Diaphragmatic leaks are uncommon but associated with a classic presentation of pleural effusion, almost always right sided, occurring in the first few weeks of PD and made worse by use of more hypertonic fluids.[301] Initial treatment is to drain the abdomen and either hold PD or switch to hemodialysis. Only a minority of these heal with rest because they are likely related to congenital defects in the diaphragm. Most require surgical or chemical pleurodesis if PD is to be successfully resumed. [301] [302]

All of these hernias and leaks can be imaged by instilling contrast dye or radioactive isotope into the PD solution before infusion and then by performing CT or magnetic resonance imaging or nuclear scanning. [296] [303] [304] This is most useful with genital swelling when it is often unclear whether the problem is a patent processus vaginalis or a pericatheter leak.[305]

Other mechanical complications include back pain, gastric reflux, and a sensation of abdominal fullness or of dyspnea.[296] Management typically involves temporary or permanent reduction of dwell volumes, especially in the daytime, when the patient is upright and intraperitoneal pressure is higher.

Encapsulating Peritoneal Sclerosis

Encapsulating peritoneal sclerosis (EPS), the most feared complication of PD, is an uncommon but dramatic condition in which gross sclerosis of the peritoneal membrane occurs to a degree that causes encapsulation of the intestines with consequent obstructive ileus, leading to anorexia and weight loss. [306] [307] There is typically hemorrhagic ascites, and there are often features of systemic inflammation including anemia, high serum C-reactive protein, low serum albumin. and low-grade fever.[307] Radiologic features include calcification of the peritoneum with formation of an encapsulating fibrotic cocoon around the bowels, loculated ascites, adherent and dilated bowel loops, and air-fluid levels.[308] The mortality rate is in the range of 30% to 50%. [306] [307] [309]

The cause of this condition is unknown, but it is clear that its incidence rises with time on PD. [307] [309] [310] It is very rare in the first 2 years and uncommon before 5 years but the cumulative incidence in one Australian study reached 10% at 6 years and almost 20% in patients on PD for more than 8 years.[309] A Japanese study suggests a rate of 6% by 10 years and of almost 20% after more than 15 years.[311] The condition often presents only after PD is discontinued and the patient is on hemodialysis. It may begin as recurrent ascites, progressing to bowel obstruction.

The cause is unknown. A similar condition labeled idiopathic occurs in patients without renal failure and other cases occur in association with use of β-blockers, autoimmune disease, and malignancy. [306] [309] The PD cases may follow severe peritonitis episodes but some have occurred in PD patients who never had peritonitis. A role for bioincompatible PD solutions is often invoked, with blame being attributed to the glucose, the GDPs, or the buffer but without any conclusive evidence. A large proportion of reported cases come from Japan, where there has been immense interest in, and preoccupation with, the condition. [306] [307] [309] In contrast, EPS appears to be uncommon in North America. This has led to speculation about genetic or ethnic predisposition, but all this may simply reflect the greater longevity of both patient and technique survival in Japan relative to North America. Some of the international differences in EPS rates may reflect differences in awareness of the condition and in thresholds for diagnosing it. In this regard, it should be stated clearly that the presence of peritoneal fibrosis or sclerosis or even ascites is insufficient to make a diagnosis of EPS. Evidence on imaging of encapsulation is required. Laparoscopy and biopsy to confirm the diagnosis are helpful but are certainly not necessary. [306] [307]

Concerns about EPS in Japan, particularly, have led to interest in identification of early stages of the condition. Suggested markers such as falling ultrafiltration, rising D/P Cr and decreasing effluent levels of CA210 are all common and nonspecific for EPS.[309]

Recently, there have been concerns that the incidence of EPS may be rising in European centers, and there are renewed calls for the sort of multicenter research initiative needed to accumulate enough cases to investigate risk factors and genetic predisposition. [312] [313]

Treatment in the past was generally supportive and largely ineffective. [306] [307] Surgical interventions to try to free up bowel and lyse adhesions were notably unsuccessful. In the last 15 years, more success has been reported, however. Steroid use has become common and appears to help, although there are no randomized trials.[314] Other immunosuppressives have also been administered, and recently, tamoxifen has been used based on its efficacy in other fibrosing diseases.[315] Surgical techniques have also improved, and Kawanishi and colleagues[316] from Japan have reported much better results with enterolysis. Total parenteral nutrition is often required.[314]

Some centers, both in Japan and elsewhere, have been sufficiently concerned to advocate switching all patients to hemodialysis once they have been on PD for more than 5 or 7 years. Others believe that this is not justified, and some wonder if the switch is itself more likely to precipitate EPS in that many cases first presenting within 6 months of discontinuation of PD.

Metabolic Complications

These mainly relate to the metabolic effects of the systemic glucose absorption that occurs in conventional PD.

Hyperlipidemia in PD is often quite marked and com-prises both hypertriglyceridemia and high low-density lipoprotein cholesterol, related, in turn, to high apolipoprotein B levels. [317] [318] [319] It is presumed that glucose absorption is the main cause of these abnormalities, although dialysate protein losses may also contribute. The lipid profile is quite atherogenic and may contribute to the perceived higher risk of mortality in patients with heart disease on PD noted in some studies. The tendency is to treat with statins, primarily to reduce the low density lipoprotein cholesterol, or with fibrates if the high triglycerides is the more marked abnormality. [210] [319] There is reluctance to use both types of agent together for fear of myositis or other side effects. There is no proof that lipid-lowering agents improve outcomes or prolong life in PD, and one study in hemodialysis was actually negative.[239]

Induction or exacerbation of hyperglycemia resulting from the glucose absorption is common in PD because many patients, if not frankly diabetic, have impaired glucose tolerance. Some previously nondiabetic patients with PD develop hyperglycemia, sufficient to require oral hypoglycemic agents or insulin. Metformin is contraindicated in renal failure, but sulphonylureas and thiazolidinediones can be used.[320] Intraperitoneal insulin was widely used in the early days of CAPD because it avoided the need for insulin injections and appeared to be more physiologic. In recent times, however, it has become less popular due to the risks of contamination during injection of bags.

Glucose absorption contributes about 500 to 800 kilocalories per day to a PD patient's caloric intake, a significant proportion of the daily recommended 3500 kilocalories.[321] A weight gain of 10% is common in the first 6 months of PD and may be welcome in some malnourished patients.[322] However, in others, it causes or, more often, exacerbates obesity, and this may decrease mobility and increase cardiovascular risk. Efforts to prevent obesity include dietary and exercise advice and glucose-sparing approaches including consideration of the use of nonglucose solutions specifically to avoid progressive weigh gain.[322]

Hypoalbuminemia is common in PD patients with an average value of 32 to 34 g/L. It is predominantly related to two factors, dialysate protein losses and systemic inflammation. [47] [179] Dialysate protein losses are mainly albumin and are typically 6 to 8 g/day. They are greater in high transporters and in patients with peritonitis. High serum C-reactive protein levels are also associated with low serum albumin, related to the altered pattern of protein synthesis seen in inflammation. Dietary protein intake has, contrary to common opinion, little influence on serum albumin. Just as in other patient populations, hypoalbuminemia is predictive of worse clinical outcomes in PD but there is no proven intervention to correct it. [47] [48]

Calcium/phosphate abnormalities are invariable in ESRD and are reviewed elsewhere. [323] [324] However, in PD, there are some specific features. Phosphate control tends to be better than in hemodialysis, because of the continuous nature of the dialysis and perhaps because of a lower dietary protein intake.[323] Hypercalcemia is also frequent and is mainly related to calcium carbonate and vitamin D intake with a contribution from adynamic bone disease in some cases. The latter entity is particularly well recognized in PD patients because of more consistent suppression of parathyroid hormone secretion. In PD, there is often a need therefore to stop vitamin D analogs and to switch to non-calcium-containing phosphate binders. [323] [324] [325]

Potassium disorders are are also frequent in patients receiving PD. Hyperkalemia is less common than hypokalemia and is usually related to dietary excess and ACE inhibitor therapy.[326] It is usually self-limiting and so not an emergency as long as the patient is compliant with the dialysis prescription. Hypokalemia usually reflects poor dietary intake although it can be exacerbated by high-dose loop diuretic therapy.[327] Dietary counseling is important, and potassium supplements may be needed. Hypokalemia is an adverse prognostic factor for patient survival.

Serum sodium in PD tends to be a little lower than in normal individuals, owing to fluid intake against a background of renal failure and due to the dialysate sodium of 132 mmol/L. Marked hyponatremia usually reflects hyperglycemia or very high water intake.[328] Icodextrin causes a modest, clinically unimportant fall in serum sodium related to serum levels of maltose and other metabolites. Hypernatremia is rare but can occur in the context of rapid cycling when dwell times are short and the dominant factor is removal of a low sodium ultrafiltrate due to sodium sieving with consequent rise in the serum sodium, especially if the patient does not sense thirst or have access to water.[329]

Acute Peritoneal Dialysis

The use of PD to treat acute renal failure is much less common than was the case 2 decades ago. The widespread availability of conventional hemodialysis and, more recently, of continuous hemofiltration and hemodiafiltration, combined with better strategies to avoid bleeding and maintain hemodynamic stability have led to a decrease in emergent PD use. Nevertheless, the modality, especially in pediatrics and in adults in developing countries, is still regularly used. [330] [331] [332]

Peritoneal access is either by an acute PD stylet or, increasingly, via a Tenchkoff catheter as the latter is more reliable and durable. Most commonly, acute PD is delivered as exchanges every 1 to 2 hours, done either manually or with a cycler. In less catabolic patients, a regimen known as continuous equilibration PD may be used and involves providing 4 to 8 dwells daily, each of 3 to 6 hours duration.[333] The advantage of acute PD is its simplicity, the lack of a need for anticoagulation, and the hemodynamic stability. The disadvantage is the risk of leaks, especially with an acute stylet, and of peritonitis, which is relatively common in this setting. Other problems that often arise are catheter drainage problems requiring intraperitoneal heparin, hypokalemia requiring intraperitoneal potassium supplementation, and hyperglycemia requiring insulin. Another concern is that the clearance provided by acute PD may not be as good as that possible with continuous hemodialysis type techniques and this may become an issue in a very catabolic patient. A recent trial comparing acute PD and hemodialysis in a Vietnamese population in which malaria was the main cause of acute renal failure showed inferior outcomes with PD but it is unclear how much these results can be extrapolated to other settings.[334]

COST-EFFECTIVENESS OF PERITONEAL DIALYSIS

Patient Outcomes with Peritoneal Dialysis

Since the development of CAPD into a major renal replacement therapy in the early 1980s, patient outcomes have inevitably been compared with those on hemodialysis. It is well recognized that technique failure is more common on PD than on hemodialysis. Typically, once transplantation and death are censored out, about 15% of prevalent PD patients switch to hemodialysis annually, most often due to recurrent or refractory peritonitis, accumulating comorbidity, or failure to thrive.[335] Switches in the other direction are less frequent and typically are cases in which modality selection was delayed until after initiation of dialysis; a small number are related to cardiovascular intolerance of hemodialysis or to vascular access failures. This is well understood, and any large PD program has to have easy access to back up hemodialysis. What is less clear and what has caused long-standing controversy is how patient survival compares between the two modalities.

Ideally, a randomized controlled trial would be carried out to answer the question of whether patients do better initiating dialysis on PD versus hemodialysis. However, such trials have not been feasible to complete due mainly to unwillingness of patients to be randomized.[336] Therefore, most comparative analyses have been retrospective studies of registry-type databases whereas a small number have been prospective cohort studies. Registry-based studies have provided conflicting results, and some of this has been related to differences in methodology. A few general comments should be kept in mind when assessing this literature. First, PD patients have better outcomes relative to their hemodialysis counterparts during the first 1 to 3 years on PD as compared with subsequent time periods ( Fig. 59-10 ).[336] The mechanism of this may be better preservation of residual function, or it may partly reflect differences in patient mix that persist even after adjustments for demographics and comorbidity are made. Regardless, it means that studies based on incident patients will show better outcomes for PD than those based on prevalent patients. Of course, incident studies are more appropriate for comparing outcomes, anyway. Second, PD appears to be relatively better in younger, nondiabetic patients than in older, diabetic ones. [336] [337] So, there may not be one overall answer to the question of which modality is best and by how much, and therefore, results are best broken down by age and diabetic status. Third, intent to treat and as treated analyses should be clearly distinguished.[336] The former are more relevant to answering the key question as to whether the initial modality allocation affects outcome. The latter, however, may help to answer the different question of whether patients are more likely to die on one modality than the other. Fourth, the approach to dealing with modality switches differs between studies and is complex because many more patients switch from PD to hemodialysis as compared with vice versa.[335] Generally, some period of grace is allowed so that outcomes occurring shortly after a switch are allocated to the previous rather than the new modality. Finally, all such studies need to try and correct for baseline demographics and comorbidity although this is always difficult and always incomplete. However, they must not attempt to correct for differences that arise after initiation of dialysis because these may be inherently related to the modality. Examples are residual renal function and clearance, both of which are influenced by the modality used.

FIGURE 59-10  Adjusted mortality rate ratios for peritoneal dialysis relative to hemodialysis with time after initiation of dialysis.  (From Schaubel DE, Morrisson HI, Fenton SS: Comparing mortality rates on CAPD/CCPD and hemodialysis. The Canadian experience: fact or fiction? Perit Dial Int 18:478–484, 1998, with permission).

 

 

Initial comparative studies did not suggest a major difference in outcomes between PD and hemodialysis. However, in 1995, Bloembergen and co-workers[338] reported a 17% higher mortality in prevalent U.S. PD patients compared with those on hemodialysis. The methodology used in the study was unusual and Vonesh and associates[339] redid the analysis with some changes, used more contemporary cohorts, and found no significant difference in outcomes. Collins and associates[337] looked at incident U.S. patients over the first 2 years of dialysis and found a survival advantage for PD in younger and in nondiabetic patients and a small advantage for hemodialysis in older diabetics. More recently, Stack and co-workers[238] and Ganesh and colleagues[237] found that in U.S. patients with heart failure and ischemic heart disease, there was an advantage for hemodialysis.

In Canada, analyses done by different methods have shown a consistent survival advantage for PD, most marked with an intent to treat approach, in younger patients and in the first 2 years of dialysis.[336] Heaf and co-workers[340]have reported a similar advantage for PD in Denmark.

Two major prospective cohort studies have been done comparing PD and hemodialysis. [341] [342] The first was conducted in Canada in the 1990s, and it focused on more than 800 incident patients in 11 large centers, followed for 2 years. Approximately half were on each modality. The analysis showed a survival advantage for PD that disappeared after correction for baseline comorbidity.[341]

The second cohort study is Choice, which was performed in the United States, published in 2004, and had more than 1000 patients at 81 centers followed for a mean of 2.4 years. However, much fewer were on PD and a somewhat contrived recruitment plan had to be introduced to provide sufficient PD patients to make the comparison valid.[342] The Choice study found no difference in unadjusted survival, but after adjustment, there was a significant survival advantage for hemodialysis. This was most apparent after the first year and only became marked after adjustment for laboratory values, not all of which were measured at baseline. The Choice study also used propensity scores in its analysis and found that survival was equivalent between the modalities in the tercile of patients most likely to do PD.[342]

In summary, there is no conclusive evidence that one modality is superior to the other in survival. However, there is some concern that PD is associated with higher risk of mortality in older PD patients with diabetes or overt heart disease and this emphasizes the need to address the potentially adverse effects of glucose absorption on the cardiovascular risk profile. The apparent survival benefit for the different modalities in different age groups is neither consistent nor large enough to justify using the data to make blanket decisions concerning modality selection. It would no more be justified to put every older diabetic on hemodialysis than it would be to put every young nondiabetic on PD. Factors such as social circumstances and patient choice should continue to be given more weight in these decisions.

Economics of Peritoneal Dialyisis

The economics of dialysis generally is a complex area, and only brief comments relevant to PD are made here. In the developed world, PD is generally less expensive to deliver than hemodialysis, and this makes the modality attractive to those funding renal replacement therapy ( Table 59-7 ). [343] [344] The lower cost is primarily related to the fact that in PD, the patient or family member administers the dialysis, whereas in hemodialysis, it is performed by a relatively expensive trained staff. Also, there are much greater capital costs in setting up a hemdialysis as compared with a PD unit. PD does require expensive sterile solutions and disposable tubing, but the end result is that the total cost of delivering PD is 50% to 70% of the cost of delivering hemodialysis.


TABLE 59-7   -- Annual Costs of Hemodialysis versus Peritoneal Dialysis in Published Studies

 

 

HD

PD

Cost Ratio

Bruns et al[344]

United States

U.S. $68,891

U.S. $45,420

1.52

Goeree et al[343]

Canada

CAN $88,585

CAN $44,790

1.98

 

HD, hemodialysis; PD, peritoneal dialysis.

 

 

 

In countries such as Canada, Australia, United Kingdom, and Hong Kong where dialysis is funded and delivered by public institutions, PD use is encouraged and is relatively high, ranging from about 20% in Canada to almost 80% in Hong Kong. Countries in which dialysis is mainly provided by private units tend to have much lower utilization rates, varying from under 5% in Japan to about 8% in the United States and Germany.[345] The question that arises is if PD is less costly, why are private providers so ineffective in using it. In some settings, the answer is that although PD costs less to deliver, hemodialysis is better reimbursed and so more profitable. This is particularly so when parenteral medications such as vitamin D, iron, and eryrthropoietins, which are easily administered on hemodialysis but not on PD, are billable items and so potential profit sources. More recently, funding agencies in many jurisdictions have moved to reduce these incentives to use more hemodialysis but PD use has not grown greatly in response. Private providers may have underused hemodialysis capacity, and in this setting, the marginal cost of treating the next patient with hemodialysis may be less than that of using PD. This may be particularly so if the PD program is small as the low cost of PD may not be apparent until 20 or more patients are treated. Low PD use in private provider settings may also reflect the relative ease of setting up hemodialysis clinics and the greater complexity of providing the infrastructure to do PD effectively.[346]

In the developing world, the cost advantage for PD is often less or absent, and in some countries, PD costs more than hemodialysis. This is because staff and capital costs are relatively less in such countries, whereas the costs of importing sterile PD solutions are relatively high. Local production of PD solutions may decrease the cost difference but these issues limit PD growth in developing countries.

Therefore, there is a paradox whereby PD growth is constrained in wealthier countries because PD, although it is cheaper, it is less profitable or less convenient; at the same time, PD is constrained in poorer countries because PD is more expensive.

Of course noneconomic factors also affect PD use, but it is striking how much of the international variation can be explained by whether providers are public or private and by national wealth.[345]

References

1. Ganter G: Uber die Beseitigung giftiger stofe aus dem Blute durch Dialyse.  Munch Med Wochensr  1923; 70:1478-1483.

2. McBride P, Russell A: Palmer—the man behind today's permanent peritoneal catheter.  Perit Dial Bull  1981; 1:159-161.

3. Fine J, Frank H, Seligman AM: The treatment of acute renal failure by peritoneal irrigation.  Ann Surg;  1946; 124:857-863.

4. Derot M, Tanzet P, Rousillillon J, Bernier JJ: La dialyse peritoneale dans le traitement de l'uremie aigue.  J Urol  1949; 55:113-119.

5. Odel HM, Ferris DO, Power MH: Peritoneal lavage as an effective means of extra renal excretion.  Am J Med  1950; 9:63-77.

6. Boen ST, Mulinari AS, Dillard DH, Scribner BH: Periodic peritoneal dialysis in the management of chronic uremia.  Trans Am Soc Artif Intern Org  1962; 8:256-263.

7. Tenckhoff H, Shilipetar G, Boen ST: One year's experience with home peritoneal dialysis.  Trans Am Soc Artif Intern Org  1965; 11:11-17.

8. McDonald HP: An automatic peritoneal dialysis machine; preliminary report.  Trans Am Soc Artif Intern Org  1965; 11:83-85.

9. Popovich RP, Moncrief JW, Decherd JF: The definition of a novel portable-wearable equilibrium peritoneal technique.  Abst Am Soc Artif Intern Org  1977; 5:64.

10. Popovich RP, Moncrief JW, Nolph KD, et al: Continuous ambulatory peritoneal dialysis.  Ann Intern Med  1978; 88:449-456.

11. Oreopouolos DG, Robson M, Izatt S: A simple and safe technique for continuous ambulatory peritoneal dialysis.  Trans Am Soc Artif Intern Org  1978; 24:484-489.

12. Moncrief JW, Popovich RP, Nolph KD: The history and current status of continuous ambulatory peritoneal dialysis.  Am J Kidney Dis  1990; 16:579-584.

13. Blake PG, Finkelstein F: Why is the proportion of patients doing peritoneal dialysis declining in North America?.  Perit Dial Int  2001; 21:107-114.

14. Moeller S, Gioberge S, Brown G: ESRD patients in 2001: Global overview of patients, treatment modalities and development trends.  Nephrol Dial Transplant  2002; 17:2071-2076.

15. Pawlaczyk K, Kuzlan M, Wieczorowska-Tobis K, et al: Species-dependent topography of the peritoneum.  Adv Perit Dial  1996; 12:3-6.

16. Nagy JA: Peritoneal membrane morphology and function.  Kidney Int  1996; 50(suppl 56):S2-S11.

17. Flessner M, Henegar J, Bigler S, Genous L: Is the peritoneum a significant transport barrier in peritoneal dialysis?.  Perit Dial Int  2003; 23:542-549.

18. Flessner J: Peritoneal transport physiology: insights from basic research.  J Am Soc Nephrol  1991; 2:122-135.

19. Bouchet Y, Voilin C, Yver R: The peritoneum and its anatomy.   In: Bengmark S, ed. The peritoneum and peritoneal access,  London: Wright; 1989:1-13.

20. Flessner MF: The role of extracellular matrix in transperitoneal transport of water and solutes.  Perit Dial Int  2001; 21(suppl 3):S24-S29.

21. Khanna R, Matcier RA: Role of lymphatics in peritoneal dialysis.  Blood Purif  1992; 10:163-172.

22. Nolph KD: Peritoneal dialysis.   In: Brenner BM, Rector FC, ed. The Kidney,  3rd ed. Philadelphia: WB Saunders; 1986:1847-1906.

23. Rippe B: A three pore model of peritoneal transport.  Perit Dial Int  1999; 13(suppl 2):S35-S38.

24. Agre P, Preston GM, Smith EL: Aquaporin CHIP: The archetypal molecular water channel.  Am J Physiol  1993; 265:F463-F476.

25. La Milia V, Di Filippo S, Crepaldi M, et al: Mini-peritoneal equilibration test: A simple and fast method to assess free water and small solute transport across the peritoneal membrane.  Kidney Int  2005; 68:840-846.

26. Smit W, Struijk DJ, Ho-Dac-Pannekeet MM, Krediet RT: Quantification of free water transport in peritoneal dialysis.  Kidney Int  2004; 66:849-854.

27. Ronco C: The nearest capillary hypothesis: a novel approach to peritoneal transport physiology.  Perit Dial Int  1996; 16:121-125.

28. Rosengren BI, Rippe B: Blood flow limitation in vivo of small solute transfer during peritoneal dialysis in rats.  J Am Soc Nephrol  2003; 14:1599-1604.

29. Flessner MF, Lofthouse J: Blood flow does not limit peritoneal transport.  Perit Dial Int  1999; 19(suppl 2):S102-S105.

30. Hirszel P, Maher JF: Pharmacological alteration of peritoneal transport rates.   In: Nolph KD, ed. Peritoneal Dialysis,  Dordrecht: Kluwer Academic Publishers; 1989:184-198.

31. Krediet RT, Zuyderhoudt MJ, Boeschoeten EW, Arisz L: Alterations in the transport of water and solutes during peritonitis in CAPD patients.  Eur J Clin Invest  1987; 17:43-52.

32. Heimburger O, Waniewski J, Werynski A, Lindholm B: A quantitative description of solute and fluid transport during peritoneal dialysis.  Kidney Int  1992; 41:1320-1332.

33. Waniewski J, Heimburger O, Werynski A, Lindholm B: Aqueous solute concentrations and evaluation of mass transfer area coefficients in peritoneal dialysis.  Nephrol Dial Transplant  1992; 7:50-56.

34. Keshaviah P: Relationship between body size, fill volume and mass transfer area coefficient in peritoneal dialysis.  J Am Soc Nephrol  1994; 4:1820-1826.

35. Mactier RA, Twardowski ZJ: Influence of dwell time, osmolarity and volume of exchange on solute mass transfer and ultafiltration in peritoneal dialysis.  Semin Dial  1988; 1:40-49.

36. Rippe B, Perry MA, Granger DN: Permselectivity of the peritoneal membrane.  Microvasc Res  1985; 29:89-102.

37. Krediet RT, Lindholm B, Rippe B: Pathophysiology of peritoneal membrane failure.  Perit Dial Int  2000; 20(suppl 4):S22-S42.

38. Mistry CD, Gokal R: The use of glucose polymer (icodextrin) in peritoneal dialysis: An overview.  Perit Dial Int  1994; 14(suppl 3):S158-S161.

39. Imholz AL, Koomen GC, Struijk DG, et al: Effect of an increased intraperitoneal pressure on fluid and solute transport during CAPD.  Kidney Int  1993; 44:1078-1085.

40. Krediet RT: The effective lymphatic absorption rate is an accurate and useful concept in the physiology of peritoneal dialysis.  Perit Dial  2005; 24:309-313.

41. Flessner M: Effective lymphatic absorption rate is not a useful or accurate term to use in the physiology of peritoneal dialysis.  Perit Dial  2005; 24:314-316.

42. Stachowska-Pietka J, Waniewski J, Flessner MF, Lindholm B: Distributed model of peritoneal fluid absorption.  Am J Physiol Heart Circ Physiol  2006; 291:H1862-H1874.

43. Twardowski ZJ, Nolph KD, Khanna R, et al: Peritoneal Equilibration Test.  Perit Dial Bull  1987; 7:138-144.

44. Ho-dac-Pannekeet MM, Atasever B, Struijk D, Krediet R: Analysis of ultrafiltration failure in peritoneal dialysisi patients by means of standard peritoneal permeability analysis.  Perit Dial Int  1997; 17:144-150.

45. Rippe B: Peritoneal dialysis capacity.  Perit Dial Int  1997; 17(suppl 2):S131-S134.

46. Lilaj T, Vychytil A, Schneider B, et al: Influence of the preceding exchange on peritoneal equilibration test results: A prospective study.  Am J Kidney Dis  1999; 34:247-253.

47. Blake PG, Flowerdew G, Blake RM, Oreopouolos DG: Serum albumin in patients on continuous ambulatory peritoneal dialysis—predictors and correlations with outcomes.  J Am Soc Nephrol  1993; 3:1501-1507.

48. Churchill DN, Thorpe KE, Nolph KD, et al: Increased peritoneal membrane transport is associated with decreased patient and technique survival for continuous peritoneal dialysis patients. The Canada-USA (CANUSA) Peritoneal Dialysis Study Group.  J Am Soc Nephrol;  1998; 9:1285-1292.

49. Davies SJ, Phillips L, Russell GI: Peritoneal solute transport predicts survival on CAPD independently of residual renal function.  Nephrol Dial Transplant  1998; 13:962-968.

50. Blake PG: What is the problem with high transporters?.  Perit Dial Int  1997; 17:317-319.

51. Tzamaloukas AH, Saddler MC, Murata GH: Symptomatic fluid retention in patients on continuous peritoneal dialysis.  J Am Soc Nephrol  1995; 6:198-206.

52. Margetts PJ, McMullin JP, Rabbat CG, Churchill DN: Peritoneal membrane transport and hypoalbuminemia: cause or effect?.  Perit Dial Int  2000; 20:14-18.

53. Chung SH, Heimburger O, Stenwinkel P, et al: Influence of peritoneal transport rate, inflammation and fluid removal on nutritional status and clinical outcome in prevalent peritoneal dialysis patients.  Perit Dial Int  2003; 23:174-183.

54. Blake PG, Abraham G, Sombolos K, et al: Changes in peritoneal membrane transport rates in patients on long-term CAPD.  Adv Perit Dial  1989; 5:3-7.

55. Heimburger O, Wang T, Lindholm B: Alterations in water and solute transport with time on peritoneal dialysis.  Perit Dial Int  1999; 19(suppl 2):S83-S90.

56. Davies SJ, Bryan J, Phillips L, Russell GI: Longitudinal changes in peritoneal kinetics: The effects of peritoneal dialysis and peritonitis.  Nephrol Dial Transplant  1996; 11:498-506.

57. Mujais S, Nolph K, Gokal R, et al: Evaluation and management of ultrafiltration problems in peritoneal dialysis. International Society for Peritoneal Dialysis Ad Hoc Committee on Ultrafiltration Management in Peritoneal Dialysis.  Perit Dial Int  2000; 20(suppl 4):S5-S21.

58. Williams JD, Craig KG, Topley N, et al: Morphologic changes in the peritoneal membrane of patients with renal disease.  J Am Soc Nephrol  2002; 13:470-479.

59. De Vriese AS, Mortier S, Lameire NH: What happens to the peritoneal membrane in long-term peritoneal dialysis?.  Perit Dial Int  2001; 21(suppl 3):S9-S18.

60. De Vriese AS: Rage in the peritoneum.  Pert Dial Int  2005; 25:8-11.

61. Davies SJ, Phillips L, Naish PF, Russell GI: Peritoneal glucose exposure and changes in membrane solute transport with time on peritoneal dialysis.  J Am Soc Nephrol  2001; 12:1046-1051.

62. Davies SJ, Brown EA, Frandsen NE, et al: Longitudinal membrane function in functionally anuric patients treated with APD: Data from EAPOS on the effects of glucose and icodextrin prescription.  Kidney Int  2005; 67:1609-1615.

63. Gokal R, Alexander S, Ash S, et al: Peritoneal catheters and exit-site practices toward optimum peritoneal access: 1998 Update.  Perit Dial Int  1998; 18:11-33.

64. Flanigan M, Gokal R: Peritoneal catheters and exit-site practices toward optimum peritoneal access: a review of current developments.  Perit Dial Int  2005; 25:132-139.

65. Palmer RH, Quinton WE: Prolonged peritoneal dialysis for chronic renal failure.  Lancet  1964; 1:700-702.

66. Tenckhoff J, Schechter HA: Bacteriologically safe peritoneal access device.  Trans A Soc Artif Intern Organs  1968; 14:181-187.

67. Twardowski ZJ, Prowant BF, Nicholas WK, et al: Six year experience with swan neck catheter.  Perit Dial Int  1992; 12:384-389.

68. Nielsen PK, Hemmingsen C, Friis SU, et al: Comparison of straight and curled Tenckhoff peritoneal dialysis catheters implanted by percutaneous technique: a prospective randomized study.  Perit Dial Int  1995; 15:18-21.

69. Ash SR, Sutton JM, Rankus RA, et al: Clinical trials of the T-fluted (Ash Advantage) peritoneal dialysis catheter.  Adv Ren Repl Ther  2002; 9:133-143.

70. Cruz C: Cruz catheter: implementation technique and clinical results.  Perit Dial Int  1994; 14(suppl 3):S59-S63.

71. Sikaneta T, Cheung KM, Abdolell M, et al: The Toronto Western Hospital Catheter: One center's experience and review of the literature.  Int J Artif Organs  2006; 29:59-63.

72. Di Paolo N, Sansoni E, Cappelletti A: The self-locating catheter: Review and cost analysis.  Int J Artif Organs  2006; 29:113-122.

73. Moncrief JW, Popovich RP, Dasgupta M, et al: reduction in peritonitis incidence in continuous ambulatory peritoneal dialysis with a new catheter and implantation technique.  Perit Dial Int  1993; 13(suppl):S329-S331.

74. Eklund BH, Honkanen EO, Kala AR, Kyllonen LE: Peritoneal dialysis access: prospective randomized comparison of the Swan neck and Tenckhoff catheters.  Perit Dial Int  1995; 15:353-356.

75. Eklund B, Honkanen E, Kyllonen L, et al: Peritoneal dialysis access: prospective randomized comparison of single-cuff and double-cuff straight Tenckhoff catheters.  Nephrol Dial Transplant  1997; 12:2664-2666.

76. Strippoli GF, Tong A, Johnson D, et al: Catheter-related interventions to prevent peritonitis in peritoneal dialysis: A systematic review of randomized, controlled trials.  J Am Soc Nephrol  2004; 15:2735-2746.

77. Lye WC, Kour NW, van der Straaten J, et al: A prospective randomized comparison of the Swan neck, coiled and straight Tenckhoff catheters in patients on CAPD.  Perit Dial Int  1996; 16(suppl 1):S333-S335.

78. Danielsson A, Blohme L, Traneaus A, Hylander B: A prospective randomized study on the effect of a subcutaneously “buried” peritoneal dialysis catheter technique versus standard technique on the incidence of peritonitis and exit-site infection.  Perit Dial Int  2002; 22:214-219.

79. Warady BA, Sullivan EK, Alexander SR: Lessons from the peritoneal dialysis patient database: a report of North American Pediatric Renal Transplant Cooperative Study.  Kidney Int  1996; 49(suppl 53):S68-S71.

80. US Renal Data System 1992: Annual Data Report. VI. Catheter related factors and peritonitis risk in CAPD patients.  Am J Kidney Dis  1992; 19(suppl. 2):48-54.

81. Strippoli GF, Tong A, Johnson D, et al: Antimicrobial agents to prevent peritonitis in peritoneal dialysis: A systematic review of randomized controlled trials.  Am J Kidney Dis  2004; 44:591-603.

82. Gadallah MF, Ramdeen G, Mignone J, et al: Role of preoperative antibiotic prophylaxis in preventing postoperative peritonitis in newly placed peritoneal dialysis catheters.  Am J Kidney Dis  2000; 36:1014-1019.

83. Ejlersen E, Steven K, Lokkergaard H: Paramedian versus midline incision for the insertion of permanent peritoneal dialysis catheters.  Scand J Urol  1990; 24:151-154.

84. Gadallah MF, Pervez A, el-Shahawy MA, et al: Peritoneoscopic versus surgical placement of peritoneal dialysis catheters: a prospective randomized study on outcome.  Am J Kidney Dis  1999; 33:118-122.

85. Eklund B, Groop PH, Halme L, et al: Peritoneal dialysis access: a comparison of peritoneoscopic and surgical insertion techniques.  Scan J Urol Nehrol Scand  1998; 32:405-408.

86. Mital S, Fried LF, Piraino B: Bleeding complications associated with peritoneal dialysis catheter insertion.  Perit Dial Int  2004; 24:478-480.

87. Shea M, Himel SD, Beck AM: Use of tissue plasminogen activator for thrombolysis in occluded peritoneal dialysis catheters in children.  Adv Perit Dial  2001; 17:249-252.

88. Savader SJ, Lund G, Scheel PJ, et al: Guide wire directed manipulation of malfunctioning peritoneal dialysis catheters: a critical analysis.  J Vas Interv Radiol  1997; 8:957-963.

89. Brandt CP, Ricanati ES: Use of laparascopy in the management of malfunctioning peritoneal dialysis catheters.  Adv Perit Dial  1996; 12:223-226.

90. Lee M, Donovan JF: Laparascopic omentectomy for salvage of peritoneal dialysis catheters.  J Endourol  2002; 16:241-244.

91. Neri L, Viglino G, Cappelletti A, Gandolfo C: Evaluation of drainage time and alarms with various automated peritoneal dialysis modalities.  Adv Perit Dial  2001; 17:72-74.

92. Sitter T, Sauter N: Impact of glucose in peritoneal dialysis: Saint or sinner.  Perit Dial Int  2005; 25:415-425.

93. Gjessing J: Addition of amino acids to peritoneal dialysis fluid.  Lancet  1968; 2:812.

94. Jones MR: Intraperitoneal amino acids: A therapy whose time has come?.  Perit Dial Int  1995; 15(suppl 5):S67-S74.

95. Kopple JD, Bernard D, Messana J, et al: Treatment of malnourished CAPD patients with an amino acid-based dialysate.  Kidney Int  1995; 47:1148-1157.

96. Jones MR, Gehr TW, Burkart JM, et al: Replacement of amino acid and protein losses with 1.1% amino acid peritoneal dialysis solution.  Perit Dial Int  1998; 18:210-216.

97. Li FK, Chan LY, Woo JC, et al: A 3-year prospective, randomized, controlled study on amino acid dialysate in patients on CAPD.  Am J Kidney Dis  2003; 42:173-183.

98. Mistry CD, Gokal R, Peers E: A randomized multicenter clinical trial comparing isosmolar icodextrin with hyperisosmolar glucose solutions in CAPD. MIDAS Study Group. Multicenter Investigation of Icodextrin in Ambulatory Peritoneal Dialysis.  Kidney Int  1994; 46:496-503.

99. Frampton JE, Plosker GL: Icodextrin: A review of its use in peritoneal dialysis.  Drugs;  2003; 63:2079-2105.

100. Moberly JB, Mujais S, Gehr T, et al: Pharmacokinetics of icodextrin in peritoneal dialysis patients.  Kidney Int  2002; 81(suppl):S22-S33.

101. Martis L, Patel M, Jiertych J, et al: Aseptic peritonitis due to peptidoglycan contamination of pharmacopoeia standard dialysis solution.  Lancet  2005; 365:588-594.

102. Davies SJ, Woodrow G, Donovan K, et al: Icodextrin improves the fluid status of peritoneal dialysis patients: Results of a double-blind randomized controlled trial.  J Am Soc Nephrol  2003; 14:2338-2344.

103. Konings CJ, Kooman JP, Schonck M, et al: Effect of icodextrin on volume status, blood pressure and echographic parameters: A randomized study.  Kidney Int  2003; 63:1556-1563.

104. Woodrow G, Oldroyd B, Stables G, et al: Effects of icodextrin in automated peritoneal dialysis on blood pressure and bioelectrical impedance analysis.  Nephrol Dial Transplant  2000; 15:862-866.

105. Wilkie ME, Plant MJ, Edwards L, Brown CB: Icodextrin 7.5% dialysate solution (glucose polymer) in patients with ultrafiltration failure: Extension of CAPD technique survival.  Perit Dial Int  1997; 17:84-87.

106. Wolfson M, Piraino B, Hamburger RJ, Morton AR: Icodextrin Study Group. A randomized controlled trial to evaluate the efficacy and safety of icodextrin in peritoneal dialysis.  Am J Kidney Dis  2002; 40:1055-1065.

107. Sisca S, Maggiore U: Beneficial effect of icodextrin on the hypertriglyceridemia of CAPD patients.  Perit Dial Int  2002; 22:727-729.

108. Amici G, Orrasch M, Da Rin G, Bocci C: Hyperinsulinism reduction associated with icodextrin treatment in CAPD patients.  Adv Perit Dial  2001; 17:80-83.

109. Davies SJ: Exploring new evidence of the clinical benefits of icodextrin solutions.  Nephrol Dial Transplant  2006; 21(suppl 2):S47-S50.

110. Van Biesen W, Boer W, De Greve B, et al: A randomized clinical trial with a 0.6% amino acid/1.4% glycerol peritoneal dialysis solution.  Perit Dial Int  2004; 24:222-230.

111. Ledebo I, Wieslander A, Kjellistrand P: Can we prevent the degradation of glucose in peritoneal dialysis solutions?.  Perit Dial Int  2000; 20(suppl 2):S48-S51.

112. Passlick-Deetjen J, Lage C: Lactate-buffered and bicarbonate-buffered solutions with less glucose degradation products in a two-chamber system.  Perit Dial Int  2000; 20(suppl 2):S42-S47.

113. Williams JD, Topley N, Craig KJ, et al: The Euro-Balance Trial: The effect of a new biocompatible peritoneal dialysis fluid (balance) on the peritoneal membrane.  Kidney Int;  2004; 66:408-418.

114. Szeto CC, Chow KM, Lam CW, et al: Clinical biocompatibility of a neutral peritoneal dialysis solution with minimal glucose-degradation products—1-year randomized control trial.  Nephrol Dial Transplant  2007; 22:552-559.

115. Rippe B, Simonsen O, Heimburger O, et al: Long-term clinical effects of a peritoneal dialysis fluid with less glucose degradation products.  Kidney Int  2001; 59:348-357.

116. Lee HY, Park HC, Seo BJ, et al: Superior patient survival for CAPD patients treated with a peritoneal dialysis fluid with neutral pH and low glucose degradation product concentration (Balance).  Perit Dial Int  2005; 25:248-255.

117. Woodrow G: Can biocompatible dialysis fluids improve outcomes in peritoneal dialysis patients?.  Perit Dial Int  2005; 25:230-233.

118. Jones S, Holmes CJ, Krediet RT, et al: Bicarbonate/lactate-based peritoneal dialysis solution increases cancer antigen 125 and decreases hyaluronic acid levels.  Kidney Int;  2001; 59:1529-1538.

119. Fusshoeller A, Plail M, Grabensee B, Plum J: Biocompatibility pattern of a bicarbonate/lactate-buffered peritoneal dialysis fluid in APD: a prospective, randomized study.  Nephrol Dial Transplant  2004; 19:2101-2106.

120. Dratwa M, Wilkie M, Ryckelynck JP, et al: Clinical experience with two physiologic bicarbonate/lactate peritoneal dialysis solutions in automated peritoneal dialysis.  Kidney Int  2003; 88(suppl):S105-S113.

121. Nau B, Schmitt CP, Almeida N, et al: European Pediatric Peritoneal Dialysis Study Group. BIOKID: Randomized controlled trial comparing bicarbonate and lactate buffer in biocompatible peritoneal dialysis solutions in children.  BMC Nephrol  2004; 5:14.

122. Mactier RA, Sprosen TS, Gokal R: Bicarbonate and bicarbonate/slash lactate peritoneal dialysis solutions for the treatment of infusion pain.  Kidney Int  1998; 53:1061-1067.

123. Simonsen O, Venturoli D, Weislander A, et al: Mass transfer of calcium across the peritoneum at three different peritoneal dialysis fluid calcium and glucose concentrations.  Kidney Int  2003; 64:208-215.

124. Sanchez C, Lopez-Barea F, Sanchez-Cabezudo J, et al: Low versus standard calcium dialysate in peritoneal dialysis: Differences in treatment, biochemistry and bone histomorphometry. A randomized multicentre study.  Nephrol Dial Transplant  2004; 19:1587-1593.

125. Stein A, Baker F, Moorehouse J, Walls J: Peritonitis rate: Traditional versus low calcium dialysate.  Am J Kidney Dis  1995; 26:632-633.

126. Wang T, Waniewski J, Heimburger O, et al: A quantitative analysis of sodium transport and removal during peritoneal dialysis.  Kidney Int  1997; 52:1609-1616.

127. Rippe B, Venturoli D, Simonsen O, de Arteaga J: Fluid and electrolyte transport across the peritoneal membrane during CAPD according to the three-pore model.  Perit Dial Int  2004; 24:10-27.

128. Cheng HH, Wang T, Heimburger O, et al: Fluid and solute transport using different sodium concentrations in peritoneal dialysis solutions.  Perit Dial Int  2001; 21:65-71.

129. Khandelwal M, Oreopoulos DG: Is there a need for low sodium dialysis solution for peritoneal dialysis patients?.  Adv Perit Dial  2004; 20:156-162.

130. Kiernan L, Kliger A, Gorban-Brennan N, et al: Comparison of CAPD related infections with different “Y-tubing” exchange systems.  Am Soc Nephrol  1995; 5:1538-1835.

131. Li PK, Law MC, Chow KM: Comparison of clinical outcome and ease of handling in two double-bag systems in continuous ambulatory peritoneal dialysis: a prospective, randomized, controlled, multicenter study.  Am J Kidney Dis  2002; 40:373-380.

132. de Fijter CW, Oe LP, Nauta JJ, et al: Clinical efficacy and morbidity associated with continuous cyclic compared with continuous ambulatory peritoneal dialysis.  Ann Intern Med  1994; 120:264-271.

133. Ronco C, Amerling R, Dell'aquila R, et al: Evolution of technology for automated peritoneal dialysis.  Contrib Nephrol  2006; 150:291-309.

134. Diaz-Buxo JA: Enhancement of peritoneal dialysis: The PD Plus concept.  Am J Kidney Dis  1996; 27:92-98.

135. Nolph KD, Twardowski ZJ, Khanna R, et al: Tidal peritoneal dialysis with racemic or L-lactate solutions.  Perit Dial Int  1990; 10:161-164.

136. Blake PG, Floyd J, Spanner E, Peters K: How much extra does “adequate” peritoneal dialysis cost?.  Perit Dial Int  1996; 16(suppl 1):S171-S175.

137. Blake PG, Sombolos K, Abraham G, et al: Lack of correlation between urea kinetic indices. Clinical outcomes in CAPD patients.  Kidney Int  1991; 39:700-706.

138. Canada-USA (CANUSA) Peritoneal Dialysis Study Group : Adequacy of dialysis and nutrition in continuous peritoneal dialysis; association with clinical outcomes.  J Am Soc Nephrol  1996; 7:198-207.

139. NKF-K/DOQI: Clinical Practice Guildelines for Peritoneal Dialysis Adequacy: Update 2000.  Am J Kidney Dis  2001; 37(Suppl 1):S65-S136.

140. Blake PG, Bargman , Bick J, et al: Guidelines for adequacy and nutrition in peritoneal dialysis. Canadian Society of Nephrology.  J Am Soc Nephrol  1999; 10(suppl 13):S311-S321.

141. Paniagua R, Amato D, Vonesh E, et al: Mexican Nephrology Collaborative Study Group. Effects of increased peritoneal clearances on mortality rates in peritoneal dialysis: ADEMEX, a prospective, randomized, controlled trial.  Am Soc Nephrol  2002; 13:120-130.

142. Lo WK, Ho YW, Li CS, et al: Effect of Kt/V on survival and clinical outcome in CAPD patients in a randomized prospective study.  Kidney Int  2003; 64:649-656.

143. Eknoyan G, Beck GJ, Cheung AK, et al: Effect of dialysis dose and membrane flux in maintenance hemodialysis.  N Engl J Med  2002; 347:2010-2019.

144. Blake PG: Adequacy of dialysis revisited.  Kidney Int  2003; 63:1587-1599.

145. Peritoneal Dialysis Adequacy Work Group : Clinical Practice Guidelines for Peritoneal Adequacy, Update 2006.  Am J Kidney Dis  2006; 48(Suppl 1):S91-S97.

146. Lo WK, Bargman JM, Burkart J, et al: Guidelines on targets for solute and fluid removal in adult patients on chronic peritoneal dialysis.  Perit Dial Int  2006; 26:520-522.

147. Keshaviah PR, Nolph KD, Van Stone JC: The peak concentration hypothesis: A urea kinetic approach to comparing the adequacy of CAPD and hemodialysis.  Perit Dial Int  1989; 9:257-260.

148. Depner TA: Benefits of more frequent dialysis: Lower TAC at the same Kt/V.  Nephrol Dial Transplant  1998; 13(suppl 6):20-24.

149. Blake P, Burkart JM, Churchill DM, et al: Recommended clinical practices for maximizing peritoneal dialysis clearances.  Perit Dial Int  1996; 16:448-456.

150. Blake PG, Korbet SM, Blake R, et al: A multicenter study of noncompliance with CAPD exchanges in U.S. and Canadian patients.  Am J Kidney Dis  2000; 35:506-514.

151. Durand PY: Optimization of fill volumes and automated peritoneal dialysis.  Perit Dial Int  2000; 20:601-602.

152. Durand PY, Shanliau J, Kessler M: Intraperitoneal pressure/volume effects on peritoneal transport in automated peritoneal dialysis.  Contrib Nephrol  1999; 129:54-61.

153. Fischbach M, Dheu C: Hydrostatic intraperitoneal pressure: An objective tool for analyzing individual tolerance of intraperitoneal volume.  Perit Dial Int  2005; 25:338-339.

154. Rippe B: Is intraperitoneal pressure important?.  Perit Dial Int  2006; 26:317-319.

155. Pirpasopolous M, Rahman M, Lindsay RM, Kennedy AC: A cost-effectiveness study of dwell times in peritoneal dialysis.  Lancet  1972; 2:1135-1136.

156. Robson M, Oreopoulos DG, Izatt S, et al: Influence of exchange volume in dialysate flow rate on solute clearance in peritoneal dialysis.  Kidney Int  1978; 14:486-490.

157. Perez RA, Blake PG, McMurray S, et al: What is the optimal frequency of cycling in automated peritoneal dialysis?.  Perit Dial Int  2000; 20:548-556.

158. Kumano K, Yamashita A, Sakai T: Optimal number of dialysate exchanges in automated peritoneal dialysis.  Adv Perit Dial  1993; 9:110-113.

159. Piraino B, Bender F, Bernardini J: A comparison of clearances on tidal peritoneal dialysis and intermittent peritoneal dialysis.  Perit Dial Int  1994; 14:145-148.

160. Juergensen PH, Murphy AL, Pherson KA, et al: Tidal peritoneal dialysis: Comparison of different tidal regimens and automated peritoneal dialysis.  Kidney Int  2000; 57:2603-2607.

161. Aasarod K, Wideroe TE, Flakne SC: A comparison of solute clearance and ultrafiltration volume in peritoneal dialysis with total or fractional (50%) intraperitoneal volume exchange with the same dialysate flow rate.  Nephrol Dial Transplant  1997; 12:2128-2132.

162. Ates K, Nergizoglu G, Keven K, et al: Effect of fluid on sodium removal on mortality in peritoneal dialysis patients.  Kidney Int  2001; 60:767-776.

163. Brown EA, Davies SJ, Rutherford P, et al: EAPOS Group. Survival of functionally anuric patients on automated peritoneal dialysis: The European APD Outcome Study.  J Am Soc Nephrol  2003; 14:2948-2957.

164. Sharma AP, Blake PG: Should “fluid removal” be used as an adequacy target in peritoneal dialysis?.  Perit Dial Int  2003; 23:107-108.

165. Medcalf JF, Harris KP, Walls J: Role of diuretics in the preservation of residual renal function in patients on CAPD.  Kidney Int  2001; 59:1128-1133.

166. Li PK, Chow KM, Wong TY, et al: Effects of an angiotensin-converting enzyme inhibitor on residual renal function in patients receiving peritoneal dialysis. A randomized controlled study.  Ann Intern Med  2003; 139:105-112.

167. Suzuki H, Kanno Y, Sugahara S, et al: Effects of an angiotensin II receptor blocker, valsartan, on residual renal function in patients on CAPD.  Am J Kidney Dis  2004; 43:1056-1064.

168. Rodriguez-Carmona A, Perez-Fontan MP: Sodium removal in patients undergoing CAPD and automated peritoneal dialysis.  Perit Dial Int  2002; 22:705-713.

169. Heimburger O, Waniewski J, Werynski A, et al: Peritoneal transport in CAPD patients with permanent loss of ultrafiltration capacity.  Kidney Int  1990; 38:495-506.

170. Fussholler A, zur Nieden S, Grabensee B, Plum J: Peritoneal fluid and solute transport: Influence of treatment time, peritoneal dialysis modality, and peritonitis incidence.  J Am Soc Nephrol  2002; 13:1055-1060.

171. Krediet RT, Lindholm B, Rippe B: Pathophysiology of peritoneal membrane failure.  Perit Dial Int  2000; 20(Suppl 4):S22-S42.

172. Stenvinkel P, Barany P, Chung SH, et al: A comparative analysis of nutritional parameters as predictors of outcome in male and female ESRD patients.  Nephrol Dial Transplant  2002; 17:1266-1274.

173. Chung SH, Lindholm B, Lee HB: Influence of initial nutritional status on continuous ambulatory peritoneal dialysis patient survival.  Perit Dial Int  2002; 20:19-26.

174. Fenton SS, Johnston N, Delmore T, et al: Nutritional assessment of continuous ambulatory peritoneal dialysis patients.  ASAIO Trans  1987; 33:650-653.

175. Pollock CA, Ibels LS, Allen BJ, et al: Total body nitrogen as a prognostic marker in maintenance dialysis.  J Am Soc Nephrol  1995; 6:82-88.

176. Perez RA, Blake PG, Spanner E, et al: High creatinine excretion ratio predicts a good outcome in peritoneal dialysis.  Am J Kidney Dis  2000; 36:362-367.

177. Mittman N, Avram MM, Oo KK, Chattopadhyay J: Serum prealbumin predicts survival in hemodialysis and peritoneal dialysis: Ten years of prospective observation.  Am J Kidney Dis  2001; 38:1358-1364.

178. Chung SH, Lindholm B, Lee HB: Is malnutrition an independent predictor of mortality in peritoneal dialysis patients?.  Nephrol Dial Transplant  2003; 18:2134-2140.

179. Yeun JY, Kaysen GA: Acute phase proteins and peritoneal dialysate albumin loss are the main determinants of serum albumin in peritoneal dialysis patients.  Am J Kidney Dis  1997; 30:923-927.

180. Jones CH, Wells L, Stoves J, et al: Can a reduction in extra cellular fluid volume result in increased serum albumin in peritoneal dialysis patients?.  Am J Kidney Dis  2002; 39:872-875.

181. Gonzalez-Espinoza L, Gutierrez-Chavez J, del Campo FM, et al: Randomized open label controlled clinical trial of oral administration of an egg albumin-based protein supplement to patients on CAPD.  Perit Dial Int  2005; 25:173-180.

182. Teixido-Planas J, Ortiz A, Coronel F, et al: Oral protein-energy supplements in peritoneal dialysis: A multicenter study.  Perit Dial Int  2005; 25:63-72.

183. Stratton RJ, Bircher G, Fouque D, et al: Multinutrient oral supplements and tube feeding in maintenance dialysis: A systematic review on meta-analysis.  Am J Kidney Dis  2005; 46:387-405.

184. Stein A, Moorehouse J, Iles-Smith H, et al: Role of an improvement in acid-based status and nutrition in CAPD patients.  Kidney Int  1997; 52:1089-1095.

185. Szeto CC, Wong TY, Chow KM, et al: Oral sodium bicarbonate for the treatment of metabolic acidosis in peritoneal dialysis patients: A randomized placebo-control trial.  J Am Soc Nephrol  2003; 14:2119-2126.

186. May RC, Bailey JL, Mitch WE, et al: Glucocoricoids and acidosis stimulate protein and amino acid metabolism invivo.  Kidney Int  1996; 49:679-683.

187. Mitch WE: Influence of metabolic acidosis on nutrition.  Am J Kidney Dis  1997; 29:46-48.

188. Kasimatis E, Maksich D, Jassel V, et al: Predictive factors of low HCO3 levels in peritoneal dialysis patients.  Clin Nephrol  2005; 63:290-296.

189. Ikizler TA, Wingard RL, Breyer JA, et al: Short-term effects of recombinant human growth hormone in CAPD patients.  Kidney Int  1994; 46:1178-1183.

190. Fouque D, Peng SC, Shamir E, Kopple JD: Recombinant human insulin-like growth factor-one induces an anabolic response in malnurished CAPD patients.  Kidney Int  2000; 57:646-654.

191. Johnansen KL, Mulligan K, Schambelan M: Anabolic effects of nandrolone decanoate in patients receiving dialysis: A randomized controlled trial.  JAMA  1999; 281:1275-1281.

192. Harty J, Boulton H, Faragher B, et al: The influence of small solute clearance and dietary protein intake in CAPD patients. A methodologic analysis based on cross-sectional and prospective studies.  Am J Kidney Dis  1996; 28:553-560.

193. Blake PG, Oreopoulos DG: Answers to all your questions about peritoneal urea clearance and nutrition in CAPD patients.  Perit Dial Int  1996; 16:248-251.

194. Wynne K, Giannitsopoulou K, Small CJ, et al: Subcutaneous ghrelin enhances acute food intake in malnurished patients who receive maintenance peritoneal dialysis: A randomized, placeo-controlled trial.  J Am Soc Nephrol  2005; 16:2111-2118.

195. Kaysen GA, Rathore V, Shearer GC, Depner TA: Mechanisms of hypoalbuminemia in hemodialysis patients.  Kidney Int  1995; 48:510-516.

196. Qureshi AR, Alvestrand A, Danielsson A, et al: Factors predicting malnutrition in hemodialysis patients: A cross-sectional study.  Kidney Int  1998; 53:773-782.

197. Stenvinkel P, Heimburger O, Paultri F: Strong association between malnutrition, inflammation and atherosclerosis in chronic renal failure.  Kidney Int  1999; 55:1899-1911.

198. Herzig KA, Purdie DM, Chang W, et al: Is C-reactive protein a useful predictor of outcome in peritoneal dialysis patients?.  J Am Soc Nephrol  2001; 12:814-821.

199. Fine A: Relevance of C-reactive protein levels in peritoneal dialysis patients.  Kidney Int  2002; 61:615-620.

200. Ducloux D, Bresson-Vautrin C, Kribbs M, et al: C-reactive protein in cardiovascular disease in peritoneal dialysis patients.  Kidney Int  2002; 62:1417-1422.

201. Avila-Diaz M, Ventura MD, Valle D, et al: Inflammation and extracellular volume expansion are related to sodium and water removal in patients on peritoneal dialysis.  Perit Dial Int  2006; 26:574-580.

202. Wang AY, Woo J, Lam CW, et al: Associations of serum fetuin-A with malnutrition, inflammation, atherosclerosis and valvular calcification syndrome and outcome in peritoneal dialysis patients.  Nephrol Dial Transplant  2005; 20:1676-1685.

203. Wang AY, Wang M, Woo J, et al: Inflammation, residual kidney function and cardiac hypertrophy are interrelated and combine adversely to enhance mortality and cardiovascular risk of peritoneal dialysis patients.  J Am Soc Nephrol  2004; 15:2186-2194.

204. Chung SH, Heimburger O, Stenvinkel P, et al: Association between inflammation and changes in residual renal function and peritoneal transport rate during the first year of dialysis.  Nephrol Dial Transplant  2001; 16:2240-2245.

205. Pecoits-Filho R, Carvalho MJ, Stenvinkel P, et al: Systemic and intraperitoneal interleukin-6 system during the first year of peritoneal dialysis.  Perit Dial Int  2006; 26:53-63.

206. Paniagua R, Frias Y, de Ventura MJ, et al: C-reactive protein and anti-Chlamydia pneumoniae antibodies as risk factors of cardiovascular death in incident patients on peritoneal dialysis.  Perit Dial Int  2003; 23:132-137.

207. Haubitz M, Brunkhorst R: C-reactive protein and chronic Chlamydia pneumoniae infection—long-term predictors for cardiovascular disease and survival in patients on peritoneal dialysis.  Nephrol Dial Transplant  2001; 16:809-815.

208. Wong TY, Szeto CC, Chow KM, et al: Rosiglitazone reduces insulin requirement and C-reactive protein levels Type 2 diabetic patients receiving peritioneal dialysis.  Am J Kidney Dis  2005; 46:713-719.

209. Ikejiri A, Hirano T, Muriayama S, et al: Effects of atorvastatin on triglyceride-rich lipoproteins, low-density lipoprotein subclass, and C-reactive protein in hemodialysis patients.  Metabolism  2004; 53:1113-1117.

210. Shurraw S, Tonelli M: Statins for treatment of dyslipidemia in chronic kidney disease.  Perit Dial Int  2006; 26:523-539.

211. Prasad K: C-reactive protein lowering agents.  Cardiovasc Drug Rev  2006; 24:33-50.

212. Kouloris S, Symeonides P, Triantafyllou K, et al: Comparison of the effects of ramipril versus telmisartan in reducing serum levels of high sensitivity C-reactive protein and oxidized low-density lipoprotein cholestrol in patients with type 2 diabetis mellitus.  Am J Cardiol  2005; 95:1386-1388.

213. Rottembourg J, Issad B, Gallego JL, et al: Evolution of residual renal function in patients undergoing maintenance hemodialysis or continuous ambulatory peritoneal dialysis.  Proc Eur Dial Transplant  1982; 19:397-409.

214. Cancarini GC, Brunori G, Camerini G, et al: Renal fucntion recovery and maintenance of residual diuresis in CPAD and hemodialysis.  Perit Dial Bull  1986; 6:77-79.

215. Moist LM, Port FK, Orzol SM, et al: Predictors of loss of residual renal function among new dialysis patients.  J Am Soc Nephrol  2000; 11:556-564.

216. Jansen MA, Hart AA, Korevaar JC, et al: NECOSAD Study Group: Predictors of the rate of decline of residual renal function in incident dialysis patients.  Kidney Int  2002; 62:1046-1053.

217. Misra M, Vonesh E, VanStone JC, et al: Effect of cause and time of dropout on the residual GFR: A comparative analysis of the decline of GFR on dialysis.  Kidney Int  2001; 59:754-763.

218. McKane W, Shandna SM, Tattersall JE, et al: Identical decline of residual renal function in high-flux biocompatible hemodialysis and CAPD.  Kidney Int  2002; 61:256-265.

219. Bargman JM, Thorpe KE, Churchill DN: CANUSA Peritoneal Dialysis Study Group.  J Am Soc Nephrol  2001; 12:2158-2162.

220. Termorshuizen F, Korevaar JC, Dekker FW, et al: NECOSAD Study Group: The relative importance of residual renal function compared with peritoneal clearance for patient survival and quality of life: An analysis of the Netherlands Cooperative Study on the Adequacy of Dialysis (NECOSAD)—2.  Am J Kidney Dis  2003; 41:1293-1302.

221. Wang AY, Li KN: The importance of residual renal function in dialysis patients.  Kidney Int  2006; 69:1726-1732.

222. Wang AY, Woo J, Wang M, et al: Important differentiation of factors that predict outcome in peritoneal dialysis patients with different degrees of residual renal function.  Nephrol Dial Transplant  2005; 20:396-403.

223. Wang AY, Wang M, Woo J, et al: A novel association between residual renal function and left ventricular hypertrophy in peritoneal dialysis patients.  Kidney Int  2002; 62:639-647.

224. Wang AY, Woo J, Sea MM, Law MC, et al: Hyperphosphatemia in Chinese peritoneal dialysis patients with and without residual kidney function: What are the implications?.  Am J Dis  2004; 43:712-720.

225. Jones MR: Etiology of severe malnutrition: results of an international cross-sectional study in continuous ambulatory peritoneal dialysis patients.  Am J Kidney Dis  1994; 23:412-420.

226. Wang AY, Sea MM, Yip R, et al: Independent effects of residual renal function and dialysis adequacy on actual dietary protein, calorie, and other nutrient intake in patients on continuous ambulatory peritoneal dialysis.  J Am Soc Nephrol  2001; 12:2450-2457.

227. Chung SH, Heimburger O, Stenvinkel P, et al: Association between residual renal function, inflammation and patient survival in new peritoneal dialysis patients.  Nephrol Dial Transplant  2003; 18:590-597.

228. Hidaka H, Nako T: Preservation of residual renal function and factors affecting its decline in patients on peritoneal dialysis.  Nephrology (Carlton)  2003; 8:184-191.

229. Shin SK, Noh H, Kang SW, et al: Risk factors influencing the design of residual renal function in CAPD patients.  Perit Dial Int  1999; 19:138-142.

230. Singhal MK, Bhaskaran S, Vidgen E, et al: Rate of decline of residual renal function in patients on continuous peritoneal dialysis and factors affecting it.  Perit Dial Int  2000; 20:429-438.

231. Holley JL, Aslam N, Bernardini J, et al: The influence of demographic factors and modality on loss of residual renal function in incident peridialysis patients.  Perit Dial Int  2001; 21:302-305.

232. Baker RJ, Senior H, Clemenger M, Brown EA: Empirical aminoglycocides for peritonitis do not affect residual renal function.  Am J Kidney Dis  2003; 41:670-675.

233. Woodrow G: A randomized controlled trial to determine whether treatment with a neutral pH, low glucose degradation product dialysate (Balance) prolongs residual renal function in peritoneal dialysis patients.  Perit Dial Int  2006; 26:113-114.

234. Hirosige A, Yuu K, Soejima M, et al: Rapid decline of residual function in patients on automated peritoneal dialysis.  Perit Dial Int  1996; 16:307-315.

235. Hufnagel G, Michel C, Queffeulou G, et al: The influence of automated peritoneal dialysis on the decrease in residual renal function.  Nephrol Dial Transplant  1999; 14:1224-1228.

236. Rodriguez-Carmona A, Perez-Fontan M, Garca-Naverio R, et al: Paired time profiles of ultrafiltration, sodium removal and renal function in incident CAPD and automated peritoneal dialysis patients.  Am J Kidney Dis  2004; 41:132-145.

237. Ganesh SK, Hulbert-Sheron T, Port FK, et al: Mortality differences by dialysis modality among incident ESRD patients with and without coronary artery disease.  J Am Soc Nephrol  2003; 14:415-424.

238. Stack AG, Malony DA, Rahman NS, et al: Impact of dialysis modality on survival of new ESRD patients with congestive heart failure in the United States.  Kidney Int  2003; 64:1071-1079.

239. Wanner C, Krane V, Marz W, et al: Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis.  N Engl J Med  2005; 21:238-248.

240. Enia G, Mallamachi F, Benedetto FA, et al: Long-term CAPD patients are volume expanded and display more severe left ventricular left hypertrophy than haemodialysis patients.  Nephrol Dial Transplant  2001; 16:1459-1464.

241. Gunal AI, Ilkay E, Kirciman E, et al: Blood pressure control and left ventricular hypertrophy in long-term CAPD and hemodialysis patients: A cross-sectional study.  Perit Dial Int  2003; 23:563-567.

242. Piraino B, Bailie GR, Bernardini J, et al: ISPD guidelines/recommendations. Peritoneal dialysis—related infections recommendations: 2005 update.  Perit Dial Int  2005; 25:107-131.

243. Perez-Fontan M, Rodriguez-Carmona A, Garcia-Naveiro R, et al: Peritonitis-related mortality in patients undergoing chronic peritoneal dialysis.  Perit Dial Int  2005; 25:274-284.

244. Kim DK, Yoo TH, Ryu DR, et al: Changes in causative organisms and their antimicrobial susceptibilities in CAPD peritonitis: a single center's experience over one decade.  Perit Dial Int  2004; 24:424-432.

245. Szeto CC, Leung CB, Chow KM, et al: Change in bacterial etiology of peritoneal dialysis-related peritonitis over 10 years: experience from a centre in South-East Asia.  Clin Microbiol Infect  2005; 11:837-839.

246. Szeto CC, Wong TY, Chow KM, et al: The clinical course of culture-negative peritonitis complicating peritoneal dialysis.  Am J Kidney Dis  2003; 42:567-574.

247. Rocklin MA, Teitelbaum I: Noninfectious causes of cloudy peritoneal dialysate.  Semin Dial  2001; 14:37-40.

248. de Freitas DG, Gokal R: Sterile peritonitis in the peritoneal dialysis patient.  Perit Dial Int  2005; 25:146-151.

249. Morduchowicz G, Van Dyk DJ, Wittenberg C, et al: Bacteremia complicating peritonitis in peritoneal dialysis patients.  Am J Nephrol  1993; 13:278.

250. Sewell DL, Golper TA, Hulman PB, et al: Comparison of large volume culture to other methods of isolation of microorganisms from dialysate.  Perit Dial Int  1990; 10:49-52.

251. Alfa MJ, Degagne P, Olson N, Harding GK: Improved detection of bacterial growth in continuous ambulatory peritoneal dialysis effluent by use of BacT/Alert FAN bottles.  J Clin Microbiol  1997; 35:862-866.

252. Szeto CC, Chow KM, Chung KY, et al: The clinical course of peritoneal dialysis-related peritonitis caused by Corynebacterium species.  Nephrol Dial Transplant  2005; 20:2793-2796.

253. Peacock SJ, Howe PA, Day NP, et al: Outcome following staphylococcal peritonitis.  Perit Dial Int  2000; 20:215-219.

254. Shukla A, Abreu Z, Bargman JM: Streptococcal PD peritonitis—A 10 year review of one centre's experience.  Nephrol Dial Transplant  2006; 21:3545-3549.

255. Troidle L, Kliger AS, Gorban-Brennan N, et al: Nine episodes of CPD-associated peritonitis with vancomycin resistant enterococci.  Kidney Int  1996; 50:1368-1372.

256. Szeto CC, Chow KM, Leung CB, et al: Clinical course of peritonitis due to pseudomonas species complicating peritoneal dialysis: A review of 104 cases.  Kidney Int  2001; 59:2309-2315.

257. Szeto CC, Chow VC, Chow KM, et al: Enterobacteriaceae peritonitis complicating peritioneal dialysis: A review of 210 consecutive cases.  Kidney Int  2006; 69:1245-1252.

258. Bunke CM, Brier ME, Golper TA: Outcomes of single organism peritonitis in peritoneal dialysis: Gram negatives vs. gram positives in the Network 9. Peritonitis Study.  Kidney Int  1997; 52:524-529.

259. Goldie SJ, Kiernan-Troidle L, Torres C, et al: Fungal peritontis in a large chronic peritoneal dialysis population: A report of 55 epidsodes.  Am J Kidney Dis  1996; 28:86-91.

260. Prasad N, Gupta A: Fungal peritonitis in peritoneal dialysis patients.  Perit Dial Int  2005; 25:207-222.

261. Felgueiras J, del Peso G, Bajo A: Risk of technique failure in death in fungal peritoneal is determined mainly be duration on peritoneal dialysis: Single-center experience of 24 years.  Adv Perit Dial  2006; 22:77-81.

262. Prakash KC: Tuberculous peritonitis.  Perit Dial Int  1999; 19(suppl 2):S283-S285.

263. White R, Abreo K, Flanigan R, et al: Nontuberculous microbacterial infections in CAPD patients.  Am J Kidney Dis  1993; 22:581-587.

264. Ariano RE, Franczuk C, Fine A, et al: Challenging the current treatment paradigm for methicillin-resistant Staphylococcus epidermidis peritonitis in peritoneal dialysis patients.  Perit Dial Int  2002; 22:335-338.

265. Lye WC, Leong SO, Lee EJ: Methicillin-resistant Staphylococcus aureus nasal carriage and infections in CAPD.  Kidney Int  1993; 43:1357-1362.

266. Yip T, Tse KC, Lam FC, et al: Risk factors and outcomes of extended-spectrum beta-lactamase-producing E.coli peritonitis in CAPD patients.  Perit Dial Int  2006; 26:191-197.

267. Newman LM, Cacho CP, Schulak JA, Weiss MF: Abdominal catastrophe: definition and proposal for a new approach.  Adv Perit Dial  2001; 17:93-97.

268. Tzamaloukas , Obermiller LE, Gibler LJ: Peritonitis associated with intra-abdominal pathology in continuous ambulatory peritoneal dialysis patients.  Perit Dial Int  1993; 13(suppl 2):S335-S337.

269. Faber MD, Yee J: Diagnosis and management of enteric disease and abdominal catastrophe in peritoneal dialysis patients with peritonitis.  Adv Chronic Kidney Dis  2006; 13:271-279.

270. Kern EO, Newman LN, Cacho CP, et al: Abdominal catastrophe revisited: the risk and outcome of enteric peritoneal contamination.  Perit Dial Int  2002; 22:323-334.

271. Szeto CC, Chow KM, Wong TY, et al: Conservative management of polymicrobial peritonitis complicating peritoneal dialysis—A series of 140 consecutive cases.  Am J Med;  2002; 113:728-733.

272. Fontan MP, Rodriguez-Carmona A, Galed I, et al: Incidence and significance of peritoneal eosinophilia during peritoneal dialysis related peritonitis.  Perit Dial Int  2003; 23:460-464.

273. Charney DI, Gouge SF: Chemical peritonitis secondary to intraperitoneal vancomycin.  Am J Kidney Dis  1991; 17:76-79.

274. Glorieux G, Lameire N, van Biesen W, et al: Specific characteristics of peritoneal leukocyte populations during sterile peritontis associated with icodextrin. CAPD fluids.  Nephrol Dial Transplant  2003; 18:1648-1651.

275. Vas SI: Single daily dose of aminoglycosides in the treatment of CAPD peritonitis.  Perit Dial Int  1993; 13(Suppl 2):S355-S358.

276. Cheng IK, Fang GX, Chau PY, et al: A randomized prospective comparison of oral levofloxacin plus intraperitoneal (IP) vancomycin and IP netromycin plus IP vancomycin as primary treatment of peritonitis complicating CAPD.  Perit Dial Int  1998; 18:371-375.

277. Leung CB, Szeto CC, Chow KM, et al: Cefazolin plus ceftazidime versus imipenem/cilastatin monotherapy for treatment of CAPD peritonitis—A randomized controlled trial.  Perit Dial Int  2004; 24:440-446.

278. Wong KM, Chan YH, Cheung CY, et al: Cefepime versus vancomyin plus netili-micin therapy for CAPD associated peritonitis.  Am J Kidney Dis  2001; 38:127-131.

279. Yeung SM, Walker SE, Tailor SA, et al: Pharmacokinetics of oral ciprofloxacin in CCPD.  Perit Dial Int  2004; 24:447-453.

280. Chan MK, Cheng IK, Ng WS: A randomized prospective trial of three different regimens of treatment of peritonitis in patients on CAPD.  Am J Kidney Dis  1990; 15:155-159.

281. Boeschoten EW, Retra PJ, Krediet RT, et al: CAPD peritonitis: A prospective randomized trial of oral versus intraperitoneal treatment with cephradine.  J Anti Microb Chem Ther  1985; 16:789-797.

282. Troidle L, Finkelstein FO: Peritonitis and automated peritoneal dialysis: A therapeutic konundrum.  Perit Dial Int  2005; 25:142-145.

283. Manley HJ, Bailie GR: Treatment of peritonitis in APD: Pharmacokinetic principles.  Semin Dial  2002; 15:418-421.

284. Salzer W: Antimicrobial-resistant gram-positive bacteria in PD peritonitis and the newer antibiotics used to treat them.  Perit Dial Int  2005; 25:313-319.

285. Bailey EM, Faber MD, Nafziger DA: Linezolid for treatment of vancomycin-resistant peritonitis.  Am J Kidney Dis  2001; 38:E20.

286. Majkowski NL, Mendley SR: Simultaneous removal and replacement of infected peritoneal dialysis catheters.  Am J Kidney Dis  1997; 29:706-711.

287. Swartz R, Messana J, Reynolds J, et al: Simultaneous catheter replacement and removal in refractory peritoneal dialysis infections.  Kidney Int  1991; 40:1160-1165.

288. Lui SL, Yip T, Tse KC, et al: Treatment of refractory Pseudomonas aeruginosa exit-site infection by simultaneous removal and reinsertion of peritoneal dialysis catheter.  Perit Dial Int  2005; 25:560-563.

289. Ter Wee P: Simultaneous removal and reinsertion of peritoneal dialysis catheters: do we know why and when?.  Perit Dial Int  2005; 25:545-546.

290. Maiorca R, Cantaluppi A, Cancarini GC, et al: Prospective controlled triasl of a Y-connector and disinfectant to prevent peritonitis in continuous ambulatory peritoneal dialysis.  Lancet  1983; 2:642-644.

291. Monteon F, Corra-Rotter R, Paniagua R, et al: Prevention of peritonitis with disconnect systems in CAPD. The Mexican Nephrology Collaborative Study Group.  Kidney Int  1998; 54:2123-2128.

292. Mupirocin Study Group : Nasal mupirocin prevents Staphylococcus aureus exit site infection during peritoneal dialysis.  J Am Soc Nephrol  1996; 7:2403-2408.

293. Uttley L, Vardhan A, Mahajan S, et al: Decrease in infections with the introduction of mupirocin cream at the peritoneal dialysis catheter exit site.  J Nephrol  2004; 17:242-245.

294. Ritzay J, Hoffman RM, Tzamaloukas AH: Effect of preventing Staphylococcus aureus carriage on rates of peritoneal catheter-related staphylococcal infections. Literature synthesis.  Perit Dial Int  2001; 21:471-479.

295. Bernardini J, Bender F, Florio T, et al: Randomized, double-blind trial of antibiotic exit site cream for prevention of exit site infection in peritoneal dialysis patients.  J Am Soc Nephrol  2005; 16:539-545.

296. Bargman JM: Complications of peritoneal dialysis related to increased intraabdominal pressure.  Kidney Int  1993; 40(suppl):S75-S80.

297. Garcia-Urena MA, Rodriguez CR, Vega Ruiz V, et al: Prevalence and management of hernias in peritoneal dialysis patients.  Perit Dial Int  2006; 26:198-202.

298. Crabtree JH: Hernia repair without delay in initiating or continuing peritoneal dialysis.  Perit Dial Int  2006; 26:178-182.

299. Del Peso G, Bajo MA, Costero O, et al: Risk factors for abdominal wall complications in peritoneal dialysis patients.  Perit Dial Int  2003; 23:249-254.

300. Girault-Lataste A, Abaza M, Valentin JF: Small volume APD as alternative treatment for peritoneal leaks.  Perit Dial Int  2004; 24:294-296.

301. Szeto CC, Chow KM: Pathogenesis and management of hydrothorax complicating peritoneal dialysis.  Curr Opinion Pulm Med  2004; 10:315-319.

302. Scarpioni R: Acute hydrothorax in a peritoneal dialysis patient: long-term efficacy of autologous blood cell pleurodesis associated with small-volume peritoneal exchanges.  Nephrol Dial Transplant  2003; 18:2200-2201.

303. Cochran ST, Do HM, Ronaghi A, et al: Complications of peritoneal dialysis: evaluation with CT peritoneography.  Radiographics  1997; 17:869-878.

304. Yavuz K, Erden A, Ates K, Erden I: MR peritoneography in complications of CAPD.  Abdom Imaging  2005; 30:361-368.

305. Abraham G, Blake PG, Matthews RE, et al: Genital swelling as a surgical complication of continuous ambulatory peritoneal dialysis.  Surg Gynecol Obstet  1990; 170:306-308.

306. Kwaguchi Y, Kawanishi H, Mujais S, et al: Encapsulating peritoneal sclerosis: definition, etiology, diagnosis and treatment. ISPD Ad Hoc Committee on Ultrafiltration Management in Peritoneal Dialysis.  Perit Dial Int  2000; 20(suppl 4):S43-S55.

307. Nakamoto H: Encapsulating peritoneal sclerosis—a clinician's approach to diagnosis and medical treatment.  Perit Dial Int  2005; 25(suppl 4):S30-S38.

308. Campbell S, Clarke P, Hawley C, et al: Sclerosing peritonitis: identification of diagnostic, clinical and radiological features.  Am J Kidney Dis  1994; 24:819-825.

309. Kawanishi H, Moriishi M: Epidemiology of encapsulating peritoneal sclerosis in Japan.  Perit Dial Int  2005; 25(suppl 4):S14-S18.

310. Rigby RJ, Hawley CM: Sclerosing peritonitis: the experience in Australia.  Nephrol Dial Transplant  1998; 13:154-159.

311. Nomoto Y, Kawaguchi Y, Kubo H, et al: Sclerosing encapsulating peritonitis in patients undergoing CAPD; a report of the Japanese Sclerosing Encapsulating Peritonitis Study Group.  Am J Kidney Dis  1996; 28:420-427.

312. Summers AM, Clancy MJ, Syed F, et al: Single center experience of encapsulating peritoneal sclerosis in patients on peritoneal dialysis for end-stage renal failure.  Kidney Int  2005; 68:2381-2385.

313. Summers AM, Brenchley PEC: An international encapsulating peritoneal sclerosis registry and DNA bank: Why we need one now.  Perit Dial Int  2006; 26:559-563.

314. Kawaguchi Y, Saito A, Kawanishi H, et al: Recommendations on the management of encapsulating peritoneal sclerosis in Japan, 2005: diagnosis, predictive markers, treatment, and preventive measures.  Perit Dial Int  2005; 25(suppl 40):S83-S95.

315. Eltoum MA, Wright S, Atchley J, Mason JC: Four consecutive cases of peritoneal dialysis-related encapsulating peritoneal sclerosis treated successfully with tamoxifen.  Perit Dial Int  2006; 26:203-206.

316. Kawanishi H, Watanabe H, Moriishi M, Tsuchiya S: Successful surgical manage-ment of encapsulating peritoneal sclerosis.  Perit Dial Int  2005; 25(suppl 40):S39-S47.

317. Cocchi R, Viglino G, Cancarini G, et al: Prevalence of hyperlipidemia in a cohort of CAPD patients. Italian Cooperative Peritoneal Dialysis Study Group (ICPDSG).  Miner Electrolyte Metab  1996; 22:22-25.

318. Little J, Phillips L, Russell L, et al: Longitudinal lipid profiles on CAPD: their relationship to weight gain, comorbidity, and dialysis factors.  J Am Soc Nephrol  1998; 9:1931-1939.

319. Fried L, Hutchison A, Stegmayr B, et al: Recommendations for the treatment of lipid disorders in patients on peritoneal dialysis. ISPD guidelines/recommendations. International Society for Peritoneal Dialysis.  Perit Dial Int  1999; 19:7-16.

320. Dasgupta MK: Strategies for managing diabetic patients on peritoneal dialysis.  Adv Perit Dial  2004; 20:200-202.

321. Ahmad S, Salahudeen AK: Effect of obesity on PD versus HD survival: is caloric intake the discriminating factor?.  Kidney Int  2004; 66:866.

322. Heimburger O: Obesity on PD patients: causes and management.  Contrib Nephrol  2003; 140:91-97.

323. Cozzolino M, Gallieni M, Chiarelli G, Brancaccio D: Calcium and phosphate handling in peritoneal dialysis.  Contrib Nephrol  2006; 150:214-225.

324. Moe SM: Management of renal osteodystrophy in peritoneal dialysis patients.  Perit Dial Int  2004; 24:209-216.

325. Coutennye MM, D'Haese PC, Deng JT, et al: High prevalence of adynamic bone disease diagnosed by biochemical markers in a wide sample of the European CAPD population.  Nephrol Dial Transplant  1997; 12:2144-2150.

326. Phakdeekitcharoen B, Leelasa-nguan P: Effects of an ACE inhibitor or angiotensin receptor blocker on potassium in CAPD patients.  Am J Kidney Dis  2004; 44:738-746.

327. Khan AN, Bernardini J, Johnston JR, Piraino B: Hypokalemia in peritoneal dialysis patients.  Perit Dial Int  1996; 16:652.

328. Uribarri J, Prabhakar S, Kahn T: Hyponatremia in peritoneal dialysis patients.  Clin Nephrol  2004; 61:54-58.

329. Moritz ML, del Rio M, Crooke GA, Singer LP: Acute peritoneal dialysis as both cause and treatment of hypernatremia in an infant.  Pediatr Nephrol  2001; 16:697-700.

330. Gabriel DP, Nascimento GV, Caramori JT, et al: Peritoneal dialysis in acute renal failure.  Ren Fail  2006; 28:451-456.

331. Ash S: Peritoneal dialysis in acute renal failure in adults: the underutilized modality.  Contrib Nephrol  2004; 144:239-254.

332. Chitalia VC, Almeida AF, Rai H, et al: Is peritoneal dialysis adequate for hypercatabolic acute renal failure in developing countries?.  Kidney Int  2002; 61:747-757.

333. Bohorques R, Rivas R, Martinez A: Continuous equilibration peritoneal dialysis (CEPD) in acute renal failure.  Perit Dial Int  1990; 10:183.

334. Phu NH, Hien TT, Mai NT, et al: Hemofiltration and peritoneal dialysis in infection-associated acute renal failure in Vietnam.  N Engl J Med  2002; 347:895-902.

335. Schaubel DA, Blake PG, Fenton SS: Trends in CAPD technique failure: Canada 1981-1997.  Perit Dial Int  2001; 21:365-371.

336. Fenton SS, Schaubel DA, Desmeules M, et al: Hemodialysis versus peritoneal dialysis: a comparison of adjusted mortality rates.  Am J Kidney Dis  1997; 30:334-342.

337. Collins AJ, Hao W, Xia H, et al: Mortality risks of peritoneal dialysis and hemodialysis.  Am J Kidney Dis  1999; 34:1065-1074.

338. Bloembergen WE, Port FK, Mauger EA, Wolfe RA: A comparison of mortality between patients treated with hemodialysis and peritoneal dialysis.  J Am Soc Nephrol  1995; 6:177-183.

339. Vonesh EF, Moran J: Mortality in end-stage renal disease: a reassessment of differences between patients treated with hemodialysis and peritoneal dialysis.  J Am Soc Nephrol  1999; 10:345-356.

340. Heaf JG, Lokkegaard HL, Madsen M: Initial survival advantage of peritoneal dialysis relative to haemodialysis.  Nephrol Dial Transplant  2002; 17:112-117.

341. Murphy SW, Foley RN, Barrett BJ, et al: Comparative mortality of hemodialysis and peritoneal dialysis in Canada.  Kidney Int  2000; 57:1720-1726.

342. Jaar BG, Coresh J, Plantinga LC, et al: Comparing the risk for death with peritoneal dialysis and hemodialysis in a national cohort of patients with chronic kidney disease.  Ann Intern Med  2005; 143:174-183.

343. Goeree R, Manalich J, Grootendorst P, et al: Cost analysis of dialysis treatments for end-stage renal disease (ESRD).  Clin Invest Med  1995; 18:455-464.

344. Bruns FJ, Seddon P, Saul M, Zeidel ML: The cost of caring for end-stage kidney disease patients: an analysis based on hospital financial transaction records.  J Am Soc Nephrol  1998; 9:884-890.

345. De Vecchi AF, Dratwa M, Wiedemann ME: Healthcare systems and end-stage renal disease (ESRD) therapies—an international review: costs and reimbursement/funding of ESRD therapies.  Nephrol Dial Transplant  1999; 14(suppl 6):31-41.

346. Blake PG: The complex economics of modality selection.  Perit Dial Int  2004; 24:509-511.