Show Summary Details
Page of

Peritoneal dialysis 

Peritoneal dialysis
Peritoneal dialysis

Simon Davies



Chapter updated throughout, including additional information on the use of peritoneal dialysis in renal replacement, including the results of the EVEREST study, and expanded content on Encapsulating peritoneal sclerosis.

Updated further reading.

Updated on 22 September 2016. The previous version of this content can be found here.
Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 23 January 2019

Abbreviations: CAPD, Continuous ambulatory peritoneal dialysis; APD, automated peritoneal dialysis; CKD Chronic Kidney Disease; AKI, Acute kidney injury; EPS, Encapsulating Peritoneal Sclerosis.


Peritoneal dialysis is achieved by repeated cycles of instillation and drainage of dialysis fluid within the peritoneal cavity, with the two main functions of dialysis—solute and fluid removal—occurring due to the contact between dialysis fluid and the capillary circulation of the parietal and visceral peritoneum across the peritoneal membrane. It can be used to provide renal replacement therapy in acute kidney injury or chronic kidney disease.

Practical aspects—choice of peritoneal dialysis as an effective modality for renal replacement in the short to medium term (i.e. several years) is, for most patients, a lifestyle issue. Typically, a patient on continuous ambulatory peritoneal dialysis (CAPD) will require three to four exchanges of 1.5 to 2.5 litres of dialysate per day. Automated peritoneal dialysis (APD) and use of the glucose-polymer dialysis solution icodextrin enables flexibility of prescription that can mitigate the effects of membrane function (high solute transport). Peritonitis—this remains the commonest complication of peritoneal dialysis, presenting with cloudy dialysis effluent, with or without abdominal pain and/or fever, and confirmed by a leucocyte count greater than 100 cells/µl in the peritoneal fluid. Empirical antibiotic treatment, either intraperitoneal or systemic, with cover for both Gram-positive and Gram-negative organisms, should be commenced immediately while awaiting specific cultures and sensitivities.

Long-term changes in peritoneal membrane function influence survival on peritoneal dialysis if fluid removal is less efficient (ultrafiltration failure), especially in the absence of residual kidney function. This is the main limitation of treatment, along with avoiding the risk of encapsulating peritoneal sclerosis—a life-threatening complication of peritoneal dialysis, particularly if of long duration (15–20% incidence after 10 years) that is characterized by severe inflammatory thickening, especially of the mesenteric peritoneum, resulting in an encapsulation and progressive obstruction of the bowel.


Peritoneal dialysis, along with haemodialysis and renal transplantation, is an effective form of renal replacement therapy. Worldwide it is used for both acute kidney injury (AKI) and chronic renal failure, although patterns of use vary considerably. In the treatment of chronic kidney disease (CKD), peritoneal dialysis should be considered as a short- to medium-term (several years) treatment option, to be used in the context of an integrated approach to renal replacement. For example, it is of value both before and after transplantation in a lifetime of treatment that might require more than one period on each of the renal replacement modalities. The only absolute contraindication is the lack of a peritoneal cavity.

Historical perspective

In the 1890s, Henry Starling, while conducting experiments from which he made his seminal observations on the forces that govern transcapillary fluid transport, provided evidence that the peritoneal membrane could be used to remove fluid and solutes—the fundamental requirements of renal replacement therapy. By the 1940s, peritoneal dialysis was established in the treatment of acute kidney injury (AKI), and a decade later intermittent treatments, typically twice weekly, were used to maintain patients with CKD. Critical to this was the development in 1968 of the Tenckhoff catheter, a soft silicone-based tube that could create permanent access to the peritoneal membrane. In 1978, Popovitch and Moncreif argued that peritoneal dialysis could be performed continuously in ambulant patients with permanent renal failure, coining the term ‘continuous ambulatory peritoneal dialysis’ (CAPD), following which peritoneal dialysis became established as a renal replacement modality that could be undertaken in the home environment. Modifications of this approach, using a device that delivers multiple exchanges of dialysis fluid overnight, so reducing the number of exchanges required during the daytime, are termed ‘automated peritoneal dialysis’ (APD). In addition to the reduced daytime burden of the treatment, the advantages of this approach are that it can facilitate delivery of care by a parent, partner, or assistant. Assisted peritoneal dialysis is the treatment of choice for elderly frail patients in some countries, notably France, providing a viable alternative to centre-based haemodialysis for this patient population.

The use of peritoneal dialysis in renal replacement

Acute kidney injury

Peritoneal dialysis is an important component of the International Society of Nephrology’s 0by25 strategy to reduce the global burden of AKI-related deaths. There is much greater reliance on peritoneal dialysis to treat AKI in developing countries, the exception to this being in infants and small children in whom it remains the treatment of choice in the developed world. Advantages of peritoneal dialysis in AKI include lack of reliance on a clean water supply, relatively stable haemodynamics, no need for vascular access, and the avoidance of anticoagulation and its complications. The International Society for Peritoneal Dialysis has published guidelines for peritoneal dialysis in AKI, freely available at; key recommendations include the routine use of tunnelled soft Tenckhoff catheters, antibiotic prophylaxis, and advice on the constitution of dialysis fluids.

Chronic kidney disease

The EVEREST study (Explaining the Variation in Epidemiology of RRT through Expert opinion, Secondary data sources and Trends over time) found that 70% of the international variability in the use of peritoneal dialysis to treat CKD (the proportion on peritoneal dialysis differs from 4 to 80% by country) is explained by health-care systems, and clinical and macro-economic factors. Independent determinants of lower peritoneal dialysis use include a high proportion of incident patients with diabetes as primary renal disease, a higher percentage of GDP per capita spent on health care, a larger private-for-profit share of haemodialysis facilities, and the high cost of peritoneal dialysis consumables relative to staffing. Currently, the highest growth rates for peritoneal dialysis are seen in the developing world, especially Indo-Asia.

Predictors of survival

As for all types of renal replacement therapy, the most important factors associated with survival are age, inflammation, and comorbidity, the latter being weighted towards cardiovascular disease. As for haemodialysis patients, the continued presence of residual renal function, a benefit that can be demonstrated with creatinine clearances as low as 1 to 2 ml/min, or just 250 ml/day of urinary volume, also improves survival. In addition, there is strong evidence that peritoneal membrane function has an effect. Membranes associated with less efficient fluid removal, due to high solute transport characteristics or low ultrafiltration capacity (see below), are associated with worse outcomes.

Comparison with haemodialysis

Current evidence supports the view that when both haemodialysis and peritoneal dialysis are equally available, the choice between the therapies is largely determined by lifestyle issues (Table The failure to date to perform a randomized controlled trial comparing modalities lies in the fact that most patients (96%) demonstrate a strong modality preference when the choice is made freely available. Modality comparisons have thus relied on observational studies from registry data, with adjustment for baseline factors. Over the last 20 years, survival has improved dramatically (>50% reduction in mortality), such that early survival (1–3 years) tends to be better on peritoneal dialysis than haemodialysis, with equivalence at 5 years. The early survival advantages, most apparent in non-diabetic patients, are likely to reflect relatively better preservation of residual kidney function, avoidance of central venous catheters, and residual confounding. Technique failure is more likely to occur in peritoneal dialysis than haemodialysis, reflecting the effects of membrane damage associated with infection and long-term exposure to dialysis fluid, as well as changes in a patient’s circumstances, such that continuing with home treatment becomes untenable.

Table Typical reasons for choosing peritoneal dialysis or haemodialysis as renal replacement modality

Reasons favouring peritoneal dialysis

Reasons favouring haemodialysis

No vascular access

No peritoneal access

Haemodialysis-induced hypotension

Multiple comorbidities

Preference for home treatment

Increasing age


Living alone

Distant from haemodialysis facility


Family and work commitments

Close to haemodialysis facility

Dependent, when assisted APD performed by a carer or parent is available

Home haemodialysis option preferred and available

APD, automated peritoneal dialysis.

Taken overall, these patient and technique survival differences are not sufficient to override patient preferences associated with lifestyle choices, provided there is an understanding that modality switching may well be required. APD extends choice to the patient or their carer, so accommodating a variety of lifestyles; it is the preferred dialysis treatment for children. Choosing the right dialysis modality is an important component to preparing for renal replacement therapy and can be facilitated by the use of a dialysis decision aid, freely available at

Principles of therapy

Peritoneal dialysis is achieved by repeated cycles of instillation and drainage of dialysis fluid within the peritoneal cavity. The fluid is heat-sterilized, a process that requires an acidic pH; ideally a pH of 2 to 3 should be used, but this requires two or more dialysis bag compartments to enable reconstitution just prior to infusion. This approach enables a physiological pH solution with low concentrations of glucose degradation products and in some cases substitution of bicarbonate for lactate as the buffer. The benefits of biocompatible solutions include reduced infusion pain, better preservation of residual kidney function, and more stable peritoneal solute transport characteristics. They are not universally available, however, and most conventional peritoneal dialysis solutions have a pH of 5.2, with 35 to 40 mmol/litre of lactate as the buffer.

Typically, a patient on peritoneal dialysis will require three or more exchanges of 1.5 to 2.5 litres of dialysate per day. CAPD regimens usually comprise three 4- to 6-hour daytime exchanges and an overnight exchange; APD regimens may use up to six exchanges overnight and a long daytime dwell period. The frequency of overnight exchanges can be further increased by using a ‘tidal’ programme, which exchanges a proportion of the intraperitoneal fluid more rapidly without losing dialysis time due to complete emptying. APD can be further augmented by adding an additional—typically evening—exchange, so increasing the dose of dialysis given. These volumes will be very different for children in whom they are adjusted according to body surface area.

The patient or their carer is trained in the sterile procedure of dialysis fluid exchange by a specialist peritoneal dialysis nurse, a process that generally takes a few days and can be achieved by a wide variety of individuals, despite physical and educational disabilities.

Peritoneal access

In order to drain peritoneal dialysis fluid in and out of the peritoneal cavity, a flexible permanent dialysis catheter is required, and this is also to be preferred when peritoneal dialysis is used for AKI. Correct insertion and subsequent catheter management is critical for the success of peritoneal dialysis. Insertion should be performed as a planned procedure by an experienced operator and may be achieved using either the Seldinger technique, open surgery, or with the use of a laparoscope. Catheters usually have two Dacron cuffs, one at each end of the subcutaneous tunnel: the deep cuff is placed at entry to the peritoneum, whereas the superficial cuff should be situated about 1.5 cm deep to the exit site, which should be positioned on the abdominal wall so that it is visible to the patient. Up-to-date recommendations for peritoneal dialysis catheter insertion (also for the management of peritonitis, see later) are freely available at the International Society for Peritoneal Dialysis website (

Peritoneal physiology

The two main functions of dialysis, solute and fluid removal, occur owing to the contact between dialysis fluid and the capillary circulation of the parietal and visceral peritoneum. In a typical adult, approximately 0.5 m2 of the peritoneum is in contact with fluid, representing about one-third of the anatomical membrane.

Solute clearance

Removal of solute is predominantly by diffusion across the membrane and is thus governed by the concentration gradient, the number and density of capillaries in contact with dialysate (often termed effective peritoneal surface area), and the size of the solute in question. Equilibration time for urea is typically within 4 to 5 h, hence removal is limited by the drained dialysate volume rather than length of dwell. Equilibration is more time-limited for larger molecules such as creatinine and glucose, e.g. the average dialysate-to-plasma ratio of creatinine at 4 h is 0.65, with a range of 0.4 to 0.85 that is normally distributed. This measurement is termed the solute transport rate and is used to measure the diffusive component of membrane function in the ‘peritoneal equilibration test’. There is considerable variability between patients, which appears to be related to local production of the inflammatory cytokine interleukin-6 rather than clinical variables such as body size or comorbidity. Whereas systemic inflammation is as independent predictor of patient survival, the local inflammatory response is not, provided appropriate dialysis regimes are prescribed.


Fluid transport across the membrane is governed mainly by pressure gradients, including hydrostatic, osmotic, and oncotic, as well as some fluid reabsorption due to lymphatics. Instillation of fluid within the peritoneal cavity creates a positive pressure that is close to capillary pressure, resulting in little net fluid movement by this mechanism unless excessive intraperitoneal pressures are created. Conventional glucose-containing dialysis fluids create an osmotic pressure gradient proportional to glucose concentration (1.5–4.25%) that results in a peak net ultrafiltration volume between 2 and 4 h, depending on solute transport characteristics of the membrane (Fig. It is by prescription of exchanges containing different glucose concentrations (‘weak’, ‘medium’, and ‘strong’ bags) that the patient’s fluid status can be controlled to achieve the target weight specified by their physician. There are two pathways of water transport by this mechanism, intercellular and transcellular, each contributing about half the fluid removal, the latter being via aquaporins and thus water-exclusive. Once the osmotic gradient has dissipated, fluid reabsorption occurs due to a combination of transcapillary oncotic (Starling) forces and lymphatics. Removal of ions such as sodium and calcium is predominantly by convection because the diffusion gradient is small (typical dialysate sodium concentration is 132 mmol/litre). Owing to the existence of the aquaporin pathway, there is uncoupling of water and sodium removal, termed sodium sieving, that is most pronounced in the early part of the dwell. As a consequence, peritoneal dialysis is always at risk of removing excess water compared with sodium, and treatment regimens that rely on too many short, hypertonic glucose exchanges are to be avoided. By the same token, lack of sodium sieving is an indicator of poor free-water transport and thus membrane failure.

Fig. Pathways of fluid transport across the peritoneal membrane. The net changes in intraperitoneal fluid volume and thus ultrafiltration achieved at each time point when using glucose (3.86%) as the osmotic agent are the result of fluid transport via at least three pathways: aquaporins (water-exclusive), small intercellular pores, and lymphatic reabsorption. Fluid is reabsorbed once the intraperitoneal glucose has equilibrated with plasma. By contrast, when using the glucose polymer icodextrin, sustained ultrafiltration occurs for several hours.

Pathways of fluid transport across the peritoneal membrane. The net changes in intraperitoneal fluid volume and thus ultrafiltration achieved at each time point when using glucose (3.86%) as the osmotic agent are the result of fluid transport via at least three pathways: aquaporins (water-exclusive), small intercellular pores, and lymphatic reabsorption. Fluid is reabsorbed once the intraperitoneal glucose has equilibrated with plasma. By contrast, when using the glucose polymer icodextrin, sustained ultrafiltration occurs for several hours.

Monitoring treatment quality

Solute clearance

Residual renal and peritoneal solute clearances are not equivalent in their impact on patient survival. To date, in CKD, no randomized study has been able to demonstrate an association between survival and peritoneal clearances, and there are no studies published in AKI. It is important, therefore, that both residual and peritoneal clearances are measured. Both urea (Kt/Vurea) and creatinine clearances can be used to monitor treatment, and there is general agreement in international guidelines—based on at least two large randomized trials—on minimum treatment targets. For Kt/Vurea, a combined residual and peritoneal clearance of 1.7 using the Watson formula to calculate V (total body water) should be achieved. Alternatively, a combined creatinine clearance of 50 litres/week per 1.73 m2 can be used. In either case, however, dialysis dose should be increased in the presence of symptoms attributable to uraemia.


Several observational studies have shown that a reduced peritoneal ultrafiltration and by implication sodium removal is associated with reduced survival, especially in anuric patients. Several reasons for this association have been proposed, but there is little doubt that a failure to remove adequate fluid owing to membrane characteristics that result in less good ultrafiltration or excessive fluid reabsorption are important. The simplest way to monitor membrane function is to perform a regular peritoneal equilibration test that measures solute transport rate and net ultrafiltration capacity in a standardized 4-h dwell. A membrane that results in less than 400 ml ultrafiltration using 3.86% or less than 100 ml using 2.27% glucose is unlikely to achieve adequate fluid removal once the patient becomes anuric. If this is associated with a fast rate of solute transport (4-h dialysate:plasma creatinine ratio above 0.65), and thus early loss of the glucose gradient and more rapid fluid reabsorption, then a combination of APD delivering shorter exchanges overnight and icodextrin (a polydispersed glucose polymer derived from starch) in the long exchange will result in adequate fluid removal in most cases. Use of APD in this situation is associated with better patient survival, mitigating the observed increased mortality seen in patients with rapid transport membranes when using CAPD. Icodextrin achieves sustained ultrafiltration in the long dwell as it acts rather like a colloidal agent, thus counterbalancing Starling forces. It has been shown to improve fluid status, reduce peritoneal glucose exposure, and enhance diabetic control, and it represents a major advance in peritoneal dialysis therapy. Note, however, that icodextrin metabolites in blood can interfere with estimation of blood glucose by monitors that use glucose dehydrogenase, such that there is a risk of failing to diagnose hypoglycaemia.

Complications and their management


Peritonitis is the commonest and potentially serious complication of peritoneal dialysis, resulting in up to 50% of technique failures. With enhanced training, patients can be expected to have one episode of peritonitis for every 3 years of treatment. Presentation is with cloudy dialysis effluent, with or without abdominal pain and/or fever, and the diagnosis is confirmed if the leucocyte count is above 100/µl in the peritoneal fluid. Clinical assessment should include examination of the exit site and tunnel for infection, and the abdomen for signs of intra-abdominal pathology. At least 20 ml of freshly drained dialysate should be sent for culture by appropriate methods, which are absolutely crucial. The technique recommended by the International Society for Peritoneal Dialysis (freely accessible at is for 50 ml of peritoneal effluent to be centrifuged at 3000 g for 15 min, followed by resuspension of the sediment in 3 to 5 ml of sterile saline and inoculation of material on both solid culture media and into a standard blood culture medium. With this method, fewer than 5% of cases should be culture-negative.

Empirical antibiotic treatment, either intraperitoneal or systemic, with cover for both Gram-positive and Gram-negative organisms should be commenced immediately while awaiting specific cultures and sensitivities. The reader is again directed to the International Society for Peritoneal Dialysis website for detailed up-to-date guidelines on management, but most typically Gram-positive organisms will be covered by vancomycin or a cephalosporin, and Gram-negative infections by a third-generation cephalosporin, aminoglycoside, or (if local sensitivities support such use) quinolone, with the regimen adjusted as and when results of culture and sensitivity of peritoneal fluid are available. A mixed bacterial growth, especially associated with anaerobes, should dictate early surgical assessment, catheter removal, and laparotomy. Catheter removal is also likely to be necessary if there is an associated catheter-tunnel infection or a failure to respond to antibiotics within a few days.

Exit-site infection

A simple scoring system has been devised to assess the exit site (Table

Table Classification and treatment of exit-site infections


Points awarded






At exit

Includes tunnel



<0.5 cm

>0.5 cm



<0.5 cm

>0.5 cm









An exit site with a score of 4 or greater, or a purulent discharge, should be treated with empirical antibiotics that cover coagulase-positive Staphylococcus and Pseudomonas until sensitivities are available.

Mechanical failure

Mechanical failure is common in the early stages of a patient’s career on peritoneal dialysis, and can be due either to catheter displacement (outflow affected), wrapping of the catheter by the omentum (in- and outflow), or leakage. The latter can occur at the deep cuff or at any hernia site and may present as genital swelling. Diagnosis is either obvious or requires appropriate imaging, usually a CT scan with contrast introduced into the catheter.

Ultrafiltration failure

Changes can occur in membrane function with time on peritoneal dialysis that result in less good ultrafiltration. Drivers of this change appear to be early loss of residual renal function and use of more hypertonic glucose dialysis solutions. Most frequently, this is due to progressive increase in solute transport that can be addressed using APD and icodextrin (see above). In some patients, however, more serious membrane damage occurs which results in reduced osmotic conductance (efficiency) of the membrane for a given osmotic gradient due to progressive membrane fibrosis. These patients require transfer to haemodialysis.

Encapsulating peritoneal sclerosis

Encapsulating peritoneal sclerosis is a relatively uncommon but life-threatening complication of peritoneal dialysis, characterized by severe inflammatory thickening, especially of the visceral peritoneum, resulting in an encapsulation and progressive obstruction of the bowel.

Diagnosis is made from CT scan or laparotomy confirmation of bowel obstruction due to macroscopic appearances of thickened membrane, with cocoon-like encapsulation of the entire intestine in severe cases. Risk factors identified include prolonged time on treatment (incidence 15–20% after 10 years), severe and protracted peritonitis, and acquired severe ultrafiltration failure characterized by loss of membrane efficiency and in particular reduced free water transport.

Encapsulating peritoneal sclerosis (EPS) may resolve slowly after stopping peritoneal dialysis, but often continues to progress and presents after modality transfer, including after renal transplantation. Corticosteroids and immunosuppressants (both anti-inflammatory) and tamoxifen (anti-fibrotic) have been used as treatments, but without convincing evidence of benefit. Management includes nutritional support, parenteral feeding in the presence of vomiting, and in severe cases—especially with life-threatening obstruction—extensive surgical enterolysis. This should be performed in specialized centres which report good primary cure rates when the treatment is planned, although recurrence can require repeat procedures. Emergency surgery, especially following bowel perforation or infarction, has a very high mortality.

EPS is an awful disease; case series report an overall mortality of 26 to 58%, although many deaths within such series were not from EPS itself. Because of the risk of the condition, there has been much discussion in the renal community as to whether patients should be advised against remaining on peritoneal dialysis for more than a particular number of years. However, the risk of developing EPS is extremely low in patients who have been on peritoneal dialysis for less than 5 years; most long-term peritoneal dialysis patients are not affected, and the risks of EPS must be balanced against those of any alternative renal replacement therapy for the individual patient. Furthermore, there are no prospective data supporting a benefit of pre-emptively transferring long-term peritoneal dialysis patients to haemodialysis.

Further reading

Ballinger AE, et al. (2014) Treatment for peritoneal dialysis-associated peritonitis. Cochrane Database Syst Rev, 4, CD005284.Find this resource:

Brimble KS, et al. (2006). Meta-analysis: peritoneal membrane transport, mortality, and technique failure in peritoneal dialysis. J Am Soc Nephrol, 17, 2591–8.Find this resource:

Brown EA, et al. (2003). Survival of functionally anuric patients on automated peritoneal dialysis: the European APD Outcome Study. J Am Soc Nephrol, 14, 2948–57.Find this resource:

Cho Y, et al. (2014). Biocompatible dialysis fluids for peritoneal dialysis. Cochrane Database Syst Rev, 3, CD007554.Find this resource:

Crabtree JH, Burchette RJ, Siddiqi NA (2005). Optimal peritoneal dialysis catheter type and exit site location: an anthropometric analysis. ASAIO J, 51, 743–7.Find this resource:

Cullis B, et al. (2014). Peritoneal dialysis for acute kidney injury. Perit Dial Int, 34, 494–517.Find this resource:

Davies SJ (2010). Peritoneal dialysis solutions. In: Himmelfarb J, Sayegh MH (eds). Chronic kidney disease, dialysis and transplantation – companion to Brenner and Rector’s. The kidney. 3rd edition. pp. 417–31. Elsevier Saunders, Philadelphia.Find this resource:

    Davies SJ (2013). Peritoneal dialysis—current status and future challenges. Nat Rev Nephrol, 9, 399–408.Find this resource:

    Jager KJ, et al. (2004). The effect of contraindications and patient preference on dialysis modality selection in ESRD patients in The Netherlands. Am J Kidney Dis, 43, 891–9.Find this resource:

    Johnson DW, et al. (2012). Effects of biocompatible versus standard fluid on peritoneal dialysis outcomes. J Am Soc Nephrol, 23, 1097–107Find this resource:

    Lambie M, et al. (2013). Independent effects of systemic and peritoneal inflammation on peritoneal dialysis survival. J Am Soc Nephrol, 24, 2071–80.Find this resource:

    Li PK, et al. (2010). Peritoneal dialysis-related infections. Recommendations: 2010 update. Perit Dial Int, 30, 393–423.Find this resource:

    Mehta R, et al. (2015). International Society of Nephrology’s 0by25 initiative for acute kidney injury: a human rights case for nephrology. Lancet, 385, 2616–43.Find this resource:

    Paniagua R, et al. (2002). Effects of increased peritoneal clearances on mortality rates in peritoneal dialysis: ADEMEX, a prospective, randomized, controlled trial. J Am Soc Nephrol, 13, 1307–20.Find this resource:

    Piraino B, et al. (2011). ISPD position statement on preventing the risks of peritoneal dialysis related infections. Perit Dial Int, 31, 614–30.Find this resource:

    Rippe B, et al. (2004). Fluid and electrolyte transport across the peritoneal membrane during CAPD according to the three-pore model. Perit Dial Int, 24, 10–27.Find this resource:

    van de Luijtgaarden MW, et al. (2013). Global differences in dialysis modality mix: the role of patient characteristics, macroeconomics and renal service indicators. Nephrol Dial Transplant, 28, 1264–75.Find this resource:

    Weinhandl ED, et al. (2010). Propensity-matched mortality comparison of incident hemodialysis and peritoneal dialysis patients. J Am Soc Nephrol, 21, 499–506.Find this resource: