An official website of the United States government
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Peritoneal Dialysis: Peritoneal Dialysis in the Treatment of Stage 5 Chronic Kidney Disease. London: National Institute for Health and Clinical Excellence (NICE); 2011 Jul. (NICE Clinical Guidelines, No. 125.)
‘Peritoneal dialysis: peritoneal dialysis in the treatment of stage 5 chronic kidney disease’ (NICE clinical guideline 125) is a NICE short clinical guideline. For a full explanation of how this type of guideline is developed, see ‘The guidelines manual’ (2009) at www.nice.org.uk/GuidelinesManual.
The Guideline Development Group agreed the following outcomes as relevant, and the importance of each outcome was agreed through informal consensus.
health-related quality of life
patient involvement and satisfaction
mortality (where reported, including deaths in the first 3 months of treatment)
preservation of renal function
technique failure or switch
resource use and costs including hospitalisation.
adverse events
adequacy rates
staff attitude and skills
nutritional status.
anaemia.
The minimally important difference was agreed as a 10% difference in relative risk.
What are the barriers and facilitators for peritoneal dialysis for:
Through understanding the barriers and facilitators to the use of peritoneal dialysis, the Guideline Development Group was able to draft recommendations on the information and support needs of patients and their families and carers to improve decision-making.
A total of 6183 articles were found by systematic searches, a further six systematic reviews were suggested by the Guideline Development Group (four of which had not been identified in the searches – Furr 1998; Kaptein et al. 2009; Morton et al. 2010b; Murray et al. 2009) and one further review was identified through background searching (Mason et al. 2008). Full text was ordered for 168 articles based on the title and abstract. Thirteen papers (Bass et al. 1999; de Paula et al. 2008a; de Paula et al. 2008b; Hislop and Lansing 1983; Lee et al. 2008; Mason et al. 2008; McLaughlin et al. 2008; Morton et al. 2010a; Morton et al. 2010b; Murray et al. 2009; Oliver et al. 2010; Portoles et al. 2009; Tong et al. 2008) met the eligibility criteria (see appendix A for review protocol and inclusion and exclusion criteria) and described the experience of decision-making when starting dialysis. Although many studies reported factors associated with the choice of dialysis, they were excluded because the focus was the experience of the patient, family or healthcare professional as explored using qualitative methods. However, some surveys evaluating patient-centred factors associated with the choice of dialysis are included because the Guideline Development Group considered that they contributed important information.
GRADE assessment was adapted, and the following variables were considered: limitations, inconsistency and indirectness. Imprecision was rated as not relevant throughout because it does not apply to the type of evidence considered in this question. The following principles were applied to assess quality: a systematic review of qualitative studies started as high, and a single qualitative study started as moderate, with downgrading as appropriate.
The evidence base describing barriers and facilitators for adults and their families or carers included two systematic reviews (Morton et al. 2010b; Murray et al. 2009) and five primary studies. Three of the primary studies were directly relevant (Lee et al. 2008; McLaughlin et al. 2008; Morton et al. 2010a), Two studies were identified as supporting evidence (Oliver et al. 2010; Portoles et al. 2009), but the GDG decided not to consider them any further.
Only the evidence considered to be directly relevant is summarised in the GRADE tables.
The evidence base describing barriers and facilitators for children and their families or carers included one systematic review of the experiences of parents (Tong et al. 2008) and two primary studies (one with reports in two papers) (de Paula et al. 2008a; de Paula et al. 2008b; Hislop and Lansing 1983).
GRADE profile 2. Patient and carer perspective (children)
GRADE profile 3. Patient and carer perspective (parents)
GRADE profile 4. Patient and family perspective (parents and other family members of children)
The evidence base for the perceptions of healthcare professionals on barriers and facilitators to peritoneal dialysis use included one primary study on the experience of healthcare professionals supporting or providing care to people who need dialysis (Bass et al. 1999).
High-quality evidence (from one systematic review) found that the most frequent decisions cited in the literature were about the type of renal replacement therapy, withholding or withdrawing treatment, and renal transplantation (Murray et al. 2009).
High-quality evidence (from one systematic review) showed that confronting mortality, lack of choice, gaining knowledge about options, and weighing alternatives were important themes for adult patients and carers when making decisions about treatment (Morton et al. 2010b).
High-quality evidence (from one systematic review) found that patients wanted information on managing their condition, social and lifestyle factors, general knowledge about CKD, treatment options and renal transplant, lifestyle management, self care, and end-of-life planning. Patients were greatly influenced by exposure to others’ (patient peers) opinions and experiences (Murray et al. 2009).
High-quality evidence (from one systematic review) identified the following as patient perceptions that influence treatment decisions: interpersonal relationships, trust in providers, preserving current wellbeing, normality and quality of life, the need for control, and being personally responsible (Murray et al. 2009).
High-quality evidence (from one systematic review) showed that major factors influencing decision-making by patients and carers were peer influence, the timing of information and duration over which it’s given, and maintaining the status quo (Morton et al. 2010b).
Low-quality evidence (from three qualitative studies) showed that issues for patients, families and carers were flexibility and independence or confinement, freedom and control or risk and subordination, self care, feelings of insecurity or security, physical space and noise in the location of treatment delivery, home modification, maintaining a normal life, lifestyle, social inclusion or family burden, convenience or time commitment, pre-dialysis information, involvement in choosing dialysis, effectiveness, simplicity, pain, access for dialysis, and impermanence of the treatment and access (Lee et al. 2008; McLaughlin et al. 2008; Morton et al. 2010a).
Very-low-quality evidence (from one qualitative study) found that children preferred different types of dialysis because of the impact on normal activities (Hislop and Lansing 1983).
High-quality evidence (from one systematic review) showed that self-accusation and blame, depression and generalised anxiety, uncertainty around diagnosis, uncertainty around the child’s future, emotional turmoil, uncertainty around prognosis and surrendering control of the child to clinical staff were issues at pre-dialysis and dialysis for parents of children and young people with CKD (Tong et al. 2008).
Very-low-quality evidence (from one qualitative study) found that families, including siblings, reported changes and adaptations when the child was treated with peritoneal dialysis. Social supports were very meaningful and were varied: type of supports were categorised as emotional, informational, instrumental, and appraisal. Such support was accessed through many and diverse sources, but it was generally thought that it should be accessible, and provide a ‘depth’ of support (de Paula et al. 2008a; de Paula et al. 2008b).
Very-low-quality evidence (from one qualitative study) showed that healthcare professionals perceived the following to be factors related to decision-making for dialysis (Bass et al. 1999):
This was not considered to be a health economic question.
For most people the choice of dialysis is not clinically driven but is related to individual characteristics and preferences and the impact on their own and their families’ lives. Kidney disease is a lifelong condition with varying modes of treatment, including dialysis and/or transplantation, with the option of conservative management if appropriate.
The disease is never ‘cured’, it is merely treated with these different therapies. It is therefore imperative that the choices offered and made are in line with what the patient needs and wants.
The Guideline Development Group noted that the decision about renal replacement therapy requires patients and professionals to consider this potential change to the patient’s treatment regimen in advance of a deterioration in the patient’s health. This requires effort by all parties to consider and discuss these treatment options over and above routine care practices, and to recognise that how the patient currently manages the impact of CKD on their life will change.
The evidence suggests that in most cases the dialysis choice depends more on how the dialysis treatment will fit into the patient’s life rather than on clinical indicators. However, there are some patient groups for whom clinical considerations do have an impact on the choice of modality, and these are discussed below.
Overall, the Guideline Development Group noted the consistency of the evidence, particularly for the experience of adults when choosing the modality of dialysis. The factors identified as being important to patients were also consistent with the experience and perceptions of the Guideline Development Group.
However, there was significantly less evidence on the experience of children and their parents and families. However, the Guideline Development Group considered that they could use experience from other areas of paediatric care to support the recommendations on decision-making in this group.
The Guideline Development Group agreed that information should be given proactively. Choosing dialysis is difficult and people, and their families, need help to collate all relevant information and tailor it to their needs. Supporting informed decision-making is not simply about providing information; if people are not able to use the information meaningfully, providing or repeating information alone will not help.
The approach to providing information to support informed decision-making should be structured and timely, including for those patients who have presented late or started dialysis treatment urgently. This is consistent with other initiatives to support people with kidney disease, for example the use of kidney care plans. The NICE guidelines on chronic kidney disease (www.nice.org.uk/guidance/CG73) also make recommendations on tailored education and information to be provided at appropriate times during the patient’s journey.
There are significant challenges in providing information about choosing dialysis; there are many important factors and it can be difficult to understand how these will affect a person’s daily life. The Guideline Development Group considered that these challenges could be addressed by encouraging the patient to think actively about those aspects of life that will change or stay the same as a result of the treatment choices. Also, because dialysis can be a long-term intervention and circumstances may change over time, a patient’s values may also change, so information, evaluation of that information and the decisions about dialysis should be an ongoing process. This is also relevant to starting dialysis and information should be provided before dialysis is needed; that is, as part of the pre-dialysis education programmes.
Another key principle underpinning the recommendations is the need to give people enough information for them to make the right decision for them and their families, rather than the healthcare professional ruling out options before a full discussion of all appropriate choices. Healthcare professionals may have different perceptions of what is important, and these may be valid (for example, if based on clinical factors). But this difference needs to be acknowledged and understood so that an informed decision can be reached.
The Guideline Development Group therefore made recommendations on the information and support that patients need to facilitate informed decision-making.
What is the effectiveness of interventions (specifically education, decision-support tools and aids) to improve decisions about the initial choice of dialysis?
A total of 6183 articles were found by systematic searches, a further six systematic reviews were suggested by the Guideline Development Group (four of which were not identified in the searches: Furr 1998; Kaptein et al. 2009; Morton et al. 2010b; Murray et al. 2009) and one further review was identified through background searching (Mason et al. 2008). Full text was ordered for 168 articles based on the title and abstract. Two papers (Keshaviah 1997; Murray et al. 2009) evaluated the effectiveness of interventions to improve decision-making when starting dialysis. The following studies did not meet the eligibility criteria (see appendix 1 for review protocol and inclusion and exclusion criteria) and were not considered further: Furr (1998), Kaptein et al. (2009) and Keshaviah (1997).
Although the systematic reviews looked at interventions to improve decisions throughout the patient journey of CKD, only those studies that focused on choosing dialysis were considered for this review.
GRADE profile 6. Interventions to improve decision-making on choosing dialysis
No other outcomes were reported.
Low-quality evidence (from one systematic review) showed no difference in measures of depression and anxiety between adults who had undergone pre-dialysis education and those in the control group (Mason et al. 2008).
Low-quality evidence (from two systematic reviews) showed that efficacy for performing self care was increased in adults who had undergone pre-dialysis education compared with those in the control group (Mason et al. 2008). However, there was no increase in the level of social support between groups (Mason et al. 2008), and no studies reported the impact of decision-support interventions on measures of decision quality (Mason et al. 2008; Murray et al. 2009).
Low-quality evidence (from one systematic review) showed that the median survival (after starting dialysis and at 20 years’ follow-up) was increased in adults who had undergone pre-dialysis education compared with those in the control group (Mason et al. 2008).
No studies were identified on the impact of pre-dialysis education or decision-support interventions on the other critical outcomes preserving renal function, technique failure or switch, resource use and costs including hospitalisation, or on the important outcomes of adverse events, adequacy rates, nutritional status or staff attitude and skills.
No studies were identified on the impact of interventions to improve decision-making in children and/or their families.
This was not considered to be a health economic question.
The Guideline Development Group noted the lack of evidence on interventions to improve decision-making for dialysis; however, it recognised that there is a large body of evidence on decision-making in general. The recommendations are therefore based on the reviewed evidence, specific to this topic, and the GDG’s knowledge of generic decision-support interventions that were not reviewed in this guideline.
Specific interventions considered by the Guideline Development Group included play therapy for children and peer support. However, no evidence on them was identified. It was agreed that children need decision-making methods suitable for their age. Peer information giving was discussed in detail. Again, although there was no evidence, the Guideline Development Group considered that structured information from peers in a formal programme could be effective in improving patients’ understanding of the impact of options. But if managed poorly it could negatively affect the decision-making process. If peer information is used, healthcare professionals should discuss this information with the patient and family and explore how the information has influenced their decision-making.
Offer patients with stage 5 chronic kidney disease (CKD) and their families and carers information and support in line with ‘Chronic kidney disease’ (NICE clinical guideline 73, 2008).
Offer patients and their families and carers oral and written information about pre-emptive transplant, dialysis, and conservative care to allow them to make informed decisions about their treatment.
To enable patients to make informed decisions, offer balanced and accurate information about all dialysis options. The information should include:
efficacy
risks
potential benefits, based on the person’s prognosis
potential side effects and their severity
changing the modality of dialysis and the possible consequences (that is, the impact on the person’s life or how this may affect future treatment or outcomes)
the patient’s and/or carer’s ability to carry out and adjust the treatment themselves
integration with daily activities such as work, school, hobbies, family commitments and travel for work or leisure
opportunities to maintain social interaction
the impact on body image
how the dialysis access point on the body may restrict physical activity
if their home will need to be modified to accommodate treatment
distance and time spent travelling for treatment
flexibility of treatment regimen
any additional support or services that might be needed from others.
Explain to patients and check they understand that CKD is a lifelong disease, and that during the course of renal replacement therapy they are likely to need to switch between treatment modalities depending on clinical or personal circumstances.
When providing information about treatment options, healthcare professionals should discuss and take into account any information the patient has obtained from other patients, families and carers and all other sources, and how this information has influenced their decision.
Make sure that healthcare professionals offering information have specialist knowledge about CKD and the necessary skills to support decision-making. This may include training in:
Trained healthcare professionals (see recommendation 1.1.6) should be available to discuss the information provided both before and after the start of dialysis.
Offer all patients who have presented late or started dialysis treatment urgently an enhanced programme of information, at an appropriate time, that offers the same information and choices as those who present at an earlier stage of chronic kidney disease.
What is the effectiveness of peritoneal dialysis compared with haemodialysis in people with stage 5 CKD who need dialysis?
A total of 5149 articles were found by systematic searches. Full text was ordered for 320 articles (comparing peritoneal dialysis with any other type of dialysis, including haemodialysis) based on the title and abstract. Only one primary study (Korevaar et al. 2003) met the eligibility criteria (for review protocol and inclusion and exclusion criteria, see appendix A) and evaluated the effectiveness of peritoneal dialysis compared with haemodialysis for adults with stage 5 CKD. A Cochrane systematic review was found that compared peritoneal dialysis with haemodialysis (Vale et al. 2004) and this included the one study identified in our reviews, so only the results from the primary study are presented.
No randomised controlled trial (RCT) evidence was found for children with stage 5 CKD.
Because of the lack of RCT evidence, the Guideline Development Group asked the technical team to search for publications from national renal registries that would provide further information on the outcomes agreed important for this guideline.
Systematic searches for registry data reported in published articles found 1672 articles. Full text was ordered for 261 articles based on the title and abstract. In addition, we looked for annual reports for those not identified by the searches; this included the 2008 annual report of the North American Pediatric Renal Trials and Collaborative Studies1. Of these publications, 53 papers met the eligibility criteria for registry publications; 42 evaluated the effectiveness of peritoneal dialysis or haemodialysis, either with comparative analyses or single intervention reports for adults and children. We also re-checked studies identified for other questions (for example, switching) and included those that were relevant.
Table 1 summarises the included studies. See appendix A for more details.
Summary of studies of registry data for peritoneal dialysis and haemodialysis.
Of the 42 studies, 11 included children (31 were adults alone); 25 reported outcomes for both peritoneal dialysis and haemodialysis, with 14 on peritoneal dialysis alone and 3 on haemodialysis alone.
For the GRADE assessment, registry studies were assessed as low-quality evidence, with downgrading as appropriate.
GRADE profile 13. Effectiveness of peritoneal dialysis – mortality – PD (adults and children)
GRADE profile 14. Effectiveness of peritoneal dialysis – mortality – PD (adults)
GRADE profile 15. Effectiveness of peritoneal dialysis – mortality – PD (children)
GRADE profile 16. Effectiveness of peritoneal dialysis – mortality – HD (adults and children)
GRADE profile 24. Effectiveness of peritoneal dialysis – technique failure or switch (adults)
GRADE profile 25. Effectiveness of peritoneal dialysis – technique failure or switch (children)
GRADE profile 28. Effectiveness of peritoneal dialysis – resource use and costs (adults)
GRADE profile 29. Effectiveness of peritoneal dialysis – resource use and costs (children)
GRADE profile 34. Effectiveness of peritoneal dialysis – adverse events (adults and children)
GRADE profile 35. Effectiveness of peritoneal dialysis – adverse events (adults)
GRADE profile 36. Effectiveness of peritoneal dialysis – adverse events (children)
GRADE profile 41. Effectiveness of peritoneal dialysis – adequacy of dialysis (adults and children)
GRADE profile 42. Effectiveness of peritoneal dialysis – adequacy of dialysis (adults)
There is no consistent evidence on the impact of different types of dialysis modalities (peritoneal dialysis or haemodialysis) on health-related quality of life.
Very-low-quality evidence (from one RCT) showed no difference in health-related quality of life in adults undergoing peritoneal dialysis (CAPD and APD) compared with those on haemodialysis (Korevaar et al. 2003). However, very-low-quality evidence from one national registry data showed that adults on CAPD tended to have a higher quality of life than those on haemodialysis (Lim et al. 2008).
No evidence on health-related quality of life in children was identified.
No evidence on patient involvement and satisfaction was identified.
Very-low-quality evidence (from one RCT and 17 national registries) showed a tendency for no difference in long-term mortality in adults and children undergoing peritoneal dialysis compared with those on haemodialysis. Where differences were seen (particularly in short-term mortality), they were associated with patient factors.
Survival on peritoneal dialysis tended to be better for adults younger than 55 years and for those without diabetes or other comorbidities, such as congestive heart failure (very-low-quality evidence from seven studies).
Low-quality evidence (from 12 national registries) showed similar rates of peritonitis-related deaths. Encapsulating peritoneal sclerosis (EPS) was very rare, but the incidence did increase over time.
No evidence on preservation of renal function was identified.
Very-low-quality evidence (from one RCT and 20 national registries) showed that rates of dialysis failure or switch did not seem to differ between peritoneal dialysis and haemodialysis. However, reasons for switching differed by type of dialysis (for example as a result of different complications).
Very-low-quality evidence (from six national registries) showed that rates of hospital admissions did not appear to differ between peritoneal dialysis and haemodialysis. However, reasons for admissions differed by type of dialysis.
Very-low-quality evidence (from 18 national registries) showed that rates of adverse events did not appear to differ between peritoneal dialysis and haemodialysis. However, the type of adverse events differed by type of dialysis.
Adequacy rates were rarely reported in RCTs; when reported, rates were only for people on haemodialysis. The evidence that was available (very-low-quality evidence from six national registries) showed no clear difference between peritoneal dialysis and haemodialysis.
No evidence on staff attitude and skills was identified.
Low-quality evidence (from one national registry) showed no significant difference in mean phosphate levels between people on peritoneal dialysis and those on haemodialysis (Dawnay et al. 2010).
Very-low-quality evidence (from three national registries) showed that erythropoietin use tended to be higher in people on haemodialysis, although this difference between types of dialysis may disappear over time.
This question was considered a priority for health economic work. Instead of a straight comparison between peritoneal dialysis and haemodialysis, the analysis focused on comparing different service provisions between peritoneal dialysis and haemodialysis. A search was made for any appropriate studies. Five studies were identified that were considered suitable. These are outlined in the health economic appendix 2.
The paper by Kirby et al. (1996) indicates that haemodialysis was the most cost-effective treatment option (lower costs and more effective). However, the paper is relatively old and costing appears to be outdated. The other papers indicate that increasing the use of peritoneal dialysis is more cost effective. However, these papers are not UK based.
Therefore, a new economic analysis was produced. The model was based on previous models, specifically Kirby et al. (1996). A diagram of the model is presented in figure 1.
Model diagram.
Transition probabilities were derived from the clinical review and from the renal registry and NHS blood and transplant report 2009. Cost data were obtained from Baboolal et al. (2009) and utilities from a meta-analysis by Liem et al. (2008). Costs from the payment by results working party were not used because they did not include additional costs such as treatment for adverse events. Full details are provided in appendix B.
The model examines several scenarios comparing hypothetical services providing a range of proportions of haemodialysis and peritoneal dialysis. The baseline is based on the average proportions of people starting dialysis from the renal registry report 2009.
Subset analysis 1 examines hypothetical centres based on amalgams of NHS trusts that are focused on providing one type of dialysis. Subset analysis 2 examines percentage increases over the baseline and subset analysis 3 examines absolute numbers starting on a treatment modality. The scenarios are outlined in table 2
Scenarios examined in health economic analysis.
The deterministic and probabilistic results over 10 years are presented in tables 3 and 4.
Base-case deterministic results.
Base-case probabilistic results.
Table 5 outlines the results of cost effectiveness acceptability curves at £20,000 and £30,000 per quality-adjusted life year (QALY) thresholds.
Cost-effectiveness acceptability results at £20,000 and £30,000 per QALY gained thresholds.
These results indicate that increasing the use of peritoneal dialysis is a cost-effective policy. Sensitivity analysis indicates that this is robust to changes in utilities, costs and assuming equivalence. The full sensitivity analysis is presented in appendix 2. Of note are the scatter plots of the simulations from the probabilistic analysis. One is reproduced in figure 2.
Scatter plot for starting 70% of people on peritoneal dialysis.
The simulations are spread across the cost-effectiveness plane, most points are below the horizontal axis, suggesting it is likely to be cost saving. However, there is significant variation in terms of effectiveness. Therefore, to reduce uncertainty in the decision peritoneal dialysis should be targeted at those who could benefit most.
The Guideline Development Group agreed that overall there was no evidence of significant differences between the modalities for the critical outcomes; therefore recommendations were made allowing all patients the option of either peritoneal or haemodialysis. However, some patient factors (such as the absence of significant comorbidities) were associated with a short-term survival benefit with peritoneal dialysis. The Guideline Development Group therefore made recommendations, based on the evidence and their clinical expertise, on those patient groups in which peritoneal dialysis was likely to be the preferred option.
The Guideline Development Group noted the lack of evidence of the effect of dialysis modality on the preservation of renal function. Although this outcome was not reported in the reviewed evidence, the Guideline Development Group considered that the general clinical view is that residual renal function is preserved more effectively in peritoneal dialysis than in haemodialysis (Misra, V 2001; Moist 2000; Rottembourg et al. 1983). A recommendation was therefore made on the preferred use of peritoneal dialysis as the initial treatment modality in people with residual renal function.
The Guideline Development Group considered the health economic analysis that had been conducted. It noted that it was based on very poor data of clinical effectiveness and could not identify clinical data to differentiate between peritoneal or haemodialysis. It noted that the cost-effectiveness results indicate that increasing the proportion of people who start on peritoneal dialysis would result in cost savings to the NHS, with potential minor increases in health related quality of life. It noted from the sensitivity analyses that the results were robust to changes to survival estimates, quality of life estimates and alternative costing. The Guideline Development Group also noted that probabilistic sensitivity analysis suggested that the conclusions were associated with probabilities of being cost effective of more than 50% and close to 90%.
The Guideline Development Group noted that the cost savings were greater with CAPD than APD. However, it was mindful that the model could not differentiate between CAPD and APD in terms of clinical effectiveness and quality of life. Therefore, it was not possible to recommend one over the other. The Guideline Development Group concluded that, given the available information, increasing the use of peritoneal dialysis would be a cost-effective use of NHS resources.
The Guideline Development Group acknowledged that the analysis did not consider variation in home haemodialysis because of the absence during guideline development of appropriate data on effectiveness and costs. It was aware of potential future developments in technology, especially for home haemodialysis. It concluded that this should be an area for future work. Therefore, based on this evidence and the related evidence on decision-making, the Guideline Development Group made the following recommendations on the use of peritoneal dialysis for people with stage 5 CKD.
Offer all people with stage 5 CKD a choice of peritoneal dialysis or haemodialysis, if appropriate, but consider peritoneal dialysis as the first choice of treatment modality for:
When discussing choice of treatment modalities, healthcare professionals should take into account that people’s priorities are not necessarily the same as their own clinical priorities.
What is the effectiveness of different types of peritoneal dialysis in people with stage 5 CKD who need dialysis?
A total of 5149 articles were found by systematic searches. Full text was ordered for 320 articles (comparing peritoneal dialysis with any other modality of dialysis, including haemodialysis) based on the title and abstract. Only four papers from three studies (Bro et al. 1999; de Fijter et al. 1991; de Fijter et al. 1994; Iles-Smith et al. 1999) met the eligibility criteria (see appendix 1 for review protocol and inclusion and exclusion criteria) and evaluated the effectiveness of different types of peritoneal dialysis for stage 5 CKD. The same three studies were also included in a systematic review (Rabindranath et al. 2007a; Rabindranath et al. 2007b) identified through our searches. The results of the systematic review based on the three included primary studies are presented below.
No RCT evidence was found for children with stage 5 CKD.
Because of the lack of RCT evidence, the Guideline Development Group asked the technical team to search for publications from national renal registries that would provide further information on the outcomes considered important for this guideline.
A total of 1672 articles were found by systematic searches focused on the retrieval of registry data reported in published articles. Full text was ordered for 261 articles based on the title and abstract. In addition, we looked for annual reports for those not identified through the searches; this included the 2008 North American Pediatric Renal Trials and Collaborative Studies annual report2. Of these publications, 53 papers met the eligibility criteria for registry publications; 11 evaluated the effectiveness of different types of peritoneal dialysis, either with comparative analyses or single intervention reports for adults and children. We also re-checked the studies identified for other questions (for example, switching) and included those that were relevant.
Table 6 summarises the included studies. See appendix 1 for more details.
Summary of studies of registry data for types of peritoneal dialysis.
Of these 11 studies, four included children (seven were adults alone); eight reported outcomes for both APD and CAPD, two for CAPD alone and one for APD alone.
For the GRADE assessment, registry studies were assessed as low-quality evidence, with downgrading as appropriate.
GRADE profile 50. Effectiveness of APD compared with CAPD – health related quality of life
GRADE profile 51. Effectiveness of APD compared with CAPD – patient involvement and satisfaction
GRADE profile 52. Effectiveness of APD compared with CAPD – mortality – RCTs (adults)
GRADE profile 55. Effectiveness of peritoneal dialysis – mortality and associated patient factors
GRADE profile 56. Effectiveness of APD compared with CAPD – preservation of renal function
GRADE profile 59. Effectiveness of APD compared with CAPD – technique failure or switch (children)
GRADE profile 61. Effectiveness of APD compared with CAPD – associated patient factors
GRADE profile 62. Effectiveness of APD compared with CAPD – resource use and costs
GRADE profile 63. Effectiveness of APD compared with CAPD – adverse events – RCTs (adults)
GRADE profile 65. Effectiveness of APD compared with CAPD – adverse events (children)
GRADE profile 66. Effectiveness of APD compared with CAPD – adverse events – CAPD alone (adults)
GRADE profile 67. Effectiveness of APD compared with CAPD – adverse events – CAPD alone (children)
GRADE profile 68. Effectiveness of APD compared with CAPD – adverse events – APD alone (adults)
GRADE profile 69. Effectiveness of APD compared with CAPD – adequacy rates (adults)
GRADE profile 70. Effectiveness of APD compared with CAPD – adequacy rates (children)
GRADE profile 71. Effectiveness of APD compared with CAPD – staff attitude
GRADE profile 72. Effectiveness of APD compared with CAPD – nutritional status
GRADE profile 73. Effectiveness of APD compared with CAPD – anaemia
Low-quality evidence (from three RCTs) showed no difference in health-related quality of life in adults on APD compared with those on CAPD.
No evidence on health-related quality of life in children was identified.
Very-low-quality evidence (from one RCT) found that adults on APD reported significantly more time for work, family and social activities than those on CAPD (p < 0.0005). Although the difference was not significant, there was a tendency for less physical and emotional discomfort caused by the dialysis fluid in the APD group. Sleep problems, on the other hand, tended to be more marked in the APD group.
Low to very-low-quality evidence (from three RCTs and seven national registries) showed a tendency for no difference in mortality in adults and children on APD compared with those on CAPD. Where differences were seen, they were associated with patient factors.
Survival on APD and CAPD tended to be better for younger adults and for those without diabetes (very low quality evidence from one national registry) (Guo and Mujais 2003). There were no comparative data to suggest that there were any differences between the types of peritoneal dialysis.
Very-low-quality evidence (from one national registry) showed higher rates of mortality for adults older than 75 on assisted (family or nurse) peritoneal dialysis compared with unassisted peritoneal dialysis. There were similar rates of peritonitis, although rates tended to be higher in people unassisted or nurse-assisted than in those assisted by family members) (Castrale et al. 2010).
Low to very-low-quality evidence (from two RCTs) showed no difference in preservation of renal function in APD or CAPD.
Very-low-quality evidence (from three RCTs and seven national registries) showed that rates of dialysis failure or switch did not seem to differ between APD and CAPD. Where differences were seen, they were associated with patient factors.
Technique survival for APD and CAPD tended to be better for younger adults and for those without diabetes (very-low-quality evidence from one national registry) (Guo and Mujais 2003). There were no comparative data to suggest that there were any differences between the types of peritoneal dialysis.
Very-low-quality evidence (from two RCTs) showed that rates of hospital admissions did not seem to differ between APD and CAPD.
Very-low-quality evidence (from three RCTs and six national registries) showed that rates of adverse events did not appear to differ between APD and CAPD.
Low to very-low-quality evidence (from two RCTs and one national registry) showed no difference between APD and CAPD in dialysis adequacy.
No evidence on staff attitude and skills was identified.
Very-low-quality evidence (from one RCT) showed no significant difference in measures of nutritional status between adults on APD and those on CAPD (Bro et al. 1999).
Very-low-quality evidence (from one RCT) showed that erythropoietin use did not differ between adults on APD and those on CAPD (de Fijter et al. 1991).
This was not considered to be a health economic question.
The Guideline Development Group agreed that overall there was no evidence of significant differences between the modalities for the critical outcomes.
The evidence on assisted peritoneal dialysis was considered to be of limited relevance to current practice in the UK. This was specifically because assisted peritoneal dialysis tended to be used only in those patients who were most frail (older, or with more health problems) and the organisation of care in the country where this was assessed (France) was not directly applicable to current practice in the UK.
Therefore, based on this evidence and the related evidence on decision-making, the Guideline Development Group considered that adults and children choosing peritoneal dialysis should have the choice of CAPD and APD, with aAPD as needed. However, based on clinical experience, there was one situation in which APD was recommended as the preferred option – for infants and children on a liquid diet.
Before starting peritoneal dialysis offer all patients a choice, if appropriate, between CAPD and APD (or aAPD if necessary).
For children for whom peritoneal dialysis is appropriate offer APD in preference to CAPD if they are on a liquid diet, especially if they have low residual renal function.
What is the effectiveness of different sequences of treatment that include peritoneal dialysis in people with stage 5 CKD who need dialysis?
A total of 1669 articles were found by systematic searches. Full text papers were ordered for 50 articles (comparing sequences of treatment that include peritoneal dialysis) based on the title and abstract. Seven studies (Badve et al. 2006b; Guo and Mujais 2003; Heaf et al. 2002b; Huisman et al. 2002b; Jaar et al. 2009; Mujais and Story 2006; Rao et al. 2009) met the eligibility criteria (see appendix 1 for review protocol and inclusion and exclusion criteria) and evaluated the effectiveness of treatment sequences that include peritoneal dialysis. Different sequences were compared in the published studies. Only one study included children (Badve et al. 2006b).
As for the other reviews of dialysis in this guideline, because of the lack of RCT evidence, we included publications from national renal registries that would provide further information on the outcomes considered important for this guideline. We also re-checked the studies identified for other questions (for example, effectiveness of peritoneal dialysis) and included those that were relevant. Table 7 shows a summary of the studies that were included.
Summary of included studies on treatment sequences.
GRADE profile 75. Effectiveness of sequences and impact of switching – patient involvement
GRADE profile 76. Effectiveness of sequences and impact of switching – mortality
GRADE profile 80. Effectiveness of sequences and impact of switching – resource use and costs
GRADE profile 81. Effectiveness of sequences and impact of switching – adverse events
GRADE profile 82. Effectiveness of sequences and impact of switching – adequacy rates
GRADE profile 83. Effectiveness of sequences and impact of switching – staff attitude and skills
GRADE profile 84. Effectiveness of sequences and impact of switching – nutritional status
GRADE profile 85. Effectiveness of sequences and impact of switching – anaemia
Overall, there was a lack of high-quality evidence on the effectiveness of sequences or the impact of switching.
No studies were identified on sequences or switching in children alone.
No studies reported the effect of different sequences on health-related quality of life, patient involvement, preserving renal function, resource use and costs, adequacy rates, staff attitude, nutritional status or anaemia.
Overall, there was no clear evidence of different sequences having different mortality rates. Similarly there was no clear evidence for the impact of switching (very-low-quality evidence based on six national registries).
Overall, there was no clear evidence of different sequences having different dialysis survival rates or rates of switching. Similarly there was no clear evidence for the impact of switching (very-low-quality evidence based on three national registries).
Overall, there was no clear evidence of different sequences having different rates of adverse events. Similarly there was no clear evidence for the impact of switching (very-low-quality evidence based on two national registries).
This was not considered to be a health economic question.
The Guideline Development Group recognised that duration of peritoneal dialysis is often used as a reason to switch treatment in anticipation of rare but significant adverse events, such as encapsulating peritoneal sclerosis (EPS; for example, in one study, the overall incidence of EPS in adults over the course of 4 years was 2.5%, and was lower in children at 1.6% in 5 years) (Hoshii et al. 2000; Kawanishi et al. 2004).
Although the duration of peritoneal dialysis is associated with the development of EPS, there is no evidence that people should be switched to prevent such adverse events. The Guideline Development Group considered that people should remain on the dialysis modality that is most effective and not switch unless there are clinical reasons to do so, such as progressive loss of solute transport or of ultrafiltration, or the patient or carers express a wish to switch.
Similarly, there was no conclusive evidence on the most effective sequence of treatments. However, there are clinical implications of sequencing; for example haemodialysis requires vascular access, so access for fistula formation is limited. Therefore, people who wish to switch should be aware of any implications this may have for future options, noting that people may have dialysis over many years. Where possible, switching should be a planned process.
Based on this evidence and the related evidence on effectiveness on choice of dialysis, the Guideline Development Group considered that adults and children should remain on the dialysis modality unless there are clear clinical reasons to switch, or the patient or carer expresses a wish to do so.
Do not routinely switch patients on peritoneal dialysis to a different treatment modality in anticipation of potential future complications such as encapsulating peritoneal sclerosis. However, healthcare professionals should monitor risk factors such as loss of ultrafiltration and discuss with patients regularly the efficacy of all aspects of their treatment.
Consider switching treatment modality if the patient, their family or carer asks.
When considering switching treatment modality, offer information on treatment options described in recommendations 1.1.1–1.1.8. This should also include how any decision to switch may affect future treatment options.
Switching between treatment modalities should be planned if possible.
All rights reserved. This material may be freely reproduced for educational and not-for-profit purposes. No reproduction by or for commercial organisations, or for commercial purposes, is allowed without the express written permission of NICE.
Your browsing activity is empty.
Activity recording is turned off.
See more...
