KDOQI (Kidney Disease Outcomes Quality Initiative)
NKF KDOQI GUIDELINES

Executive Summaries | Anemia | Hemodialysis | Peritoneal Dialysis |
Vascular Access | Nutrition | CKD 2002 | Dyslipidemias | Bone Metabolism | Hypertension and Antihypertensive Agents | Cardiovascular Disease in Dialysis Patients | History of KDOQI | Pediatric Bone | Anemia 2006 |
Updates 2006

Clinical Practice Guidelines and Clinical Practice Recommendations
2006 Updates
Hemodialysis Adequacy
Peritoneal Dialysis Adequacy
Vascular Access


CLINICAL PRACTICE GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

III. RESEARCH RECOMMENDATIONS

RANKING OF RECOMMENDATIONS
Research recommendations have been grouped into 3 categories: critical research, important research, and research of interest. These rankings were made by the Work Group based on current evidence and on the need for research to provide additional evidence.

The recognized lack of prospective randomized trials and level A medical evidence was noted when the original KDOQI Guidelines for PD Adequacy were formulated. As a result, most of the guidelines were opinion based, and important areas for research were identified. Some of those questions have been answered with well-conducted, prospective, randomized trials so that new guidelines can be formulated with grade A and B medical evidence. Subsequent studies have identified further questions and deficiencies in current medical knowledge that will hopefully stimulate further research.

CRITICAL RESEARCH RECOMMENDATIONS
Guideline 1. Initiation of Dialysis
Although it is recognized that the patient's clinical condition at the start of KRT is an important predictor of outcome, there are no data to confirm whether an “earlier” (in terms of kidney progression) or a “healthier” (less advanced comorbidities) start results in a survival advantage or just a lead-time bias. Furthermore, is the answer to that question dependent on prior rate of progression of kidney disease, cause of kidney disease, the same for different ethnic groups, or dependent on comorbidities present? Given the cost of KRT to society, it is important to know whether, in general, the timing of the start of dialysis therapy improves total lifespan or only increases time on dialysis therapy, but not total lifespan. If it is the latter, data to show that the patient otherwise would tend to be healthier with less hospitalization, better QOL, or rehabilitation also would be important to know.

Guideline 2. PD Solute Clearance Targets and Measurements
It now is well documented that the presence of RKF offers the typical patient on KRT an important survival advantage. What is not known is why that is true. Is it caused by better blood pressure or volume control, more small-solute removal, removal of middle molecules, or some other poorly recognized function of metabolic or paracrine function of the kidney tubules? Additional research to define the effects of RKF would be most important. Results may influence clinical practice, guidelines on initiation of dialysis therapy (can it be done in an incremental manner?), and further determine how one best includes the residual kidney component of total solute clearance in dose calculations.

As noted in the text of these guidelines, 2 recent prospective randomized trials suggested that, over the range of solute clearance studies and using current standard PD technologies (mainly CAPD), trying to achieve higher solute clearance goals had little clinical benefit for the population as a whole. Therefore, considerably more research is needed in the area of adequacy of PD. Additional randomized trials, optimally multicentered, to examine different PD doses are needed to evaluate lower Kt/Vurea in populations with larger patients with more comorbidity. A study that compares a group that maintains a peritoneal Kt/Vurea of 1.7 from the start of dialysis therapy (disregarding RKF) with a group that has a total Kt/Vurea of 1.7 (kidney plus peritoneal, which would require starting PD therapy with a minimal prescription with subsequent adjustment upward as RKF is lost) would be helpful. In addition, a randomized trial of different levels of small-molecule clearances is needed specifically for anuric patients on PD therapy. Trials with a longer follow-up than 2 years with assessment of nerve conduction to evaluate for neuropathy would be helpful. Markers of middle-molecule clearances also should be obtained long term. A randomized study to evaluate the influence of middle-molecule clearance or of full- versus partial-duration day dwells on patient outcomes would be valuable.

Trials are needed in APD, with both dry day and wet days. A trial that compares outcomes with beginning PD on APD with a dry day versus beginning PD with a wet day (controlling for peritoneal dose), with the subsequent adjustment of the prescription (including the addition of a wet day), would be informative in evaluating the potential benefit of a dry abdomen for part of the day on protection of the peritoneal membrane and immune function. Such a study would need to include markers of the peritoneal membrane, as well as determination of middle molecules and neuropathy.

Studies must be designed that separate the effects of volume control from those of small-molecule clearances. It is clear from the studies that have been done that volume overload sometimes is a consequence of using a limited number of exchanges in CAPD and perhaps a consequence of excessively short nighttime exchanges in APD, in which the ultrafiltration volume is likely to be 50% sodium free.

Because increasing small-molecule clearance does not appear to be the path to improved survival, studies investigating other maneuvers to decrease mortality should be investigated. Attention should be focused on specific causes of mortality. These studies could include use of an ACE inhibitor in combination with a lipid-lowering drug versus ACE inhibitor alone, monthly follow-up to assess and adjust the prescription to maximize volume status versus less frequent visits, and to evaluate cardiovascular deaths. Anuric patients are more likely to die a sudden death.69 Data from the same group indicate that hypokalemia is a risk factor for death; in this study, hypokalemia was defined by 3 measurements of potassium during 12 weeks, and sudden death was not more frequent in this group.270 Therefore, it seems possible that hypokalemia might be more common in anuric patients, possibly because of dietary and nutritional issues, and contribute to sudden death, but this needs to be studied.

Another area that might prove fruitful to decrease morbidity and mortality is further research on decreasing the risk for, and managing, peritonitis. The risk for death related to peritonitis is variable from a low of 3% of deaths in Canada to 16.6% of deaths in Hong Kong.69 Aggressive catheter removal for refractory peritonitis versus delayed catheter removal (in an attempt to decrease mortality related to peritonitis) may result in a decrease in peritonitis-related deaths. Peritonitis remains the leading cause of technique failure271 and affects peritoneal function during the first year on PD therapy.272 Additional research on training methods and exit-site care may prove fruitful.

Last, studies of maneuvers to improve adherence with the prescription and diet are much needed in PD patients, especially in such countries as the United States, where adherence is less than optimal. Such maneuvers might include closer monitoring, treatment of depression, evaluation of supplies with home visits, etc.

Guideline 3. Preservation of RKF
Rigorous studies are needed to examine whether the use of radiocontrast dye affects RKF in dialysis patients and whether renoprotective strategies in the nondialysis population also apply to those on dialysis therapy. Although controversial, it was suggested that the rate of decrease in RKF in those on APD therapy compared with those on CAPD therapy is faster. More data are needed. Because of financial issues and ease of administration, use of aminoglycoside antibiotics for the treatment of peritonitis has been recommended. Therefore, data about whether long- or short-term use of aminoglycosides is associated with a more rapid decrease in RKF would be helpful. The USRDS analysis80 showed an association between use of ACE inhibitors and also use of calcium channel blockers with better preservation of RKF. Subsequent studies examined ACE inhibitors and ARBs, but not the use of calcium channel blockers; therefore, additional studies are needed. Clinical evaluation of the continuing use of immunosuppressive therapy (other than calcineurin inhibitors) to maintain residual kidney allograft function in patients on dialysis therapy is lacking. Also unclear is whether the benefit of attaining normotension by vigorous ultrafiltration is offset by the decrease in RKF from the attendant volume depletion.

Guideline 4. Maintenance of Euvolemia
Randomized trials to determine optimal blood pressure targets for PD patients are required. Larger randomized trials looking at the effect of newer dialysis solutions on important patient outcomes also would be helpful. Studies looking at the relationship between peritoneal hypertonic glucose exposure and metabolic and cardiovascular outcomes, as well as patient survival, would be valuable.

Guideline 5. Quality Improvement Programs
CQI programs were shown to improve specific outcomes for subgroups of patients, such as peritonitis rates, exit-site infection rates, technique failure rates, etc. It would be important to develop a better understanding about which factors also improve patient well-being and satisfaction with their modality. Current guidelines recommend assessing peritoneal transport status by using PET. They subsequently recommend a hypertonic dwell (4.25% dextrose) to work up a patient with ultrafiltration failure. Studies that compare 1.36%/1.5% dextrose or 2.27%/2.5% dextrose PET with 3.86%/4.25% dextrose PET are minimal. Because the 3.86%/4.25% test is recommended for the workup of ultrafiltration failure, more comparison data are needed. Furthermore, most kinetic modeling programs use data from 2.27%/2.5% dextrose PET to predict solute clearance and ultrafiltration. One needs to evaluate whether current kinetic modeling programs are as accurate if 4.25% PET is used; alternatively, if not, one may want to develop programs that use 4.25% dextrose PET data specifically. Once done, the standard PET may be changed to a 4.25% dextrose PET.

Guideline 6. Pediatric PD
Pediatric data are sparse, in part because there are few clinical trials using RR for death as an outcome for adequacy. However, there are other important aspects of overall patient care that need to be considered and evaluated. These include the development of a simplified means to estimate glomerular rate in children that precludes the need for urine collection and that is accurate at low levels (stages 4 to 5 CKD) of kidney function, determination of adequate and optimal total solute clearance in children receiving PD, comparison of the impact of peritoneal solute clearance versus RKF on patient outcome, evaluation of PD and the longevity of dialysis therapy on QOL of pediatric patients and their families, determination of the ability of icodextrin-based dialysis solutions to enhance ultrafiltration across the age/size spectrum of pediatrics, and evaluation of the safety and efficacy of ACE-inhibitor, ARB, and diuretic therapy in children with CKD stage 5 receiving PD.

IMPORTANT RESEARCH RECOMMENDATIONS
Guideline 1. Initiation of Dialysis
Much more research is needed regarding the impact on the patient of the period leading up to dialysis therapy and the period just after starting dialysis therapy. Additional research is needed on mood disorders, particularly depression and anger, that may develop during this period and the impact such disorders may have on outcomes after dialysis therapy is initiated.

Guideline 2. PD Solute Clearance Targets and Measurements
The presence of RKF was rather arbitrarily defined in this document by the Work Group as 100 mL of urine output per day. This was chosen because many of the studies on clearances chose 100 mL/d as the cutoff value. However, it is not clear that this is the most appropriate level of urine output to use, or even if urine volume, rather than measured GFR, would be preferable. Additional research is needed in this area.

Guideline 3. Preservation of RKF
Data for the effect of peritonitis on RKF are contradictory. Studies examining the impact of peritonitis, as well as the treatment approach, on RKF are needed. In particular, the severity of peritonitis may relate to loss of RKF with more severe episodes (for example, fungal or those caused by gram-negative bacilli) perhaps more likely leading to loss of RKF.

Guideline 4. Maintenance of Euvolemia
Euvolemia in home dialysis patients is not always readily achieved because patients may not be knowledgeable about this aspect of PD. A study examining training methods emphasizing evaluation of “euvolemia” as done by the patient on impact of blood pressure and volume status would be worthwhile. In addition, there are few, if any, studies of interventions to enhance patients' abilities to follow a rather rigorous diet in regard to sodium intake. Such studies should be undertaken.

Guideline 5. Quality Improvement Programs
Quality improvement programs are rather time consuming and therefore costly. A cost analysis of the impact of aggressive interventions by a program on outcomes should be carried out.

CPR for Guideline 3
Current guidelines recommend assessing peritoneal transport status by using PET. They subsequently recommend a hypertonic dwell (4.25% dextrose) to work up a patient with ultrafiltration failure. Studies that compare 1.36%/1.5% dextrose or 2.27%/2.5% dextrose PET with 3.86%/4.25% dextrose PET are minimal. Because the 3.86%/4.25% test is recommended for the workup of ultrafiltration failure, more comparison data are needed. Furthermore, most kinetic modeling programs use data from 2.27%/2.5% dextrose PET to predict solute clearance and ultrafiltration. One needs to evaluate whether current kinetic modeling programs are as accurate if 4.25% PET is used; alternatively, if not, one may want to develop programs that use 4.25% dextrose PET data specifically. Once done, the standard PET may be changed to a 4.25% dextrose PET.

RESEARCH RECOMMENDATIONS OF INTEREST
Guideline 1. Initiation of Dialysis
It is unclear whether it is advisable to start patients who chose cycler dialysis on therapy with a dry day. Theoretically, this might enhance long-term preservation of the peritoneal membrane, but there are no data concerning this. In addition, it is unclear whether a patient who wishes to dialyze at home using the cycler should be started initially on CAPD. The Work Group sees no reason for such an approach, but a study might be carried out to investigate the impact of each approach on QOL. Research on the desirability of placing a “backup” fistula in patients who chose PD therapy would be of interest.

Guideline 2. PD Solute Clearance Targets and Measurements
HD patients have a measure of small-solute adequacy carried out each month. There is considerable controversy among PD experts about the desirable frequency of measurements of peritoneal clearance, which tends not to change much over time. Studies evaluating this more carefully are warranted.

Guideline 3. Preservation of RKF
Because PD prescriptions often are dependent on RKF, it is important that the health care team recognize when RKF decreases. A study in which the PD patient measures the volume of urine output daily (such as transplant recipients historically were asked to do in the period immediately after transplantation) as a marker of RKF and to notify the dialysis program if there is a substantial change would be of some interest to see if this approach impacts on earlier recognition of an important change in RKF.

Guideline 4. Maintenance of Euvolemia
Patients on PD therapy traditionally are asked to monitor their own blood pressure at home, at a minimum daily. A study in which the patient measures blood pressure more often and takes an as-needed extra blood pressure medication for an elevation would be of interest to determine if this impacted on RKF or other outcomes.

Guideline 5. Quality Improvement Programs
Quality improvement programs traditionally are multidisciplinary. However, it is unclear how involved the physician or physician assistant typically is in many programs in the CQI efforts. An evaluation of the importance of having a “physician PD champion” in the multidisciplinary CQI process would be of interest.