NKF KDOQI GUIDELINES 2000
GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY
V. Adequate Dose of Peritoneal Dialysis
Weekly Dose of CAPD (Evidence)
For CAPD, the delivered PD dose should be a total Kt/Vurea of at least 2.0 per week and a total creatinine clearance (CCr) of at least 60 L/wk/1.73 m2 for high and high-average transporters, and 50 L/wk/1.73 m2 in low and low-average transporters.
Rationale A detailed rationale is presented in Appendix G. The following is a summary.
Theoretical constructs predict that a weekly peritoneal Kt/Vurea between 2.0 and 2.25 will provide adequate dialysis. These constructs assume no residual renal function, full equilibration of plasma and dialysate urea, a target serum urea nitrogen concentration between 60 and 80 mg/dL, and nPCR between 1.0 and 1.2 g/kg/day.
Clinical studies addressing the validity of these predictions can be divided into those using univariate and those using multivariate statistical analyses. The former are methodologically weaker. Four studies which used univariate analysis suggest that total (renal and peritoneal) weekly Kt/Vurea values greater than 1.5, 1.89, 2.0, and 2.0, respectively, are associated with better patient survival than lower values.
Three studies from France, Italy, and North America (CANUSA) have used multivariate statistical analysis. The French study found better survival among patients with an initial weekly Kt/Vurea >1.7 but did not evaluate changes in Kt/Vurea associated with loss of residual kidney function. The Italian study evaluated prevalent CAPD patients with minimal residual kidney function. Improved patient survival was observed with a weekly Kt/Vurea >1.96. Values higher than 1.96 were not associated with increased survival but the statistical power to detect this association was low. The CANUSA study of 680 incident continuous peritoneal dialysis patients reported a 5% decrease in patient survival in association with every 0.1 decrease in total weekly Kt/Vurea, for Kt/Vurea between 1.5 and 2.3. There was no association between Kt/Vurea and technique failure or hospitalization. The predicted 2-year survival associated with a constant total Kt/Vurea of 2.1 was 78%. These predictions assume that renal and peritoneal Kt/Vurea are equivalent.
Clinical experience suggests that a total weekly creatinine clearance >50 L/1.73 m2 is required for adequate dialysis. Among patients with minimal residual function in the Italian study, a weekly Kt/Vurea of 1.96 correlated with a weekly creatinine clearance (CCr) of 58 L. The CANUSA study reported a 7% decrease in patient survival in association with a 5 L/1.73 m2/wk decrease in CCr. Unlike the situation for Kt/Vurea, both technique failure and hospitalization were worse with decreased weekly creatinine clearance. The predicted 2-year survival of 78% was associated with a weekly Kt/Vurea of 2.1 or a weekly CCr of 70 L.
There are insufficient data to address the issue of adequate compared to optimal dialysis (see Introduction). The latter is in part defined as the dialysis dose above which the incremental clinical benefit does not justify the patient burden or financial costs. Nor are there sufficient data to evaluate the relative importance of renal and peritoneal clearances. The recommendations assume equivalence but this requires further study. The correlation between Kt/Vurea and CCr will vary with residual renal function (see Fig II-2, Appendix A). The higher CCr observed in the CANUSA study compared to the Italian study was due to the greater residual renal function in the former.
Based on the available evidence, the minimum delivered dialysis dose target Kt/Vurea should be 2.0 per week; the minimum weekly target CCr should be 60 L/1.73 m2. If there is discordance in achieving these targets, the Kt/Vurea should be the immediate determinant of adequacy because it directly reflects protein metabolism and is less affected by extreme variations in residual renal function (see Appendix A and Fig II-2). However, a cause for the discrepancy should be sought and the patient followed closely for signs of underdialysis.
A special case is the underweight patient, defined in Appendix E: Detailed Rationale for Guideline 9. Successful efforts to restore weight to a normal level in such a patient will result in a rising V, and consequently in a proportionally declining Kprt/Vurea. To achieve a weekly Kprt/Vurea of 2.0 at the increased weight, the weekly target Kprt/Vurea provided during the malnourished state must be greater than 2.0. The Work Group recommends that the target Kprt/Vurea should be raised in a malnourished CAPD patient to the level that would provide a weekly Kprt/Vurea of 2.0 for that patient if he or she were at the desired weight. That level is calculated by multiplying the target of 2.0 for CAPD times the ratio of Vdesired/Vactual. This is described in detail in Appendix E: Detailed Rationale for Guideline 9, and discussed in Guideline 17: PD Dose in Subpopulations. The same upward target adjustment should be made in creatinine clearance. The target creatinine clearance should be adjusted upward by multiplying the target for that therapy (CAPD or APD) by the ratio of BSAdesired/BSAactual.
Even after controlling for delivered dose, low and low-average transporters have better patient and technique survival than do high and high-average transporters.69 In the absence of adequate residual kidney function, low and low-average transporters may not be able to achieve a CCr of 60 L/wk/1.73 m2 on any reasonable dialysis prescription. However, because urea clearance is less affected than creatinine clearance by transport status, low and low-average transporters can achieve a weekly Kt/V of 2.0. Therefore, it seems reasonable to lower the CCr target in low and low-average transporters without fear of jeopardizing patient outcome. These patients must be observed closely for evidence of inadequate dialysis.
Clinical judgment suggests that the target doses of PD for children should meet or exceed the adult standards. However, there are currently no definitive outcome data in pediatric patients to suggest that any measure of dialysis adequacy is predictive of well-being, morbidity, or mortality.69a There also are minimal data regarding the real protein needs of children, especially young children, on dialysis.69b It is the opinion of the Work Group that the nutritional requirements per kilogram of body weight are higher in children than in adults. Therefore, PD doses in children, and especially small infants who have very high protein requirements, may have to be higher than PD doses in adults.
Weekly Dose of NIPD and CCPD (Opinion)
For NIPD, the weekly delivered PD dose should be a total Kt/Vurea of at least 2.2 and a weekly total creatinine clearance of at least 66 L/1.73 m2.
For CCPD, the weekly delivered PD dose should be a total Kt/Vurea of at least 2.1 and a weekly total creatinine clearance of at least 63 L/1.73 m2.
Rationale In the absence of data that relate delivered dose of automated PD (APD) to patient outcomes, targets for NIPD and CCPD are based on opinion.
Theoretically, there is an 8% difference in clearance between CAPD and NIPD. This difference is based on calculations which describe a 200% increase in the intermittent HD clearance required to achieve the same solute removal as in continuous dialysis (Kt/Vurea of 4.0 in HD and 2.0 in CAPD), holding protein intake constant.70,71 The Work Group assumed that the delivered dose of NIPD would need to be 8% higher than the CAPD dose (108% of 2.0 = 2.16, rounded up to 2.2). The Work Group assumed that the requisite delivered dose of CCPD would be intermediate between those for CAPD and NIPD. Some variations of CCPD with diurnal exchanges of less duration than the nocturnal exchange of CAPD may be considered equal to CAPD. However, in order to simplify recommendations, the target weekly total dose for CCPD is 2.1. The recommendations for creatinine clearance are percentage adjustments corresponding to the changes in Kt/Vurea targets for these groups.
Clinical judgment suggests that the target doses of PD for children should meet or exceed the adult target doses. There are no definitive outcome data in pediatrics to suggest that any measure of dialysis adequacy is predictive of well-being, morbidity, or mortality.
PD Dose in Subpopulations (Opinion)
There is no adequate basis for recommending any change in the target doses of dialysis discussed in Guidelines 15: Weekly Dose of CAPD, and Guideline 16: Weekly Dose of NIPD and CCPD, for various patient subpopulations (eg, patients with diabetes or who are elderly), with the exception of the malnourished patient, whose target dose is increased by the ratio of the Vdesired/Vactual for Kt/Vurea. For creatinine clearance, the target dose in a malnourished patient is increased by the ratio BSAdesired/BSAactual. Transport status is not considered a subpopulation in the context of this guideline.
Rationale There are no data available in the literature on which to base a recommendation for different adequacy targets for patients with diabetes or for the elderly. However, it must be remembered that malnourished patients may appear to have an adequate Kt/Vurea due to calculation of V from the actual or malnourished body weight. If V were calculated from an estimate of desired body weight, the target would reflect that target body weight. This is discussed in Guidelines 9: Estimating Total Body Water and Body Surface Area, and Guideline 15: Weekly Dose of CAPD, and in detail in Appendix E: Detailed Rationale for Guideline 9.
The fact that historically the size indicator to normalize CCr is BSA and for Kturea has been total body water (V) contributes to the discrepancy between these clearance measures and confounds population comparisons (male versus female, obese versus lean, edematous versus nonedematous).
In the absence of data in these subpopulations, if no other cause of malnutrition is discovered, the target delivered dose of dialysis should be increased by multiplying the target Kt/Vurea for a normally nourished patient by the ratio of the Vdesired/Vactual, and the target creatinine clearance should be increased by the ratio of BSAdesired/BSAactual. These modifications are described in Guideline 15: Weekly Dose of CAPD, and Appendices E and G.
Use of Empiric and Computer Modeling of PD Dose (Evidence)
Both empiric and computer modeling methods can be used to estimate adequate doses of PD. Specific prescriptions are described below.
Rationale The Work Group has elected to describe these two empiric and computer modeling approaches in detail. They are by no means mutually exclusive.
EMPIRIC APPROACH FOR DETERMINATION OF DOSE OF PERITONEAL DIALYSIS
A. General Evaluation of the Patient With Kidney Failure
1. Explain all options (transplant, HD, and PD) to patients/parents/caregivers in a nonbiased manner.
2. Review medical condition/comorbidities to determine if contraindications, relative or absolute, exist for any modality (see Section VIII: Suitable Patients for PD).
3. If no medical contraindications exist and the patient is a candidate for self therapy, allow patient to choose a modality.
4. Place the chronic dialysis access (PD or HD). The Vascular Access Work group recommends that vascular accesses be placed in patients on PD. The PD Adequacy Work Group feels that this decision should be made on an individual patient basis, but our position does not necessarily disagree with the recommendations of the Vascular Access Work Group.
5. If dialysis is needed at the time of presentation, place the temporary HD access, or after placing the PD catheter, initiate therapy as suggested under point B.2, below.
B. Initiation of Peritoneal Dialysis
1. If possible, wait 10 days to 2 weeks after catheter placement to start PD.
2. If PD must be started in less than 10 days following catheter placement, do low-volume, supine dialysis.
3. Obtain baseline 24-hour urine collection for urea and creatinine clearance (see Guideline 6: Assessing Residual Kidney Function). These collections are for solute clearance calculations, assessment of creatinine generation, and PNA determinations.
4. Note patients weight and the presence or absence of edema.
5. At initiation of dialysis, explain to patient/parents/caregivers that the patients prescription will be individualized. Specifically, state that their instilled volume almost certainly will need to increase over time. For patients who choose Automated Peritoneal Dialysis (APD), one or more daytime dwells will be needed in approximately 85% of patients. Patients should know from the start of PD that their total solute clearance will be monitored and that, if their residual kidney function or peritoneal transport changes over time, their prescription may need to change as well.
C. Initial Dialysis Prescription for Adults
Initial dialysis can be prescribed empirically based on patients weight, amount of residual kidney function, and lifestyle constraints. These empiric recommendations should be implemented prior to peritoneal equilibration testing.
PD may be initiated incrementally, or as full therapy, depending on RKF at the time of initiation (see Guideline 1: When to Initiate DialysisKt/Vurea Criterion). For example, if Krt/Vurea is 1.8 per week, only 0.2 Kpt/Vurea is needed per week. Assuming complete urea equilibration (serum to dialysate) at 6 hours, a single 2-L overnight exchange would contribute 14 L per week. If V is 40 L, this contributes a Kpt/Vurea of 14/40 or 0.35 per week. Any ultrafiltrate would add further to total solute removal. That, plus the Krt/Vurea of 1.8, brings the Kprt/Vurea to at least 2.15, satisfying the target requirement. This approach uses basic principles of dialysis prescription development. Thus, the dose of Kpt/Vurea depends on the Krt/Vurea as the Work Group has emphasized throughout these guidelines. Keeping in mind that the weekly Kprt/Vurea goal is at least 2.0, the following more intense empiric approach is reasonable:
1. Patients with an estimated underlying GFR >2 mL/min
b. If patients lifestyle choice is CCPD:
BSA <1.7 m2 → 4 × 2.0 L (9 hours/night) + 2.0 L/day
BSA 1.7 to 2.0 m2 → 4 × 2.5 L (9 hours/night) + 2.0 L/day
BSA >2.0 m2 → 4 × 3.0 L (9 hours/night) + 3.0 L/day
c. If patients lifestyle choice is NIPD: Specific attention to certain details will be required. Nightly intermittent peritoneal dialysis (NIPD) is not a therapy that is typically used at the initiation of dialysis. It has been reserved for high or rapid transporters. However, in patients with significant RKF (and ability to diurese), they may initially do well on nightly exchanges only (dry day) because of the supplemental clearance provided by the patients RKF. See further comments on NIPD under point 2.c, below.
2. Patients with an estimated underlying GFR < mL/min
b. If patients lifestyle choice is CCPD:
BSA <1.7 m2 → 4 × 2.5 L (9 hours/night) + 2.0 L/day
BSA 1.7 to 2.0 m2 → 4 _ 3.0 L (9 hours/night) + 2.5 L/day
BSA >2.0 m2 → 4 _ 3.0 L (10 hours/night) + 2 _ 3.0 L/day (Consider combined HD/PD or transfer to HD if clinical situation suggests need.)
c. If patients lifestyle choice is NIPD:
Many of the issues discussed above for patients with an estimated underlying GFR >2 mL/min still apply to urine volume. Namely, if RKF provides enough diuresis, NIPD may provide enough solute removal for a while. This should be tested early on. If during training, it is noted that a patient has very low drain volumes with no apparent mechanical problem or leak, a PET should be done to determine if the patient is a rapid transporter. If so, NIPD can be prescribed using kinetic modeling.
D. Initial Dialysis Prescription for Children
In view of the close, age-independent relationship between peritoneal surface area and body surface area (BSA), the use of BSA as a normalization factor for the prescribed exchange volume in children is preferred. An instilled volume of at least 1,100 mL/m2 is recommended for most pediatric patients, although individual tolerance must be considered.72
It should be emphasized that the preceding prescriptive guidelines are general empiric guidelines for patients initiating PD, generally as first renal replacement therapy. For patients transferring from HD with minimal RKF, prompt adequacy testing is required. The above empiric recommendations must be individualized and guided by documentation that the delivered dose equals the prescribed dose. Furthermore, the instilled volumes are ones that theoretically will result in a weekly target Kt/Vurea of greater than 1.9 for the average patient. Low transporters may be below creatinine targets if RKF is low. Finally, although most patients tolerate instilled volumes of greater than 2.0 L, this needs to be evaluated for each patient.
E. Observations Needed During Training
1. Determine 4-hour drain volumes during training. This is to note if drain volumes are as expected for typical 4-hour dwells with 1.5%, 2.5%, or 4.25% dextrose exchanges. This is not a formal peritoneal equilibration test (see below), but is done to determine if the patients peritoneal membrane transport characteristics are markedly different from the mean.
2. Monitor for evidence of leakage in the vicinity of the catheter.
3. Complete laboratory studies.
b. Perform serum chemistries and complete blood count.
c. If a computer-assisted kinetic modeling system is available, enter preliminary data to predict if the current prescription will be adequate.
F. Early Follow-Up
1. Perform 24-hour dialysate and urine collection for Kt/Vurea, creatinine clearance, PNA calculation, creatinine generation, and D/PCreatinine and D/PUrea values. These should be done 2 to 4 weeks following initiation (see Table II-1 and Guideline 3: Frequency of Delivered PD Dose and Total Solute Clearance Measurement Within Six Months of Initiation).
2. Perform peritoneal equilibration testing (PET) approximately 1 month following initiation of PD, an appropriate time physiologically. This baseline PET could be performed at the end of a prolonged (>1 week) training period (see Guideline 3: Frequency of Delivered PD Dose and Total Solute Clearance Measurement Within Six Months of Initiation). This PET (1 month) is used as the baseline measure of peritoneal membrane transport characteristics, not to determine total solute clearance. This PET is done to rule out unsuspected problems or deviation from mean transport characteristics. Low transporters will probably require high-dose CAPD or CCPD. High transporters will eventually have ultrafiltration problems (when RKF diuresis fails) and will need short-dwell therapy such as NIPD. Average transporters will have the most flexibility (ie, all options will be feasible).
3. Perform serum chemistries and complete blood count.
4. If a computer-assisted modeling program is available, enter baseline data. Actual data from 24-hour collection can be compared.
5. If clearances are at or above target, continue routine monitoring on a regular basis. Look for changes in 24-hour urine studies and PET data. Kinetic modeling can be used to guide future therapy.
6. If clearance is below target at 1 month, a change in prescription may be needed. Compliance issues and collection procedures should be evaluated for abnormalities.
G. Adjusting Dialysis Prescription
If kinetic modeling is not available, unless PET has changed, dialysis dose is most effectively increased by increasing the instilled volume, therefore maximizing mass transfer and dwell time. Another option would be to increase the number of exchanges/day while maintaining maximum dwell time, ie, by using a single nighttime exchange to increase to 5 equal dwells/day. To this end, simplified mechanical exchange systems have been developed to perform a nocturnal exchange.
If kinetic modeling is available, use these programs to tailor a new prescription to meet adequacy target goals and patient lifestyle issues. This is discussed in the next section.
COMPUTER-ASSISTED KINETIC MODELING APPROACH TO ACHIEVING TARGET DOSES OF PERITONEAL DIALYSIS
As mentioned above, the availability of computer-assisted kinetic modeling to tailor PD prescriptions to transport type, body size, lifestyle, etc. may have distinct advantages. Much of what is described in the preceding discussion of the empirical approach has a mathematical basis. Computer-assisted kinetic modeling is a logical extension of the empirical approach in that it uses computer calculations to speed and assist the physician in PD prescription development. PD solution manufacturers, such as Baxter and Fresenius, provide urea kinetic modeling (UKM) models without charge.
The major advantage of this approach is the flexibility and speed of the calculations of solute clearance. Recent studies have shown that certain models very accurately predict dialysis delivery.73-76 Kinetic modeling is especially important for APD therapies because the dwell times are so variable and may not approach the optimal for many patients. Since the models accurately and reliably predict the delivered dose, the patients and caregivers can discuss options in a timely manner. The trial and error empirical approach discussed above moves at a much slower rate. Even with computer-assisted UKM, actual measurements are still necessary to confirm adequate dose delivery. But with computer-assisted UKM, the process is accelerated. The only theoretical disadvantage of a computer kinetic modeling approach is that there might be a tendency for the caregivers not to learn the principles behind the modeling program which form the basis for the prescription strategies.
The use of computer-based modeling to achieve target PD doses has been applied successfully to a small number of children.77,78
The dose of PD is defined as the sum of the total daily or weekly kidney + peritoneal urea clearance normalized to the total body water, the Kprt/Vurea, which is a dimensionless parameter expressed as either the total daily or weekly fractional clearance of body water for urea.79,80 It is substantially more difficult to compute the appropriate prescribed dialysis dose in PD than in HD. In HD, Kt/Vurea is delivered in a single dialysis session and can be precisely calculated from the dialyzer mass transfer coefficient (MTC), constant blood and dialysate flows, ultrafiltration rate, and treatment time.70,81 The delivered Kt/Vurea and PCR (or PNA) can both be calculated from the predialysis and postdialysis BUNs from a single dialysis.70,81
In PD, however, the total daily peritoneal clearance, Kpt, is comprised of the sum of the clearances provided by several discrete exchanges. This therapy consists of several batch exchanges during which clearance and ultrafiltration are not constant (unlike the situation in HD) but fall exponentially to zero over the course of each exchange. The peritoneal mass transfer coefficient (MTC), which controls the rate of solute transport between blood and dialysate (and hence clearance), is an individual patient characteristic71,82 which must be determined and which can clearly vary as a function of exchange volume and body position.83,84 In HD, ultrafiltration contributes minimally to urea clearance, while in PD it contributes up to 25% or more of total clearance and must be included in calculation of the dose. The net ultrafiltration can be precisely controlled in HD, while in PD it is a complex function of glucose absorption, membrane water permeability and lymphatic flow. The PD prescription variables include MTC, which is dependent on patient, body position, and exchange volume; distribution of exchanges between ambulatory and supine cycler exchanges; exchange volume(s); exchange times; and the osmotic gradient or percent dextrose in each exchange. Additionally, residual kidney urea clearance must be measured and included in the prescription and body water or V estimated from age, gender, and surface area.39,85
In current clinical practice, the peritoneal dialysis prescription is usually based on transport categorization using the peritoneal equilibration test (PET) and subsequently more finely tuned through empirical prescription changes guided by clinical experience, as described in the preceding section of this guideline.
With computerized UKM many possible PD regimens with variable exchange schedules and volumes can be tailored to be compatible with individual patient lifestyle preferences and to minimize total dialysate volume relative to required Kprt/Vurea. A PD prescription can be quickly and rigorously evaluated mathematically using programs written for the personal computer.73-75,86 These programs all require baseline transport characterization using either the PET or peritoneal function test data.
The most common method to monitor delivered Kprt/Vurea is measurement of total daily peritoneal and kidney urea clearance by analysis of blood, total drained dialysate, and 24-hour urine for urea nitrogen. Although highly reliable for determination of Kprt/Vurea and PCR/PNA, batch analyses do not provide data to distinguish between noncompliance and/or possible changes in MTC when the delivered Kprt/Vurea deviates from that expected with the current prescription. There is further discussion of this subject in Guideline 7: PD Dose Troubleshooting, and Guideline 8: Reproducibility of Measurement.
An alternative technique is measurement of BUN, urine volume, and urea nitrogen in aliquots of each exchange87 combined with the patients report of the exchange time and volume for each exchange. A further discussion of aliquot methodology is in Guideline 8: Reproducibility of Measurement. Although this technique, based on the peritoneal function test approach, requires substantially more dialysate urea nitrogen measurements in addition to measurement of Kprt/Vurea and PCR (or PNA), it permits calculation of a MTC for each exchange. If there are important deviations from the reported exchange schedule, the deviant exchanges will be identified by markedly deviant MTCs.
There are advantages and disadvantages of both the PET and PFT measurements. The wide range of exchange times, including nearly complete equilibration in long exchanges and the assumption of constant BUN, will result in some variability in the MTCs measured with the PFT which do not reflect true differences. On the other hand, the MTC calculated with a single 2-L exchange under carefully controlled conditions will not always accurately reflect the MTC under clinical exchange conditions with variable exchange volumes and body position.
A simple kinetic technique for routine monitoring of delivered Kprt/Vurea and PCR/PNA in established patients for whom the MTCs for urea and creatinine and 24-hour dialysate creatinine have been previously established is measurement of BUN and serum creatinine and dialysate urea and creatinine from an aliquot of one exchange. This data combined with the number of exchanges, exchange times, and exchange volumes reported by the patient permits calculation of Kprt/Vurea, PCR/PNA, and expected total dialysate creatinine content.81 The validity of the data can be assessed by comparison of the calculated dialysate creatinine content to the measured historical value for the patient (see Guideline 7: PD Dose Troubleshooting). In this way, both Kprt/Vurea and PCR/PNA can be estimated from measurements of urea nitrogen and creatinine in a blood sample and a small aliquot of one exchange using computerized UKM. When there is deviation of more than 10% in the expected creatinine excretion, more complete dialysate collections would be indicated for analysis of therapy.
RECOMMENDATIONS FOR RESEARCH
The amount of dialysis required for malnourished patients is not known. While there probably is consensus that such patients need extra dialysis, the requisite increase is unclear and should be studied. Other malnutrition-related questions of interest include: Can aggressive dialysis delivery reverse malnutrition? What V is to be used in malnourished patients?
Can increasing dialysis dose improve outcomes in a linear manner, or is there a dose above which no benefit is noted, or complications or costs outweigh the benefits?
A multicentered study of pediatric patients to evaluate clinical outcome as a function of delivered PD dose should be initiated. Urea kinetic modeling computer programs specifically designed for children should be developed and validated in a prospective trial.
Although there is a database documenting the validity of UKM to describe transport in PD,71,73-75,88,89 UKM has not been widely used to prescribe and control the delivered dose. Since the risk of mortality may be highly nonlinear with increased dose of delivered dialysis,90,91 it is reasonable to assume that the coefficient of variation on mean Kprt/Vurea should not exceed 10% to 15% for individual patients. A multicenter clinical trial to study clinical outcome as a function of Kprt/Vurea, using UKM to control Kprt/Vurea prospectively in individual randomized patients is recommended.
Do patients with diabetes need higher targets for delivered dose of PD? Should PD delivered dose be increased in hospitalized patients during acute illness or stress?
© 2001 National Kidney Foundation, Inc