NKF-DOQI CLINICAL PRACTICE GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

867 Acronyms and Abbreviations

Abbreviation Term ACE angiotensin-converting enzyme BSA body surface area BUN blood urea nitrogen concentration (an old term; SUN is technically correct, see below) CANUSA Canada/USA Peritoneal Dialysis Study CAPD continuous ambulatory peritoneal dialysis CCPD continuous cycling peritoneal dialysis CCr creatinine clearance CrCr residual renal creatinine clearance DI dialysis index DPI dietary protein intake eC effective clearance ESRD end-stage renal disease GFR glomerular filtration rate HD hemodialysis Kpcr peritoneal creatinine clearance KptlVurea the peritoneal component of KtIV urea KprtlVurea the sum of peritoneal and renal KtIV urea. These terms are interchangeable in that KtIVurea is total unless otherwise noted. KtlVurea urea clearance X time normalized by total body water, the volume of distribution of urea KrtlVurea the renal component of KtIV urea MTC mass transfer coefficient n normalized nBSA normalized body surface area NIPD nightly intermittent peritoneal dialysis nPCR normalized protein catabolic rate nPNA normalized protein equivalent of total nitrogen appearance nV normalized volume PCR protein catabolic rate PD peritoneal dialysis PET peritoneal equilibration test PNA protein equivalent of nitrogen appearance QOL quality of life RRF residual renal function SGA subjective global assessment SHR standardized hospitalization rates SUN serum urea nitrogen concentration t time UKM urea kinetic modeling UNA urea nitrogen appearance URR urea reduction ratio USRDS United States Renal Data System V volume of distribution. When referring to urea, this is total body water

S68 Introduction

This Work Group was charged with preparing purpose of these guidelines, a child was considered practice guidelines for the "adequacy of perito• to be a' patient less than 19 years of age. neal dialysis," a topic which could be defined An "effective dose" is that which achieves its broadly or narrowly. The Work Group elected to stated goal. That goal is some form of outcome focus its guidelines on those areas of "ade• measure(s), and could be determined by patient, quacy" that needed the most urgent develop• provider, payer, regulator or a combination of ment, knowing that subsequent guidelines will these parties. At the lower extreme is the' 'mini• be developed or that others were currently under mal effective dose." In certain circumstances this development (e.g., for management of peritoni• may be interpreted as "adequate," At the other tis). In addition, the Work Group focused on top• extreme is the "maximal effective dose," the ics for which guidelines would likely have the dose above which there are no additional bene• greatest impact on patient outcomes. However, fits. For hemodialysis and peritoneal dialysis the the Work Group's focus should not be construed maximal effective dose is not known. Some• to mean that areas not covered are unimportant. where between these extremes is the "optimal Some external reviewers criticized these dose," the dose above which the additional de• guidelines as too complex, while others wrote rived benefit does not justify the additional cost that they were not thorough enough. Some or burden. If one accepts this definition, the Work wanted guidelines merged, and others thought Group intended to more precisely define "opti• the guidelines were too dense. The Work Groop mal dose" targets in a clinically relevant and considered all these issues. quantitative fashion. It was the intention of the We advise the reader to first become familiar Work Group to bring "adequate dose" to the with the Table of Contents, which provides a level of "optimal dose" by raising the outcome listing of the Clinical Practice Guidelines for goals or expectations. The present guidelines at• Peritoneal Dialysis Adequacy; detailed rationales tempt to make recommendations based on avail• are provided for each guideline. Redundancies able scientific/medical evidence, resorting to ex• are often intentional because it is anticipated that pert opinion only when necessary. It is clearly a reader might review only selected topics. How• stated in each guideline title when recommenda• ever, the Work Group considers these guidelines tions were based on evidence, opinion, or both. as best viewed in their entirety, rather than in Even when guidelines were based on opinion, their component parts. that opinion is supported by direct or extrapo• There is a paucity of data on children in the lated evidence. areas covered by these guidelines. Pediatricians These guidelines are intended for use by health were represented on the Work Group, and outside care professionals trained to understand varia• pediatric consultations were obtained. Because tions in the practice of medicine and the necessity some recommendations for adults do not apply to for such variation. These guidelines are not in• children, additional recommendations are included tended for punitive use by any oversight official when appropriate for pediatric patients. For the who does not understand the reasons or the ne• cessity for practice variations including varia• © 1997 by the National Kidney Foundation, Inc. tions in societies different from that of the United 0272·6386/97/3003-0210$3.00/0 States.

AmericanJournal of Kidney Diseases, Vol 30, No 3, Suppl 2 (September), 1997: p S69 S69 I. Initiation of Dialysis

BACKGROUND or peritoneal). Hence, it is likely that delays in referral for initiation of dialysis result in unneces• sary morbidity and potentially higher costs as Two clinical guidelines for when to initiate well. While seeing a nephrologist does not guar• dialysis are provided because there appear to be antee that patients will be adequately prepared two independent predictors of clinical outcome. and referred for dialysis, it dramatically increases The first guideline is based on the level of renal the likelihood that this will occur. 13 function (as measured by KrtNurea per week); the second is based on the level of protein intake (as estimated from measured 24-hour urinary urea GUIDELINE 1 nitrogen output and estimated non-urea nitrogen When to Initiate Dialysis-KtlVurea losses). Criterion (Opinion) Although less than I % of American dialysis patients begin dialysis with a serum creatinine Unless certain conditions are met, patients concentration <8.0 mg/dL or a CCr > 10 mLl should be advised to initiate some form of dial• min, approximately 60% suffer from nausea! ysis when the weekly renal KtNurea (KrtNurea ) vomiting at the time of dialysis initiation.3 Thus, falls below 2.0. The conditions that may indicate the likelihood of malnutrition in this population dialysis is not yet necessary even though the is high. Evidence from the Modification of Diet weekly KrtNurea is less than 2.0 are: in Renal Disease (MDRD) studl and a recent 1. Stable or increased edema-free body large Australian study5 clearly show that when weight. Supportive objective parameters for ade• glomerular filtration rate (GFR) decreases to 25 quate nutrition include a lean body mass >63%, to 50 mL/min, patients adapt by reducing their subjective global assessment score indicative of protein intake. Protein intake continues to decline adequate nutrition (see Guideline 12: Nutritional as renal disease progresses to end stage. These Status Assessment and Appendix B: Detailed Ra• observations have been corroborated in a pro• tionale for Guideline 2) and a serum albumin spective study.6 As renal function deteriorates, concentration in excess of the lower limit for the protein and energy intake decreases, leading to lab, and stable or rising; and changes in body weight, fat mass, serum albu• 2. nPNA ~0.8 g/kg/d (see Guideline 2: When min, and transferrin concentrations. Earlier initi• To Initiate Peritoneal Dialysis-nPNA Criteria ation of dialysis may prevent or perhaps even and Appendix B: Detailed Rationale for Guide• reverse this deterioration in nutritional status. In• line 2); and creased serum albumin concentration has been 3. Complete absence of clinical signs or shown to parallel the increase in KtNurea for HD symptoms attributable to uremia. patients.7 In addition, an editorial review cited A weekly KrtNurea of2.0 approximates a renal data8 suggesting that albumin level at initiation urea clearance of 7 mL/min and a renal creatinine of dialysis is predictive of survival.9 clearance that varies between 9 to 14 mL/mini 2 Adverse clinical and economic consequences 1.73 m • Urea clearance should be normalized of failure to properly manage patients as they to total body water (V) and creatinine clearance 2 approach ESRD and dialysis were first described should be expressed per 1.73 m of body surface in Britain.1O These observations have now been area. The GFR, which is estimated by the arith• corroborated in other regions of Britain, the metic mean of the urea and creatinine clearances, 2 United States, and France. Jl·14 Specifically, costs, will be approximately 10.5 mLiminl1.73 m hospitalization, and morbidities decrease if atten• when the KrtNurea is about 2.0. tion is paid to nutrition, acid-base status, hypo• Rationale A detailed rationale is described in calcemia, hyperphosphatemia, anemia, hyperten• Appendix A. The following is a summary. sion, volume status, and dialysis access (vascular It is paradoxical that nephrologists have fo• cused on optimizing urea clearance once patients are started on dialysis, but have accepted much © 1997 by the National Kidney Foundation, Inc. lower levels of renal urea clearance during the 0272-6386/97/3003-0211 $3.00/0 pre-dialysis phase of patient management. For·

S70 AmericanJournal of Kidney Diseases, Vol SO, No 3, Suppl 2 (September), 1997: pp S70-S73 INITIATION OF DIALYSIS S71

example, a weekly total (residual renal plus peri• (see Guidelines 3: Frequency of Delivered PD toneal dialysis) KtlVurea(KprtlVurea) of 2.0 or Dose and Total Solute Clearance Measurement higher is associated with improved outcomes in Within Six Months of Initiation and Guideline patients on PD (see Guideline 15: Weekly Dose 5: Frequency of Measurement of KtlVure .. Total of CAPD), yet dialysis is usually not initiated CCr, PNA, and Total Creatinine Appearance). Al• until weekly KrtlVurea falls to the range of 0.71 ternatively, the initiation of a "full dose" of PD to 1.3. There is no definitive direct proof for the may be offered (equivalent of four 2 L exchanges belief that a given level of urea clearance by the per day, which may yield a weekly KptIV urea of kidney is associated with better control of uremia 1.5 to 2.0, depending on transport characteristics, than PD with this same urea clearance. In fact, ultrafiltration, and body size). With initiation of recent studies suggest that the relationship be• "full dose" PD, frequency of measurement of tween protein intake and weekly KtIV urea is RRF can be less intense. nearly identical in patients with chronic renal The Work Group strongly supports the opinion failure not yet on dialysis and in patients on PD. that the PD outcome data for a weekly KtlVurea Thus, until proven otherwise, residual renal and of ~2.0 are so compelling that using the same peritoneal clearances of small solutes should be figure for initiation of dialysis justifies the small considered equivalent. risks of performing peritoneal dialysis. Those Once KrtIV urea falls below 2.0 per week, pa• risks include infections and the possibility that tients should be considered at increased risk for increasing the length of time on PD contributes malnutrition and uremic complications. With fur• to eventual patient "burn-out." If a patient is ther decreases in KrtIV urea in the absence of renal suspected to be at high risk for these complica• replacement therapy, the risk increases. Dialysis, tions, PD may not be the best choice for renal or some form of renal replacement therapy, replacement therapy. The Work Group acknowl• should be strongly considered when KrtIV urea falls edges that the risks of early initiation of PD are below 2.0 (or CCr falls in the range of 9 to 14 not clearly known, but that the risks of late initia• 2 mLJminl1.73 m ) and definitely implemented if: tion are known and are unacceptable. Further• 1. There has been an unintentional decrease more, not knowing which initiation strategy (in• in edema-free body weight. Supportive objective cremental versus full therapy initiation) is better, parameters for malnutrition include a lean body the Work Group recommends that either ap• mass <63%, subjective global assessment score proach be used to reach or exceed targets. indicative of malnutrition (see Guideline 12: Nu• Compared to CAPD, it is more complex to cal• tritional Status Assessment and Appendix B: De• culate the incremental dose of hemodialysis (HD) tailed Rationale for Guideline 2) and a serum that would be needed such that the total continuous albumin concentration that is decreasing even delivered weekly KtIV urea would be greater than within the normal range or is less than the lower 2.0. However, it can be estimated using the funda• limit for normal for that lab; or mental assumption underlying CAPD, that at the 2. nPNA <0.8 g/kgJd due to spontaneous, un• same protein catabolic rate, continuous renal re• planned, and unsupplemented protein restriction placement therapy must keep the steady state BUN (see Guideline 2: When To Initiate Peritoneal equal to the average pre-hemodialysis BUN (see Dialysis-nPNA Criteria and Appendix B: De• Appendix G for further discussion of this assump• tailed Rationale for Guideline 2); or tion). If weekly K,tIV urea is 1.6 , for example, a 3. There are clinical signs or symptoms attrib• one time per week HD treatment must deliver an utable to uremia. equilibrated (double-pool) KtIV urea of 2.0 to achieve If PD is initiated, the KptIV urea could be in• a total continuous weekly KtIV urea equivalent to creased incrementally so the combined weekly 2.0. This is quite difficult to achieve, so two HD value of KrtIV urea + KptIV urea (KprtIV urea or total treatments per week may be more realistic for this KtlVurea ) does not fall below the target level of level of RRF. If weekly KrtIV urea is 0.5, two HD 2.0. With the incremental initiation approach fre• treatments must each deliver an equilibrated (dou• quent measurement of residual renal function ble-pool) KtIV urea of 2.0 to achieve a total continu• (RRF) will be necessary to assure that total deliv• ous weekly KtIV urea equivalent to 2.0. This is also ered solute removal does not drop below targets quite difficult to achieve, so three HD treatments S72 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

per week may be more realistic for this level of progressive renal failure is associated with spon• RRF. More technical details about intermittent HD taneous anorexia and malnutrition. Under these are described in Appendix A, including the role of circumstances and in the absence of comorbid biocompatible membranes to help preserve RRF. causes of anorexia, and after unsuccessful inter• It is a general consensus that patients with vention by a registered dietitian, dialysis should diabetes should initiate dialysis at levels of RRF be started in adult patients when nPNA spontane• higher than in patients with causes of ESRD other ously falls below 0.8 g/kg/d. than diabetes. That practice is not altered by this Rationale A detailed rationale is presented in guideline. Appendix B. The following is a summary. The Work Group also recognizes that for many Decreasing renal function in chronic renal fail• clinicians, initiating dialysis based on KtNurea is ure is associated with spontaneous reductions in a new concept. Therefore, we have attempted protein intake and deterioration in nutritional sta• to equate this to the traditional measure of urea tus. Worse nutritional status at the initiation of

clearance, CCo and GFR (estimated by the arith• chronic dialysis is predictive of poorer clinical metic mean of urea and creatinine clearance). outcomes on PD over the subsequent months. A The Work Group recognizes that the patient weekly KprtN urea of 2:2.0 is typically required to will playa major role in accepting the initiation maintain a nPNA of 0.9 g/kg/d in CAPD patients of dialysis based on a certain' 'laboratory value." or those with chronic renal failure. In chronic It is the responsibility of the care providers to renal failure with negligible urinary protein make clear to the patient the rationale for initiat• losses, many patients are at increased risk for ing dialysis when the above conditions become protein malnutrition at a level of protein intake applicable. In particular, the nephrologist must less than 0.8 g/kg/d. It is recognized that an ade• explain to the patient the risk of malnutrition quately supplemented, tightly monitored, low with delayed initiation of dialysis and the strong protein diet may slow the progression of renal inferential evidence that survival might be im• failure in certain circumstances15 and may delay proved with an earlier start of dialysis. Thus, the need for dialysis. When properly performed, appropriate patient education regarding an in• such diets do not result in loss of lean body mass formed decision about dialysis is necessary. and are associated with other features of malnu• Medical conditions that may explain why dialysis trition. A registered dietitian should be involved is not being initiated when weekly KrtNurea is in the care of these patients and caloric intake less than 2.0 need to be documented. These con• should exceed 35 kcal/kg/d if the diet was pre• ditions are described above. scribed or designed specifically for a protein in• Some individuals have expressed concern that take <0.8 g/kg/d (a planned low-protein diet). this guideline will run afoul of the Health Care With urinary protein losses in the nephrotic range Financing Administration (HCFA) regulations and/or PD dialysate protein losses, a nPNA of regarding the initiation of dialysis (e.g., form 0.9 g/kg/d or more may be necessary to maintain 2728, ESRD Medicare Medical Evidence Re• nitrogen balance. Comorbid causes of anorexia port). The leadership of the NKF-DOQI is work• such as gastroparesis, infection, acidosis, and de• ing with HCFA to ensure that this will not be pression should be identified and treated. How• the case. ever, if these measures fail to restore nPNA to 0.8 g/kg/d or higher, the Work Group recom• GUIDELINE 2 mends that dialysis be initiated to maintain Kti V urea targets that are usually associated with ade• When to Initiate Dialysis-nPNA Criterion quate levels of protein intake. (Opinion) The recommendation to use an nPNA criterion It is recognized that an adequately supple• for the initiation of dialysis does not imply that mented, tightly monitored, low protein diet may a broader view of nutritional status should be slow the progression of renal failure in certain neglected. On the contrary, the presence of symp• circumstances. When properly performed, such toms, such as the following, are all factors that diets do not result in loss of lean body mass or should influence the clinical judgment regarding, other manifestations of malnutrition. However, the need for dialysis: INITIATION OF DIALYSIS S73

• anorexia RECOMMENDATIONS FOR RESEARCH • weight loss Although data from available studies indi• • nausea rectly support our recommendation to initiate • vomiting • falling caloric or protein intake by history dialysis when residual renal Kt/V urea falls be• • decreased fat-free, edema-free (lean) body low 2.0 per week, this recommendation can mass by creatinine kinetics (see Guideline 17: only be validated by a prospective, randomized Determining Fat-Free, Edema-Free Body Mass) controlled trial in which selected weekly levels or other measures of Kt/Vurea at the start of dialysis (e.g., Krtl • declining subjective global assessment of V urea 1.0 and Krt/V urea 2.0) are related to clinical nutritional status (see Appendix F: Detailed Ra• outcomes. tionale for Guideline 15) In addition, while available retrospective • decreasing serum albumin concentration and prospective cohort studies strongly suggest A critical value of nPNA below which dialysis that dialytic intervention should be initiated be• should be started has not been identified in chil• fore nutritional status is compromised (i.e., dren. However, nutritional status, specifically as when nPNA <0.8 g/kg/d), this issue will not it pertains to growth, should be considered a be resolved until a prospective randomized trial mandatory part of each child's evaluation. is undertaken which allows comparison of the The registered dietitian from the dialysis fa• clinical outcomes of patients starting on dial• cility to whom the patient will be referred ysis at nPNA levels below 0.8 g/kg/d to those should be involved in the care of the patient at whose levels are significantly higher, e.g., these early pre-dialysis stages. Doing so will ~ 1.0 g/kg/d. Data from the CANUSA Study 16 provide continuity of nutritional care for the suggest that a follow-up period of about 2 years patient as he/she enters the dialysis program would be adequate for such a study, assuming and assures that a registered nurse is actually a sufficient number of patients are enrolled in available. the trial. II. Measures of Peritoneal Dialysis Dose

GUIDELINE 3 measurements in the first 6 months is to establish a more accurate baseline excretion of creatinine. Frequency of Delivered PD Dose and Total Two measurements within the first 6 months Solute Clearance Measurement Within Six are probably sufficient if the results are similar. Months of Initiation (Opinion) Based on our collective personal experience, the Work Group believes that patient compliance The total solute clearance (delivered PD dose with a prescribed PD regimen is highest soon plus residual renal function)s should be measured after initiation of PD, e.g., within the first 6 at least twice and possibly three times within the first 6 months after initiation of PD. For patients months; hence, this period is used to establish a initiating dialysis for the first time and/or patients baseline. with substantial residual renal function, the first Delivered peritoneal dialysis dose depends on many factors, including the transport properties measurement should be performed approxi• of the peritoneal membrane, assessed by the peri• mately 2 to 4 weeks after initiation of PD. For toneal equilibration test (PET).18,19 There is evi• patients transferring from another renal replace• ment therapy to PD and/or for patients who do dence that the PET performed within the first not have substantial residual renal function, the week after initiation of PD may yield higher first measurement of delivered dose of PD should transport results than a PET performed a few weeks later. 20 This difference is statistically sig• be made by 2 weeks after initiation of PD. To nificant, but may not be clinically relevant. It establish a baseline, at least one and possibly two additional measurements will need to be per• may be more convenient to perform the first PET formed in the subsequent 5 months. The fre• at the end of training, rather than at the end of quency of measurement of residual renal function the first month, and the Work Group thinks this depends on the PD prescription of incremental is acceptable. However, the results after a month vs. full dose (see Table II-I). of PD may more accurately reflect peritoneal transport properties for the subsequent period. Rationale Adequate total solute clearance (delivered dose of PD plus residual renal func• Table 11-1. Peritoneal Dialysis Dose and Total Solute tion) will improve patient outcomes (see Guide• Clearance Measurement Schedule: Initial 6 Months line 15: Weekly Dose of CAPD). To assure deliv• ery of adequate solute clearance, measurements PD Fluid PET Urine' for solute clearance are required. In dialysis, as Month Ccrp K,W,,,,a CCrr in other human endeavors, continuous education KpWurea and repetition of a process diminish the fre• 1t X X X X X quency of errors, or at least increase the likeli• 2:\: Y Y hood of recognition of such errors and compensa• 3:\: Y Y tion for them. Furthermore, physiological 4:\: X X X X 5:\: Y Y variations occur which must be taken into ac• 6:\: X X X X count. 17 These lines of reasoning apply to mea• surement of the dose of PD. Thus, measurement NOTE. X, measurement; Y, additional measurement if of delivered PD dose should be repeated periodi• "incremental" PO utilized. cally. The recommendation to measure CCr and • For patients who void infrequently «3 times in 24 hours), collect urine over a 48-hour period. KtIV urea three times within the first 6 months re• t If possible, at the end of month 1, but at the end of lates to items discussed in Guideline 7: PD Dose training if that is more convenient. Troubleshooting, specifically, establishing a :\: The measurement interval in months 2 to 6 is flexible. baseline creatinine excretion and following resid• At least one additional measurement after the first ual renal function (RRF). The rationale for three month's measurement is necessary. If the results of the second measurement are similar to those of the first mea• surement, an adequate baseline is established, obviating the third measurement. If the result of the second mea• © 1997 by the National Kidney Foundation. Inc. surement is discrepant, a third measurement is necessary 0272-6386/97/3003-0212$3.00/0 to establish a more reliable baseline.

S74 AmericanJournal of Kidney Diseases, Vol 30, No 3, Suppl 2 (September), 1997: pp S74-S79 MEASURES OF PERITONEAL DIALYSIS DOSE S75

In patients initiating ESRD therapy for the first measures of PD dose, total weekly KtIV urea and time who have some RRF, delaying the PET and total cFeatinine clearance normalized to 1.73 m2 the first measurement of delivered dose for a BSA are the best, because they are most strongly month is safe and appropriate. However, for pa• associated with mortality and morbidity (see tients initiating PD because of transfer from HD Guideline 15: Weekly Dose of CAPD). Addition• and/or for patients who do not have substantial ally, when properly performed, these measures RRF, the first measurement of delivered dose of are reproducible enough to be useful in routine PD should be performed earlier. In the absence clinical practice. of substantial RRF, waiting a month to measure The urea-based measure, KtlVure .. measures delivered dose may result in inadequate dialysis removal of the direct product of protein catabo• for a month. Thus, the Work Group recommends lism. The creatinine clearance (CCr) measures re• that these patients undergo measurement of de• moval of a product of muscle metabolism, which livered dose of PD at 2 weeks post-initiation, provides insight into lean (ie, fat-free, edema• assuming maintenance exchange volumes have free) body mass and possibly into compliance been achieved. Patient care technicians may be (see Guideline 7: PD Dose Troubleshooting). In able to perform these measurements. Guideline 1: When to Initiate Dialysis-KtIV If "incremental" PD is initiated instead of Criterion and Guideline 15: Weekly Dose of "full dose" (see Guideline 1: When to Initiate CAPD, there is a discussion of the comparison Dialysis-KtIV urea Criterion and Appendix A: of these two different measures (see Guideline Detailed Rationale for Guideline 1), RRF must 6: Assessing Residual Renal Function, for a be followed carefully and frequently such that definition of total weekly CCr)' PD dose can be increased as RRF deteriorates. These two recommended measures have While urine production rate is presumed to be a both been used to measure delivered dialysis clue to deteriorating RRF, that is not always the dose. Since each measure provides slightly dif• case.21 Thus, for patients initiating PD with "in• ferent information, the Work Group recom• cremental" PD, the Work Group recommends mends that both measures be used. Both creati• measuring RRF every 2 months. For patients on nine and urea concentration can be obtained "full dose" PD, the Work Group recommends on the same sample of urine, blood, and dialy• measuring RRF with total solute removal mea• sate. No additional samples need to be col• surements every 4 months. If urine production lected to perform both, rather than one, of rate is decreasing, measure RRF every 2 months these measures. Most laboratories perform or as often as needed and considered helpful, but both measures simultaneously (e.g., 6160, at least every 4 months. Once weekly KrtIV falls Chem 6, etc.) on automated equipment and the to less than 0.1, RRF can be considered negligi• cost is the same for one or both measures. ble and its routine measurement can be stopped. Guideline 11: Dialysate and Urine Collections, GUIDELINE 5 addresses this subject again. Frequency of Measurement of KtIV urea, Total C PNA, and Total Creatinine GUIDELINE 4 cn Appearance (Opinion) Measures of PD Dose and Total Solute After 6 months, total KtIV urea, total CCr, and Clearance (Opinion) PNA (with all its components), should be mea• Both total weekly creatinine clearance normal• sured every 4 months unless the prescription has ized to 1.73 m2 body surface area (BSA) and been changed or there has been a significant total weekly KtIV urea should be used to measure change in clinical status (see Table 11-2). delivered PD doses. Rationale Despite the establishment in 1993 Rationale A valid and reproducible measure of ESRD NetworklHealth Care Finance Admin• of PD dose is essential to assess the quantity of istration guidelines that measurements of deliv• dialysis delivered to an individual patient. The ered PD dose and total solute clearance be per• quantity of dialysis is an important component formed twice yearly, a Network Core Indicator of the quality of dialysis. Of the few available review of 1,208 patient charts in 1995 revealed S76 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

Table 11-2. Peritoneal Dialysis Dose and Total Solute cause it strikes a balance: every 4 months is Clearance Measurement Schedule After 6 Months often enough to be clinically helpful, but not so often as to be intrusive into a patient's life• PD Fluid Urine" style or to create a burden for the dialysis facil• Month KpWurea Ccrp KrWurea Ccrr ity. Awareness of loss of RRF must be para• mount. If "incremental" PD is initiated in• 7 stead of "full dose" (see Guideline 1: When 8 Xt Xt 9 to Initiate Dialysis-KtlV Criterion and Ap• 10 X X X X pendix A: Detailed Rationale for Guideline 1), 11 RRF must be followed carefully and fre• 12 Xt Xt quently such that PD dose can be increased as 13 RRF deteriorates. While urine production rate 14 X X X X is presumed to be a clue to deteriorating RRF, 21 NOTE. X, measurement. that is not always the case. Thus for patients " If incremental PD is still being utilized at this point, the initiating PD with "incremental" PD, the frequency of RRF testing applies as described in Table 1 Work Group recommends measuring RRF ev• of Guideline 3: Frequency of Delivered PD Dose and Total ery 2 months (see Table 11-2). For patients on Solute Clearance Measurement Within Six Months of Initi• "full dose" PD, the Work Group recommends ation. For patients who void infrequently «3 times in 24 measuring RRF with total solute removal mea• hours), collect urine over a 48-hour period. Urine testing can cease when the residual renal function component is surements every 4 months. If urine production a weekly KrWurea < 0.1. rate is decreasing, measure RRF every 2 t For children who have greater difficulty with accurate months, or as often as needed and considered urine collection than adults, this may be deferred until full helpful, but at least every 4 months. Once urine and dialysate collections occur every 4 months (see weekly KrtlV falls to less than 0.1, RRF can Guideline 11, Dialysate and Urine Collections). be considered negligible and its routine mea• surement can be stopped. Guideline 11: Dialy• sate and Urine Collections, addresses this sub• ject again. that data sufficient to calculate such measures The impact of a change in prescription should were available for only one third of the patients. be assessed within 2 to 4 weeks in order to deter• The preliminary 1996 results reveal that 69% of mine if the recommended change has actually the patient charts have such data (written com• been executed and if it has accomplished its goal. munication, Diane Frankenfield, a member of the The promptness of the assessment is important HCFAlHSQB ESRD Core Indicators PD Sub• because clinical events could occur in the interval committee, December 4, 1996; also available on which could postpone the measurement or con• the Internet at http://www.hcfa.gov/medicare! found the results (see Guideline 10: Timing of hsqblhsqbl.htm). Measurement). Measurements of delivered PD dose and to• Some clinical events may impair the quality tal solute clearance are easy to perform, but of delivered PD. A change in clinical condition require attention to detail and precision in which warrants measurement of delivered PD technique for patients and dialysis staff. A va• dose is defined as any serious problem which riety of clinical and psychosocial events can affects nutritional status, the ability of the pa• interrupt and invalidate these measurements. tient to perform PD mechanically or techni• It is imperative that these measurements be• cally (such as stroke or arthritis, loss of surface come a routine for the patients and facility area from surgery, decreased exchange vol• staff. Setting a goal of performing measure• umes due to hernias, etc.), or permanently af• ments every 4 months builds flexibility and fects the transport properties of the peritoneum leeway into a complex care plan, while assur• (e.g., protracted peritonitis). Many conditions ing that a lengthy interval of possibly inade• that lead to hospitalization fall into one of quate PD does not occur. The 4-month interval these categories. Any suggestion of exacerba• between complete measurements of total Ktl tion of uremia should prompt a measurement V urea, total CCn and PNA is recommended be- of delivered dose of PD. MEASURES OF PERITONEAL DIALYSIS DOSE S77

The Work Group recognizes that technical An alternative measure of RRF is residual re• problems in urine collections in some children nal ure~ clearance, normalized to total body wa• may justifiably decrease the frequency of urine ter, KrtlVurea. This measure can be directly added collections in selected cases. to the peritoneal urea clearance component, Kptl GUIDELINE 6 V ure., to create the total urea clearance normal• ized to total body water, Kprtl V (shortened to Assessing Residual Renal Function urea KtlV )· (Evidence) urea Creatinine clearance corrected for renal secre• Residual renal function (RRF), which can pro• tion and KtlVurea are both valuable measures in vide a significant component of total solute and the management of PD patients. Each measure water removal, should be assessed by measuring offers different information. Since the dialysate the renal component of KtIV urea (KrtIV urea) and and urine collections are being performed for ei• estimating the patient's glomerular filtration rate ther measure, the Work Group recommends that (GFR) by calculating the mean of urea and creati• both measures be determined. nine clearance. Rationale A detailed rationale is presented in GUIDELINE 7 Appendix C. The following is a summary. During the first few years of dialysis therapy, PD Dose Troubleshooting (Opinion) residual renal function (RRF) contributes sig• In adult patients, a daily creatinine excretion nificantly to total solute and water removal. P~s­ in urine and dialysate that differs from the ervation of RRF may be particularly important baseline rate (as determined during the first 6 to the effectiveness of long-term PD. For solute months in Guideline 3, Table II-I) by > 15% removal targets to be met (see Guideline 15: should prompt an investigation for noncompli• Weekly Dose of CAPD and Guideline 16: ance, improper collection of drained dialysate Weekly Dose of NIPD and CCPD) in many pa• and/or urine, or altered peritoneal transport tients without a possibly unacceptable dialytic function. Compliance should not be assessed burden, there must be a substantial contribution by comparing measured to predicted creatinine from RRF. excretion. As GFR declines over time, the contribution Rationale Twenty percent of PD patients re• of secreted creatinine to total creatinine clearance port some non-compliance with their dialysis (CCr) rises disproportionately and CCr becomes 3 an inaccurate marker of GFR. Since the perito• prescription. Preliminary data suggest that to• neal membrane does not secrete solute, the GFR tal daily creatinine excretion or appearance can measure that corrects for creatinine secretion is be used as an indicator of compliance in CAPD. the preferred measure to add to peritoneal clear• The premise for such use is that, in noncompli• ance. In the case of a low GFR, the measurement ant patients who perform the proper number of of GFR with endogenous solutes is best done by exchanges only during the day of the clearance defining GFR as the arithmetic mean of urea and measurement,22 the amount of creatinine ex• creatinine clearance. This arithmetic mean essen• creted in 24 hours (equal to the daily amount tially corrects for secretion of creatinine. This of creatinine in the spent dialysate and urine GFR measure is added to peritoneal CCn normal• plus an estimated amount of creatinine lost ized to 1.73 m2 of body surface area and is totaled through other routes, primarily the gastrointes• for a week. This is the "total weekly creatinine tinal tract) will exceed the amount of creatinine clearance. ' , produced daily. Essentially, noncompliance creates an unsteady state of recently accumu• GFR lated creatinine. Thus, an increase in the daily renal urea clearance (mL/min) ] [+ renal creatinine clearance (mL/min) excretion of creatinine in dialysate plus urine (mL/min) = ------may indicate noncompliance just prior to the 2 collection.23 Other potential causes of variation Total weekly CCr = GFR + Peritoneal CCr in the measured amount of creatinine excreted normalized to 1.73 m2 of Body Surface Area include changes in muscle mass (see Guideline S78 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

13: Determining Fat-Free, Edema-Free Body eration is expressed as the ratio of measured/ Mass), improper collection of dialysate or predicted creatinine generation.22 urine due to timing errors, and inaccurate urine The use of a cut-off value of measured/pre• or dialysate creatinine measurement by the lab• dicted creatinine generation to identify non• oratory.23 Finally, another potential cause of compliance is not warranted because the mea• change in total creatinine excretion may be sured/predicted creatmme generation in peritoneal membrane transport dysfunction. compliant CAPD patients appears to vary Quantitatively, a very large transport defect widely.22,30,31 In addition, the increase in the must occur to result in 15% variation in the amount of creatinine excreted during the clear• daily creatinine excretion. However, if this is ance day in noncompliant patients is very suspected, a peritoneal equilibration test (PET) small in most cases.32-34 Measured/predicted or its alternative should be performed. creatinine generation is better used sequen• A similar approach should be considered in tially in a patient after establishing baseline children, although only a small number of pedi• values during a period of close observation atric patients have been studied in this man• (see Table 11_1).30,34 At least for short periods ner. 24 Furthermore, in growing children with of time (days, weeks), the steady-state excre• increasing muscle mass, there will be an in• tion of creatinine is constant in CAPD pa• crease in total creatinine excretion over time. tients,31,35 although a variation of 15% may be The Work Group's decision to use a variance essentially physiologic.25 However, since ab• of> 15% in creatinine appearance over the es• solute creatinine excretion is being measured tablished baseline was based on convincing but and compared to a previously established reli• indirect evidence in adult patients. There are able baseline measurement, the Work Group no data to support a similar approach in chil• considers the use of predicted creatinine pro• dren. In PD patients who appear to be stable, duction to be unnecessary. creatinine appearance may vary by up to 15%, In summary, the Work Group recommends depending on a variety of factors. 25 This vari• establishing a baseline creatinine excretion on ance of > 15% is simply a suggestion or warn• the basis of 2 to 3 measurements in the first 6 ing to the clinician that the creatinine excretion months of PD (see Guideline 3: Frequency of data are not consistent with prior evaluations Delivered PD Dose and Total Solute Clearance and suggests a need for further investigation. Measurement Within Six Months of Initia• The intensity of the investigation that the vari• tion). The Work Group feels that following ance triggers is a clinical decision that requires these measures longitudinally will be more taking many issues into consideration. helpful than comparing measured to predicted The Work Group does not recommend as• creatinine appearance. Causes for any subse• sessing compliance by comparing measured quent deviation from the baseline total creati• creatinine excretion to predicted creatinine ex• nine excretion over time should be sought, rec• cretion. Our reasoning is as follows: The esti• ognizing that noncompliance is only one of mated amount of creatinine lost in the gut is several possible explanations. equal to 0.036 X serum creatinine concentra• RECOMMENDATIONS FOR RESEARCH tion in mg/dL X body weight.26 The predicted creatinine production, in mg/day, is calculated Since preservation of RRF is important for by the Cockroft-Gault formulae27 as follows: solute removal and contributes to total renal re• placement therapy, there is a need to identify For men: (28 - 0.20 X Age) X Weight contributors to loss of RRF. For example, do antibiotics playa role? Hypotension? Other fac• For women: (24 - 0.17 X Age) X Weight tors? Does aggressive solute removal and/or where age is in years and weight is in kilo• highly efficient dialysis remove stimulatory fac• grams.22,28,29 These formulae were derived in a tors favoring remnant kidney hyperfiltration? nondialysis population. The discrepancy be• Why does GFR vary so much on a day-to• tween measured and predicted creatinine gen- day basis and how does one account for this in MEASURES OF PERITONEAL DIALYSIS DOSE S79

adequacy studies? Is this a collection artifact or children who are growing, how often should total true physiologic variation? What factors alter the creatinine excretion be measured and is it useful daily production and excretion of creatinine? Is as an assessment of compliance with dialysis pre• it a function of creatinine production or simply scription? excretion? If the latter, is it variation in renal The recommendation that > 10% variance in creatinine appearance be considered as indicative secretion, filtration, or both? Urine output also of a status change or noncompliance should be varies dramatically on a day-to-day basis. Is this validated. simply a volume phenomenon or is it a reflection An accurate, reproducible and easy-to-per• of true clearance changes? Why does creatinine form method of measuring RRF should be de• appearance vary even in compliant patients? In veloped. III. Measurement of Peritoneal Dialysis Dose

GUIDELINE 8 dwell. For most NIPD and CCPD patients, these time points occur in the early afternoon. Reproducibility of Measurement (Opinion)

Accurate measurement of total KtIV urea and to• GUIDELINE 9 tal creatinine clearance (CCr) requires collection Estimating Total Body Water and Body and analysis of urine, dialysate, and serum in a Surface Area (Opinion) way that yields reproducible and valid results. Dialysate creatinine concentration must be cor• V (total body water) should be estimated by 36 3 rected for the presence of glucose in some assays. either the Watson or Hume ? method in adults Peritonitis precludes reliable measurement of de• using actual body weight, and by the Mellits• 38 livered PD dose for up to a month. Compliance Cheek method in children using actual body with complete collections is mandatory. For pa• weight. 36 tients who void ~3 times per day, a 24-hour Watson method : urine collection is sufficient. For patients who For Men: V (liters) = 2.447 + 0.3362*Wt (kg) void less frequently, a 48-hour collection is rec• + 0.1074*Ht (cm) - 0.09516*Age (yrs) ommended. For CAPD patients, the serum sam• For Women: V = -2.097 + 0.2466*Wt + ple can be obtained at any convenient time. For 0.1069*Ht 3 NIPD patients, the serum sample should be ob• Hume method ?: tained at the midpoint of the daytime empty pe• For Men: V = -14.012934 + 0.296785*Wt riod. For CCPD patients, the serum sample + 0.192786*Ht should be obtained at the midpoint of the daytime For Women: V = -35.270121 + O.l83809*Wt + 0.344547*Ht dwell(s). 38 Rationale A detailed rationale is presented in Mellits-Cheek method for children : Appendix D. The following is a summary. For Boys: V (liters) = -1.927 + 0.465*Wt Measurement of PD dose must be performed (kg) + 0.045*Ht (cm), when Ht ::£132.7 cm in a valid and reproducible fashion. The measure V = -21.993 + 0.406*Wt + 0.209*Ht, when of creatinine concentration in effluent dialysate height is ~ 132.7 cm must be corrected for the presence of glucose For Girls: V = 0.076 + 0.507*Wt + O.013*Ht, with some creatinine assays. Each facility must when height is -s: 110.8 cm determine whether this is necessary by specifi• V = -10.313 + 0.252*Wt + 0.154*Ht, when cally inquiring of its laboratory whether the cre• height is ~ 110.8 cm Body surface area, BSA, should be estimated atinine assay used by that lab is altered by high 39 glucose concentrations. PD dose measures by either the DuBois and DuBois method , the Gehan and George method40, or the Haycock should not be made until a month after peritonitis 41 resolves, because peritonitis causes residual ef• method using actual body weight. fects on membrane transport. Either total dialy• For all formulae, Wt is in kg and Ht is in cm: sate collections or aliquots (i.e., samples of dialy• DuBois and DuBois method: 2 725 sate) can be used with proper patient training and BSA (m ) = 71.84*Wt°.425*Hto. compliance. For CAPD patients, the timing of Gehan and George method: 2 the blood sample is not important. For patients BSA (m ) = 0.0235*Wt051456*Ht°.42246 on NIPD or CCPD, the blood sample must reflect Haycock method: 2 the overall average for the entire 24 hours. For BSA (m ) = 0.024265* WtO.53?8*Ht03964 NIPD patients, the serum sample should be ob• Rationale A detailed rationale is presented tained at the midpoint of the daytime empty pe• in Appendix E. The following is a summary. riod. For CCPD patients, the serum sample The Watson and Hume formulae were de• should be obtained at the midpoint of the daytime rived by comparing total body water measure• ments to simple anthropometric measurements (weight, height, age) in subjects without © 1997 by the National Kidney Foundation, Inc. edema, volume deficit, or end-stage renal dis• 0272-6386/97/3003-0213$3.00/0 ease. In peritoneal dialysis patients, the Wat-

S80 AmericanJournal of Kidney Diseases, Vol 30, No 3, Suppl 2 (September), 1997: pp S80-S82 MEASUREMENT OF PERITONAEAL DIALYSIS DOSE S81

son and Hume formulae provide reasonable when the patient is clinically stable (eg, stable approximations of isotopic body water mea• weight,. stable BUN and creatinine concentra• surements. Volume abnormalities (edema) are ap• tions), and at least 4 weeks after resolution of parently the major cause of discrepancy. The Mel• peritonitis. lits-Cheek formulae were derived from subjects aged Following a change in prescription or a major 1 month to 34 years for males and 1 month to 31 change in clinical status (eg, hospitalization, years for females. In each case, the measurement of weight loss), but in the absence of recent perito• total body water was performed in normal subjects nitis, measurements of delivered weekly KtIV urea by the use of deuterium oxide distribution, with si• and total weekly CCr should be performed within multaneous measurement of weight and height. the next 4 weeks and then at 4-month intervals. The Work Group recommends the use of the Rationale The effect of body weight on the Watson or Hume formulae in adults and the Mellits• calculation of V is discussed in the rationale for Cheek formula in children as methods for estimating Guideline 9: Estimating Total Body Water and V. Attention should be paid to the presence of edema Body Surface Area and in Appendix E: Detailed at the time of the clearance study. A special case is the underweight patient (see Table X-I, Appendix Rationale for Guideline 9. Variations in serum E for a definition). Successful efforts to restore urea and creatinine concentration can potentially weight to a normal level in such a patient will result increase the error in the clearance calculations, in a rising V, and consequently in a proportionally and indicate that the patient is not in a steady declining KprtIV urea' This does not alter the meth¢• state. ology of estimating total body water using actual Peritonitis may, in some instances, affect peri• weight. It does affect target doses of dialysis, how• toneal solute transport for long periods. The ra• ever. This issue is discussed again in Guideline 15: tionale for waiting 4 weeks after resolution of Weekly Dose of CAPD. peritonitis to repeat the clearance studies is pre• Like the formulae above for total body water, sented in Appendix D: Detailed Rationale for the formulae for BSA were determined in a normal Guideline 8. population. Many of the disclaimers described for The Work Group recommends frequent assess• calculating V are less of an issue in calculating BSA, ment of delivered dose of PD and total solute because the relationship to the defining simple an• clearance, specifically every 4 months after the thropometric measurements is less influenced by first 6 months of PD (see Guideline 11: Dialysate clinical conditions accompanying ESRD. Histori• and Urine Collections). Major changes in clinical cally, many nephrologists have utilized the method status (eg, patient compliance, weight gain, of DuBois and DuBois, and much of our data are weight loss, technicaVmechanical complications, from its application. Only nine subjects were used 39 some causes of hospitalization) may alter PD to define this formula. More than 400 subjects, dose requirements. For example, pneumonia may including many children, were used to define the contribute to loss of residual renal function, formula of Gehan and George,40 and in an indepen• which would be undetected unless measured. dent comparison, the Gehan and George method 42 Therefore, in the absence of peritonitis, a major was preferred. The Haycock formula is based on measurements of 81 subjects ranging from prema• change in clinical status should prompt a re-eval• ture infants to adults.41 uation of weekly KtIV urea and total weekly CCr, Amputation alters the relationship between body and this should occur within one month follow• height and weight. This causes a mathematical dis• ing the change in clinical status. Within a month tortion of the calculation of both anthropometric V following a PD prescription change, weekly Kti and BSA, because the calculation of each takes this V urea and total weekly CCr should be measured 43 relationship into account. ,44 Modification of V and to demonstrate that the goals of the prescription BSA in patients with amputations are described in change have been achieved. If the patient is not detail in Appendix E. stable, all attention should be directed to de• termining the cause of the instability and to cor• GUIDELINE 10 recting it. This mayor may not include measur• Timing of Measurement (Opinion) ing the delivered dose of PD. There will be Routine measurements of total KtIV urea and to• circumstances in which the change in clinical tal creatinine clearance should be performed status might only alter RRF (eg, exposure to S82 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

nephrotoxins), not delivered dose of PD. While subjects followed longitudinally.46,47 The CA• one must be cautious in assuming that persistent NUSA study demonstrated substantial loss of re• urine flow rate implies stable RRF,21 a clinical nal function at 6-month intervals.47 The Work clue that deterioration has occurred may be a Group concludes that measurements of urinary decrease in what previously had been a stable clearances should be performed at 2-month inter• daily urine volume.45 In those settings, only mea• vals to prevent long periods of underdialysis. For surement of RRF is indicated. children in whom urine collections are difficult (requiring special collection apparatus, etc.), GUIDELINE 11 urine collections can be deferred until the next Dialysate and Urine Collections (Opinion) total solute removal measurement (see Guideline 3: Frequency of Delivered PD Dose and Total Two to three total solute removal measure• Solute Clearance Measurement within Six ments are required during the first 6 months of Months of Initiation and Guideline 5: Frequency PD (see Guideline 3: Frequency of Delivered PD of Measurement of KtIV urea, Total Cen PNA, and Dose and Total Solute Clearance Measurement Total Creatinine Appearance). Within Six Months of Initiation). After 6 months, if the dialysis prescription is unchanged: RECOMMENDATIONS FOR RESEARCH 1. Perform both complete dialysate and urine The optimal timing of blood sampling for sub• collections every 4 months; and jects on asymmetric PD (NIPD, CCPD) should 2. Perform urine collections every 2 months be determined. The recommendations we have until the renal weekly KrtIV urea is <0.1. made are based on pharmacokinetic theory. Thereafter, urine collections are no longer nec• Comparison of the "batch" and "aliquot" essary, as the RRF contribution to total KtIV urea methods of dialysate sampling should be studied. becomes negligible. In children, urine collections Development of a clinically applicable method are recommended only with complete dialysate of assessing TBW in children undergoing PD is collections (see Table 11-2 reproduced from recommended. Guideline 5.) Methodology to calculate renal urea and creat• Rationale Loss of residual renal function is inine clearance and PNA from random urine the major cause of decreasing clearance in PD samples should be developed. IV. Assessment of Nutritional Status Specifically as it Relates to Peritoneal Dialysis

GUIDELINE 12 Group's opinion that adequate renal replacement therapy is necessary for normal appetite and me• Assessment of Nutritional Status (Opinion) tabolism. Thus, nutritional problems may reflect inadequate dialysis which, if corrected, may lead Nutritional status of adult PD patients should be to subsequent improved outcomes. assessed on an ongoing basis in association with Although nutritional status is influenced by KtIV urea and CCr measurements using the Protein many nondialysis-related factors, appetite sup• equivalent of Nitrogen Appearance (PNA) and pression, nausea, and vomiting are major clinical Subjective Global Assessment (SGA). For pediatric features of uremia and inadequate dialysis. PD patients, nutritional status should be assessed Therefore, nutritional status is also an important using the PNA and other standard nutritional as• measure of PD adequacy. Of the available mea• sessments (see Guideline 14). sures of nutrition, PNA is recommended because Rationale A detailed rationale is presented in it provides an estimate of protein intake and pro• Appendix F. The following is a summary. tein losses. The SGA is recommended because There is strong indirect evidence linking sur• it is a valid clinical assessment of nutritional sta• vival on dialysis with nutritional status both at tus and is strongly associated with patient sur• initiation of dialysis (see Section I: Initiatiog. of vivaL Dialysis) and during longitudinal follow-up. Bet• Protein equivalent of total nitrogen appearance ter survival has been reported in PD patients with (PNA) Nitrogen intake is almost entirely from high normalized protein equivalent of nitrogen protein. The final product of protein catabolism is appearance (nPNA)48 (see Guideline 28: Mea• urea. Therefore, if steady-state nitrogen balance surement of Normalized PNA in PD Patients). conditions exist, one can work backward from Positive correlations between nPNA and clear• urea excretion to determine what the protein in• ance of urea or creatinine have been reported take was. Yet, other protein losses (urinary, peri• repeatedly in PD subjects.49-52 The correlation be• toneal dialysate, diarrhea) also reflect the body's tween nPNA and KtIV urea may indicate increased turnover of protein. Studies in dialysis patients appetite and dietary protein intake as KtIV urea in• show that predictable mathematical relationships creases, but also may simply reflect the fact that exist between urea excretion, protein catabolism,

nPNA and KtlVurea are mathematically linked. 53 and dietary protein intake. If peritoneal protein This mathematical linkage makes the correlation losses are greater than 15 gld, PNA should be between nPNA and KtlVurea in cross-sectional calculated as protein catabolic rate plus protein studies of questionable clinical significance. losses. If dialysate protein losses are less than 15 However, nPNA tends to increase in the same gld, the formula: subjects when KtIV urea and creatinine clearance PNA(g/d) = 1O.76*(O.69*UNA + 1.46) (CCr) are increased by increasing the dose of PD,5! especially if the increase in the dose of PD can be used to calculate PNA where UNA is was prescribed because of inadequate clear• total urea nitrogen appearance in grams per day.57 ances.54 In the latter instance, the association be• PNA is an indirect method for estimating dietary tween KtIV urea and nPNA is not the result of a protein intake, a key measure of nutritional status mathematical coupling. In addition, there is in dialysis patients. For pediatric patients, see strong evidence suggesting that quality and quan• Guideline 14: Use of the Modified Borah Equa• tity of dialysis influences nutrition.55 .56 While the tion to Assess Nutritional Status of Pediatric PD precise relationship between renal function (or Patients. renal replacement therapy) and nutrition is not Subjective Global Assessment (SGA) The SGA is yet adequately understood, it is the Work a simple assessment that uses the clinician's experi• ence to subjectively rate a patient's nutritional status based on the medical history and physical exam. © 1997 by the National Kidney Foundation, Inc. The SGA was modified for use in PD patients as 0272-6386/97/3003-0214$3.00/0 described in Appendix F: Detailed Rationale for

AmericanJournal of Kidney Diseases, Vol 30, No 3, Suppl 2 (September), 1997: pp S83-S85 S83 S84 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

Guideline 12. This assessment is valuable because steady-state of creatinine excretion should exist it does not focus on a single variable; rather, it forces for the equation result to be valid. Factors other the clinician to view the patient more broadly. SGA than muscle mass can affect the fat-free, edema• addresses four items (recent weight change, an• free body mass calculation by creatinine kinetics. orexia, subcutaneous tissue, and muscle mass) These factors include errors in the collection or scored on a 7-point Likert scale (see Appendix B: measurement of creatinine in urine or dialysate, Detailed Rationale for Guideline 2). It can be per• and large variations in the dietary intake of cre• formed by physicians, nurses, or registered dietitians atine plus creatinine (meat). Fat-free, edema-free during routine clinic visits. Several studies have vali• body mass estimates by creatinine kinetics may dated that the SGA accurately reflects nutritional be a better index of nutritional status in PD pa• status in dialysis patients and, in the CANUSA tients, because they reflect dry fat-free, edema• study, a higher SGA was associated with a lower free body mass and changes in muscle mass bet• risk of death. ter than dual-energy X-ray absorptiometry or The SGA is easy to perform and can be per• bioimpedance.6o The day-to-day variability ofto• formed within minutes during a routine clinic visit. tal creatinine excretion is only 2% to 4% over a There are no data to dictate how often to perform short time interval,31,35 but is up to 15% over the SGA, so the Work Group bases its opinion on much longer intervals.25 the following: the SGA should be done often Serum creatinine concentration is not, by itself, enough to detect changes and to intervene in a an index of adequacy of peritoneal dialysis because timely manner. It should be performed in associa• of the large variations in creatinine production be• tion with measurement of KtIV urea and Ceo every tween individuals. However, a change in serum 4 months after the initial 6 months (see Guideline creatinine concentration may indicate changes in

5: Frequency of Measurement of KtlVurea , Total creatinine and urea removal to a much larger extent Ceo PNA, and Total Creatinine Appearance). than a change in serum urea concentration.61 ,62 A The SGA has not been validated as a means rising serum creatinine concentration is usually of nutritional assessment in the pediatric PD pop• caused by a decrease in total creatinine clearance, ulation. often secondary to a loss of residual renal function, and much less frequently by an increase in muscle GUIDELINE 13 mass. A decreasing serum creatinine concentration Determining Fat-Free, Edema-Free Body is caused more often by progressive loss in muscle Mass (Opinion) mass and less often by an increase in total clear• ance. In other words, increases in peritoneal solute Total creatinine appearance should be used to transport or recovery of renal function do not occur determine fat-free, edema-free body mass. very often. Rationale Fat-free, edema-free body mass is probably a more accurate term for what had pre• GUIDELINE 14 viously been called lean body mass. It is an im• portant index of overall nutritional status. Total Use of the Modified Borah Equation to daily creatinine production, measured as the sum Assess Nutritional Status of Pediatric PD of creatinine excreted in dialysate and urine plus Patients (Opinion) the estimated creatinine lost in the gut,26 can be Nutritional status of pediatric PD patients used to calculate fat-free, edema-free body mass. should be assessed at least every 6 months by Fat-free, edema-free body mass reflects somatic standard clinical nutritional evaluations and by protein stores in the same way that serum albu• the modified Borah equation57: min reflects visceral protein stores.58 In adults, PNA (g/d) = [6.49*UNA] + [0.294*V] fat-free, edema-free body mass in kilograms is 58 computed by the equation ,59: + protein losses (g/d) Fat-free, edema-free body mass = 0.029 Rationale The equation described in Guide• line 12, Assessment of Nutritional Status, from X total creatinine production in mgld + 7.38. the glossary57 is a further modification of the Norms vary by patient gender and size. A original Borah equation.63 Since this modification ASSESSMENT OF NUTRITIONAL STATUS SPECIFICALLY AS IT RELATES TO PERITONEAL DIALYSIS S85

has not been validated in children, the Work RECOMMENDATIONS FOR RESEARCH Group recommends using the modification above The precise relationships among PNA, dialysis 57 64 from Kopple et al and Keshaviah et al. dose, and outcome are unclear. Although these Although not validated in children, this modi• relationships are currently being studied in hemo• fied Borah equation contains a factor, V, that dialysis patients, they should also be examined controls for patient size, and has been employed in PD patients. There are probably limits, for in pediatric studies. Furthermore, dialysate pro• example, to the role of increasing delivered dial• tein losses must be measured directly in the dial• ysis dose in order to improve appetite and nutri• ysis effluent and not estimated when using the tional parameters. Other questions of interest in• modified Borah equation in children. The use of clude: How does improved dialysis affect equations in which dialysate protein losses are nutrition, and above what delivered dose does estimated has been studied in very few pediatric that effect dissipate, if at all? What interventions PD patients.65 act synergistically with improved dialysis to im• The dialysate protein measurement is the only prove nutritional parameters? What complica• additional laboratory determination from stan• tions interfere with nutrition, in what manner, dard total solute removal measurements, as de• and can the interference be overridden by another scribed in Guideline 4: Measures of PD Dose type of intervention? and Total Solute Clearance. Thus, this nutritional A standardized nutritional assessment tool for assessment could be easily merged to accompany children analogous to the SGA should be devel• total solute removal measurements. - oped. V. Adequate Dose of Peritoneal Dialysis

78%. These predictions assume that renal and GUIDELINE 15 peritoneal KtlVurea are equivalent. Weekly Dose of CAPD (Evidence) Clinical experience suggests that a total weekly creatinine clearance >50 Ll1.73 m2 is required for adequate dialysis. Among patients For CAPD, the delivered PD dose should be with minimal residual function in the Italian a total KtIV urea of at least 2.0 per week and a study, a weekly KtlV of 1.96 correlated with total creatinine clearance (CCr) of at least 60 LI urea 2 a weekly creatinine clearance (CCr) of 58 L. The wkl1.73 m • CANUSA study reported a 7% decrease in pa• Rationale A detailed rationale is presented in tient survival in association with a 5 Ll1. 73 21 Appendix G. The following is a summary. m wk decrease in CCr- Unlike the situation for Kti Theoretical constructs predict that a weekly V urea. both technique failure and hospitalization peritoneal KtIV urea between 2.0 and 2.25 will pro• were worse with decreased weekly creatinine vide adequate dialysis. These constructs assume clearance. The predicted 2-year survival of 78% no residual renal function, full equilibration of was associated with a weekly KtIV urea of 2.1 or plasma and dialysate urea, a target serum urea a weekly CCr of 70 L. nitrogen concentration between 60 and 80 mgl There are insufficient data to address the issue dL, and nPCR between 1.0 and 1.2 g/kg/d. of adequate compared to optimal dialysis (see Clinical studies addressing the validity of these Introduction.) The latter is in part defined as the predictions can be divided into those using uni• dialysis dose above which the incremental clini• variate and those using multivariate statistical cal benefit does not justify the patient burden or analyses. The former are methodologically financial costs. Nor are there sufficient data to weaker. Four studies which used univariate anal• evaluate the relative importance of renal and ysis suggest that total (renal and peritoneal) peritoneal clearances. The recommendations as• weekly KtlVurea values greater than 1.5, 1.89,2.0, sume equivalence but this requires further study. and 2.0, respectively, are associated with better The correlation between KtIV urea and CCr will patient survival than lower values. vary with residual renal function (see Figure X- Three studies from France, Italy, and North 2, Appendix A). The higher CCr observed in the America (CANUSA) have used multivariate sta• CANUSA study compared to the Italian study tistical analysis. The French study found better was due to the greater residual renal function in survival among patients with an initial weekly the former. KtIV urea> 1.7 but did not evaluate changes in Kti Based on the available evidence, the minimum V urea associated with loss of residual renal func• tion. The Italian study evaluated prevalent CAPD delivered dialysis dose target KtlVurea should be 2.0 per week; the minimum weekly target CCr patients with minimal residual renal function. 2 Improved patient survival was observed with a should be 60 Ll1.73 m • If there is discordance weekly KtlVurea > 1.96. Values higher than 1.96 in achieving these targets, the KtIV urea should be were not associated with increased survival but the immediate determinant of adequacy because the statistical power to detect this association was it directly reflects protein metabolism and is less low. The CANUSA study of 680 incident contin• affected by extreme variations in residual renal uous peritoneal dialysis patients reported a 5% function (see Appendix A and Figure 11-2.) How• decrease in patient survival in association with ever, a cause for the discrepancy should be sought and the patient followed closely for signs every 0.1 decrease in total weekly KtlVurea , for of underdialysis. KtlVurea between 1.5 to 2.3. There was no associ• ation between KtIV urea and technique failure or A special case is the underweight patient, defined hospitalization. The predicted 2-year survival as• in Appendix E: Detailed Rationale for Guideline 9. sociated with a constant total KtIV urea of 2.1 was Successful efforts to restore weight to a normal level in such a patient will result in a rising V, and conse• quently in a proportionally declining ~tIVurea' To © 1997 by the National Kidney Foundation. Inc. achieve a weekly KprtIV urea of 2.0 at the increased 0272-6386/97/3003-0215$3.00/0 weight, the weekly target KprtIV urea provided during

886 AmericanJournal of Kidney Diseases, Vol 30, No 3, 8uppl 2 (8eptember), 1997: pp 886-892 ADEQUATE DOSE OF PERITONEAL DIALYSIS S87

the malnourished state must be greater than 2.0. The in continuous dialysis, (KtlVurea of 4.0 in HD and 66 67 Work Group recommends that the target KvrtIV urea 2.0 in C;APD) holding protein intake constant. • should be raised in a malnourished CAPD patient The Work Group assumed that the delivered dose to the level that would provide a weekly KprtIV urea of NIPD would need to be 8% higher than the of 2.0 for that patient if he or she were at the desired CAPD dose (108% of 2.0 = 2.16, rounded up to weight. That level is calculated by multiplying the 2.2). The Work Group assumed that the requisite target of 2.0 for CAPD times the ratio of V desrreJ delivered dose of CCPD would be intermediate Vactual . This is described in detail in Appendix E: between those for CAPD and NIPD. Some varia• Detailed Rationale for Guideline 9, and discussed tions of CCPD with diurnal exchanges of less in Guideline 17: PD Dose in Subpopulations. The duration than the nocturnal exchange of CAPD same upward target adjustment should be made in may be considered equal to CAPD. However, creatinine clearance. The target creatinine clearance in order to simplify recommendations, the target should be adjusted upward by multiplying the target weekly total dose for CCPD is 2.1. The recom• for that therapy (CAPD or APD) by the ratio of mendations for creatinine clearance are percent• BSA.Jesired /BSAactual . age adjustments corresponding to the changes in Clinical judgment suggests that the target KtIV urea targets for these groups. doses of PD for children should meet or exceed Clinical judgment suggests that the target the adult standards. However, there are currently doses of PD for children should meet or exceed no definitive outcome data in pediatric patients the adult target doses. There are no definitive to suggest that any measure of dialysis adequaty outcome data in pediatrics to suggest that any is predictive of well-being, morbidity, or mortal• measure of dialysis adequacy is predictive of ity. There also are no data regarding the real well-being, morbidity, or mortality. protein needs of children, especially young chil• dren, on dialysis. It is the opinion of the Work GUIDELINE 17 Group that the nutritional requirements per kilo• PD Dose in Subpopulations (Opinion) gram of body weight are higher in children than in adults. Therefore, PD doses in children, and There is no adequate basis for recommending especially small infants who have very high pro• any change in the target doses of dialysis dis• tein intakes, may have to be higher than PD doses cussed in Guidelines 15: Weekly Dose of CAPD, in adults. and 16: Weekly Dose of NIPD and CCPD, for various patient subpopulations (eg, patients with GUIDELINE 16 diabetes or who are elderly), with the exception of the malnourished patient, whose target dose Weekly Dose of NIPD and CCPD (Opinion) is increased by the ratio of the V desired Naclual for For NIPD, the weekly delivered PD dose KtIV urea. For creatinine clearance, the target dose should be a total KtIV urea of at least 2.2 and a in a malnourished patient is increased by the ratio weekly total creatinine clearance of at least 66 B SAdesired /BSAaclual . 2 Ll1.73 m • Rationale There are no data available in the For CCPD, the weekly delivered PD dose literature on which to base a recommendation should be a total KtIV urea of at least 2.1 and a for different adequacy targets for patients with weekly total creatinine clearance of at least 63 diabetes or for the elderly. However, it must be 2 Ll1.73 m • remembered that malnourished patients may ap• Rationale In the absence of data that relate pear to have an adequate KtIV urea due to calcula• delivered dose of automated PD (APD) to patient tion of V from the actual or malnourished body outcomes, targets for NIPD and CCPD are based weight. If V were calculated from an estimate of on opinion. desired body weight, the target would reflect that Theoretically, there is an 8% difference in target body weight. This is discussed in Guide• clearance between CAPD and NIPD. This differ• lines 9: Estimating Total Body Water and Body ence is based on calculations which describe a Surface Area, and 15: Weekly Dose of CAPD, 200% increase in the intermittent HD clearance and in detail in Appendix E, Detailed Rationale required to achieve the same solute removal as for Guideline 9. S88 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

In the absence of data in these sUbpopulations, B. Initiation of Peritoneal Dialysis if no other cause of malnutrition is discovered, 1. If possible, wait 10 days to 2 weeks after the target delivered dose of dialysis should be catheter placement to start PD. increased by multiplying the target KtIV urea for a 2. If PD must be started in less than 10 days normally nourished patient by the ratio of the following catheter placement, do low-volume, VdesiredN actual, and the target creatinine clearance supine dialysis. should be increased by the ratio of BSAdesiredl 3. Obtain baseline 24-hour urine collection for BSAactual. These modifications are described in urea and creatinine clearance (see Guideline 6: Guideline 15: Weekly Dose of CAPD, and Ap• Assessing Residual Renal Function.) These col• pendices E and G. lections are for solute clearance calculations, as• sessment of creatinine generation, and PNA de• GUIDELINE 18 terminations. 4. Note patient's weight and the presence or Use of Empiric and Computer Modeling of absence of edema. PD Dose (Evidence) 5. At initiation of dialysis, explain to patient! Both empiric and computer modeling methods parents/caregivers that the patient's prescription can be used to estimate adequate doses of PD. will be individualized. Specifically, state that Specific prescriptions are described below. their instilled volume almost certainly will need Rationale The Work Group has elected to de• to increase over time. For patients who choose scribe these two empiric and computer modeling Automated Peritoneal Dialysis (APD), one or approaches in detail. They are by no means mutu• more daytime dwells will be needed in approxi• ally exclusive. mately 85% of patients. Patients should know from the start of PD that their total solute clear• EMPIRIC APPROACH FOR ance will be monitored and that, if their residual DETERMINATION OF DOSE OF renal function or peritoneal transport changes PERITONEAL DIALYSIS over time, their prescription may need to change as well. A. General Pre-ESRD Evaluation C. Initial Dialysis Prescription for Adults 1. Explain all options (transplant, HD, and PD) to patients/parents/caregivers in a non-bi• Initial dialysis can be prescribed empirically ased manner. based on patient's weight, amount of residual 2. Review medical conditionlcomorbidities to renal function, and lifestyle constraints. These determine if contraindications, relative or abso• empiric recommendations should be imple• lute, exist for any modality (see Section VIII, mented prior to peritoneal equilibration testing. Suitable Patients for PD). PD may be initiated incrementally, or as full 3. If no medical contraindications exist and therapy, depending on RRF at the time of initia• the patient is a candidate for self therapy, allow tion (see Guideline 1: When to Initiate Dial• patient to choose a modality. ysis-KtlVurea Criterion). For example, if Krt! 4. Place the chronic dialysis access (PD or V urea is 1.8 per week, only 0.2 KptIV urea is needed HD). The Vascular Access Work group recom• per week. Assuming complete urea equilibration mends that vascular accesses be placed in pa• (serum to dialysate) at 6 hours, a single 2-L over• tients on PD. The PD Adequacy Work Group night exchange would contribute 14 L per week. feels that this decision should be made on an If V is 40 L, this contributes a KptIV urea of 141 individual patient basis, but our position does not 40 or 0.35 per week. Any ultrafiltrate would add necessarily disagree with the recommendations further to total solute removal. That, plus the Krt! of the Vascular Access Work Group. Vurea of 1.8, brings the KprtlVurea to at least 2.15, 5. If dialysis is needed at the time of presenta• satisfying the target requirement. This approach tion, place the temporary HD access, or after uses basic principles of dialysis prescription de• placing the PD catheter, initiate therapy as sug• velopment. Thus, the dose of KptIV urea depends gested under point B.2, below. on the KrtIV urea as the Work Group has empha- . ADEQUATE DOSE OF PERITONEAL DIALYSIS S89

sized throughout these guidelines. Keeping in tients with an estimated underlying GFR mind that the weekly KprtIVurea goal is 2.0, the ~ 2 mL/min still apply to urine volume. following more intense empiric approach is rea• Namely, if RRF provides enough diuresis, sonable: NIPD may provide enough solute removal 1. Patients with an estimated underlying GFR for a while. This should be tested early on. >2 mLimin If during training, it is noted that a patient a. If patient's lifestyle choice is CAPD: has very low drain volumes with no appar• 2 BSA <1.7 m ---+ 4 X 2.0 L exchanges/day ent mechanical problem or leak, a PET 2 BSA 1.7 to 2.0 m ---+ 4 X 2.5 L exchanges/ should be done to determine if the patient day is a rapid transporter. If so, NIPD can be BSA >2.0 m2 ---+ 4 X 3.0 L exchanges/day prescribed using kinetic modeling. b. If patient's lifestyle choice is CCPD: BSA <1.7 m2 ---+ 4 X 2.0 L (9 hours/night) D. Initial Dialysis Prescription for Children 2.0 Lid + In view of the close, age-independent relation• BSA 1.7 to 2.0 m2 ---+ 4 X 2.5 L (9 hours/ ship between peritoneal surface area and body night) + 2.0 Lid surface area (BSA), the use of BSA as a normal• BSA >2.0 m2 ---+ 4 X 3.0 L (9 hours/night) ization factor for the prescribed exchange volume 3.0 Lid + in children is preferred. An instilled volume of c. If patient's lifestyle choice is NIPD: at least 1100 mLlm2 is recommended for most Specific attention to certain details will"be pediatric patients, although individual tolerance required. Nightly intermittent peritoneal must be considered.68 dialysis (NIPD) is not a therapy that is typi• It should be emphasized that the preceding cally used at the initiation of dialysis. It prescriptive guidelines are general empiric guide• has been reserved for high or rapid trans• lines for patients initiating PD, generally as first porters. However, in patients with signifi• renal replacement therapy. For patients transfer• cant RRF (and ability to diurese), they may ring from HD with minimal RRF, prompt ade• initially do well on nightly exchanges only quacy testing is required. The above empiric rec• (dry day) because of the supplemental ommendations must be individualized and clearance provided by the patient's RRF. guided by documentation that the delivered dose See further comments on NIPD under point equals the prescribed dose. Furthermore, the in• 2.c, below. stilled volumes are ones that theoretically will 2. Patients with an estimated underlying GFR result in a weekly target KtIV urea of greater than :::;2 mLimin 1.9 for the average patient. High or low transport• a. If patient's lifestyle choice is CAPD: 2 ers may be below target if RRF is low. Finally, BSA <1.7 m ---+ 4 X 2.5 Lid 2 although most patients tolerate instilled volumes BSA 1.7 to 2.0 m ---+ 4 X 3.0 Lid 2 of greater than 2.0 liters, this needs to be evalu• BSA >2.0 m ---+ 4 X 3.0 Lid (consider use ated for each patient. of a simplified nocturnal exchange device to achieve optimal dwell times and to aug• E. Observations Needed During Training ment clearance). b. If patient's lifestyle choice is CCPD: 1. Determine 4-hour drain volumes during train• 2 BSA <1.7 m ---+ 4 X 2.5 L (9 hours/night) ing. This is to note if drain volumes are as + 2.0 Lid expected for typical 4-hour dwells with 1.5%, 2 BSA 1.7 to 2.0 m ---+ 4 X 3.0 L (9 hours/ 2.5%, or 4.5% dextrose exchanges. This is night) + 2.5 Lid not a formal peritoneal equilibration test (see 2 BSA >2.0 m ---+ 4 X 3.0 L (10 hours/ below), but is done to determine if the pa• night) + 2 X 3.0 Lid (consider combined tient's peritoneal membrane transport charac• HDIPD or transfer to HD if clinical situa• teristics are markedly different from the mean. tion suggests need). 2. Monitor for evidence of leakage in the vicinity c. If patient's lifestyle choice is NIPD: of the catheter. Many of the issues discussed above for pa- 3. Complete laboratory studies. S90 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

a. Delay baseline peritoneal equilibration test G. Adjusting Dialysis Prescription (PET) until after training (see F below). If kinetic modeling is not available, unless b. Perform serum chemistries and complete PET has changed, dialysis dose is most effec• blood count. tively increased by increasing the instilled vol• c. If a computer-assisted kinetic modeling sys• ume, therefore maximizing mass transfer and tem is available, enter preliminary data to dwell time. Another option would be to increase predict if the current prescription will be ade• the number of exchanges/day while maintaining quate. maximum dwell time, ie, by using a single night• F. Early Follow-Up time exchange to increase to 5 equal dwells per day. To this end, simplified mechanical exchange 1. Perform 24-hour dialysate and urine collec• systems have been developed to perform a noc• tion for KtNure., creatinine clearance, PNA cal• turnal exchange. culation, creatinine generation, and DlPcre.linine If kinetic modeling is available, use these pro• and DlPure• values. These should be done 2 to 4 grams to tailor a new prescription to meet ade• weeks following initiation (see Table II-I and quacy target goals and patient lifestyle issues. Guideline 3: Frequency of Delivered PD Dose This is discussed in the next section. and Total Solute Clearance Measurement Within Six Months of Initiation). COMPUTER-ASSISTED KINETIC 2. Perform peritoneal equilibration testing MODELING APPROACH TO (PET) approximately I month following initia• ACHIEVING TARGET DOSES OF tion of PD, an appropriate time physiologically. PERITONEAL DIALYSIS This baseline PET could be performed at the end As mentioned above, the availability of com• of a prolonged (> 1 week) training period (see puter-assisted kinetic modeling to tailor PD pre• Guideline 3: Frequency of Delivered PD Dose scriptions to transport type, body size, lifestyle, and Total Solute Clearance Measurement Within etc. may have distinct advantages. Much of what Six Months of Initiation). This PET (1 month) is described in the preceding discussion of the is used as the baseline measure of peritoneal empirical approach has a mathematical basis. membrane transport characteristics, not to deter• Computer-assisted kinetic modeling is a logical mine total solute clearance. This PET is done to extension of the empirical approach in that it rule out unsuspected problems or deviation from uses computer calculations to speed and assist mean transport characteristics. Low transporters the physician in PD prescription development. will probably require high-dose CAPD or CCPD. PD solution manufacturers, such as Baxter and High transporters will eventually have ultrafil• Fresenius, provide urea kinetic modeling (UKM) tration problems (when RRF diuresis fails) and models without charge. will need short-dwell therapy such as NIPD. Av• The major advantage of this approach is the erage transporters will have the most flexibility flexibility and speed of the calculations of solute (i.e., all options will be feasible). clearance. Recent studies have shown that certain 3. Perform serum chemistries and complete models very accurately predict dialysis deliv• blood count. ery.69-71 Kinetic modeling is especially important 4. If a computer-assisted modeling program is for APD therapies because the dwell times are available, enter baseline data. Actual data from so variable and may not approach the optimal 24-hour collection can be compared. for many patients. Since the models accurately 5. If clearances are at or above target, con• and reliably predict the delivered dose, the pa• tinue routine monitoring on a regular basis. Look tients and caregivers can discuss options in a for changes in 24-hour urine studies and PET timely manner. The trial and error empirical ap• data. Kinetic modeling can be used to guide fu• proach discussed above moves at a much slower ture therapy. rate. Even with computer-assisted UKM, actual 6. If clearance is below target at 1 month, a measurements are still necessary to confirm ade• change in prescription may be needed. Compli• quate dose delivery. But with computer-assisted ance issues and collection procedures should be UKM, the process is accelerated. The only theo• evaluated for abnormalities. retical disadvantage of a computer kinetic model- ADEQUATE DOSE OF PERITONEAL DIALYSIS S91

ing approach is that there might be a tendency In current clinical practice, the peritoneal dial• for the caregivers not to learn the principles be• ysis prescription is usually based on transport hind the modeling program which form the basis categori'zation using the peritoneal equilibration for the prescription strategies. test (PET) and subsequently more finely tuned The use of computer-based modeling to through empirical prescription changes guided by achieve target PD doses has been applied suc• clinical experience, as described in the preceding cessfully to a small number of children.72 section of this guideline. The dose of PD is defined as the sum of the With computerized UKM many possible PD total daily or weekly renal plus peritoneal urea regimens with variable exchange schedules and clearance normalized to the total body water, the volumes can be tailored to be compatible with KprtIV urea, which is a dimensionless parameter individual patient lifestyle preferences and to expressed as either the total daily or weekly frac• minimize total dialysate volume relative to re• tional clearance of body water for urea. 73.74 It is quired KprtIV urea' A PD prescription can be substantially more difficult to compute the appro• quickly and rigorously evaluated mathematically priate prescribed dialysis dose in PD than in RD. using programs written for the personal com• In RD, KtIV urea is delivered in a single dialysis puter.69-71 ,80 These programs all require baseline session and can be precisely calculated from the transport characterization using either the PET dialyzer mass transfer coefficient (MTC) , con• or peritoneal function test data. stant blood and dialysate flows, ultrafiltration The most common method to monitor deliv• 66 75 rate, and treatment time. . The delivered KtJ ered KprtIV urea is measurement of total daily peri• Vurea and PCR (or PNA) can both be calculated toneal and renal urea clearance by analysis of from the pre- and post-dialysis BUNs from a blood, total drained dialysate, and 24-hour urine 66 75 single dialysis. . for urea nitrogen. Although highly reliable for In PD, however, the total daily peritoneal clear• determination of KprtlVurea and PCRJPNA, batch ance, Kpt, is comprised of the sum of the clearances analyses do not provide data to distinguish be• provided by several discrete exchanges. This ther• tween noncompliance and/or possible changes in apy consists of several batch exchanges during MTC when the delivered KprtlVurea deviates from which clearance and ultrafiltration are not constant that expected with the current prescription. There (unlike the situation in RD) but fall exponentially is further discussion of this subject in Guideline to zero over the course of each exchange. The 7: PD Dose Troubleshooting and Guideline 8: peritoneal mass transfer coefficient (MTC), which Reproducibility of Measurement. controls the rate of solute transport between blood An alternative technique is measurement of and dialysate (and hence clearance), is an individ• BUN, urine volume, and urea nitrogen in aliquots ual patient characteristic,67,76 which must be deter• of each exchange 81 combined with the patient's mined and which can clearly vary as a function of exchange volume and body position.77,78 In RD, report of the exchange time and volume for each ultrafiltration contributes minimally to urea clear• exchange. A further discussion of aliquot meth• ance, while in PD it contributes up to 25% or more odology is in Guideline, 8: Reproducibility of of total clearance and must be included in calcula• Measurement. Although this technique, based on tion of the dose. The net ultrafiltration can be pre• the peritoneal function test approach, requires cisely controlled in RD, while in PD it is a complex substantially more dialysate urea nitrogen mea• function of glucose absorption, membrane water surements in addition to measurement of Kprtl permeability and lymphatic flow. The PD prescrip• Vurea and PCR (or PNA), it permits calculation of tion variables include MTC, which is dependent a MTC for each exchange. If there are important on patient, body position, and exchange volume; deviations from the reported exchange schedule, distribution of exchanges between ambulatory and the deviant exchanges will be identified by mark• supine cycler exchanges; exchange volume(s); ex• edly deviant MTCs. change times; and the osmotic gradient or percent There are advantages and disadvantages of dextrose in each exchange. Additionally, residual both the PET and PFT measurements. The wide renal urea clearance must be measured and in• range of exchange times, including nearly com• cluded in the prescription and body water or V plete equilibration in long exchanges and the as• estimated from age, gender, and surface area.37,79 sumption of constant BUN, will result in some S92 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

variability in the MTCs measured with the PH is consensus that such patients need extra dial• which do not reflect true differences. On the other ysis, the requisite increase is unclear and should hand, the MTC calculated with a single 2 L ex• be studied. Other malnutrition-related questions change under carefully controlled conditions will of interest include: Can aggressive dialysis deliv• not always accurately reflect the MTC under clin• ery reverse malnutrition? What V is to be used ical exchange conditions with variable exchange in malnourished patients? volumes and body position. Can increasing dialysis dose improve out• A simple kinetic technique for routine moni• comes in a linear manner, or is there a dose above toring of delivered KprtIV urea and PCRlPNA in which no benefit is noted, or complications or established patients for whom the MTCs for urea costs outweigh the benefits? and creatinine and 24-hour dialysate creatinine A multi-centered study of pediatric patients to have been previously established, is measure• evaluate clinical outcome as a function of deliv• ment of BUN and serum creatinine and dialysate ered PD dose should be initiated. Urea kinetic urea and creatinine from an aliquot of one ex• modeling computer programs specifically de• change. This data combined with the number of signed for children should be developed and vali• exchanges, exchange times, and exchange vol• dated in a prospective trial. umes reported by the patient permits calculation Although there is a database documenting of KprtlVure .. PCRlPNA and expected total dialy• the validity of UKM to describe transport in sate creatinine content.67 The validity of the data PD,67,69-71,82,83 UKM has not been widely used can be assessed by comparison of the calculated dialysate creatinine content to the measured his• to prescribe and control the delivered dose, torical value for the patient (see Guideline 7: PD Since the risk of mortality may be highly non• Dose Troubleshooting). In this way, both Kprtl linear with increased dose of delivered dial• V urea and PCRlPNA can be estimated from mea• ysis,84,85 it is reasonable to assume that the co• surements of urea nitrogen and creatinine in a efficient of variation on mean KprtlV urea should blood sample and a small aliquot of one exchange not exceed 10% to 15% for individual patients, using computerized UKM. When there is devia• A multi-center clinical trial to study clinical tion of more than 10% in the expected creatinine outcome as a function of Kprt/V urea, using UKM excretion, more complete dialysate collections to control Kprt/Vurea prospectively in individual would be indicated for analysis of therapy. randomized patients is recommended. Do patients with diabetes need higher targets RECOMMENDATIONS FOR RESEARCH for delivered dose of PD? Should PD delivered The amount of dialysis required for malnour• dose be increased in hospitalized patients during ished patients is not known. While there probably acute illness or stress? VI. Strategies for Increasing the Likelihood of Achieving the Prescribed Dose of Peritoneal Dialysis

GUIDELINE 19 the dialysate plus urine, as detailed in Guideline 7: PD Dose Troubleshooting, is recommended Identify and Correct Patient-Related Failure by the Work Group as a method for measuring to Achieve Prescribed PD Dose (Opinion) compliance.

Potential patient-related causes of failure to GUIDELINE 20 achieve prescribed peritoneal dialysis dose Identify and Correct Staff-Related Failure should be investigated and corrected. These in• to Achieve Prescribed PD Dose (Opinion) clude: • Failure to comply with the prescription Potential staff-related causes of failure to • Lack of understanding of the importance of achieve prescribed peritoneal dialysis dose adherence to the full prescription should be investigated and corrected. These in• • Sampling and collection errors. clude: Rationale A detailed rationale is presented in • Errors in prescription Appendix H. The following is a summary. • Inadequate monitoring of delivered dose Preliminary data from the USRDS DMMS • Inadequate patient education. Wave II project show that 487 CAPD patients Rationale To increase the likelihood of self-report full compliance with 82.8% of their achieving a prescribed dose of PD, it is necessary exchanges.3 One exchange per week is missed to elucidate the staff-related causes of failure to by 11.5% of patients and 2 to 3 exchanges per achieve a prescribed dose of peritoneal dialysis. week are missed by 4.5% of patients, all self• The Work Group found no reports addressing reported. The American Association of Kidney this issue in PD; the following discussion repre• Patients completed a patient self-reported survey sents the opinion of the Work Group members. about the impact of the NKF-DOQI guidelines Inadequate understanding of the physiology and came up with a very similar number for fre• and kinetic principles of PD by the physicians quency of missed exchanges. and nursing staff may result in: Conditions causing noncompliance in PD pa• • Errors in patient selection tients have not been adequately analyzed. From • Errors in the prescription of the PD dose studies on compliance with chronic drug regi• • Errors in monitoring whether the prescribed mens, it is known that patients are more compli• dose is delivered ant when they are convinced about the appropri• • Errors in PD dose modification to achieve ateness and beneficial effects of the prescribed the prescribed goal treatment and that frequent reinforcement of the • Inability to test for and recognize patient importance of the treatment is associated with noncompliance better compliance. Therefore, it is the opinion of • Inadequate patient education. the Work Group that, in addition to giving careful The impact of patient education on patient consideration to the selection of medically appro• compliance with the PD prescription was dis• priate candidates for PD, as detailed in Section cussed in Guideline 19: Identify and Correct Pa• VIII: Suitable Patients for Peritoneal Dialysis, tient-Related Failure to Achieve Prescribed PD special emphasis should be placed on education Dose. The chance of inappropriate prescription of PD patients about the importance and tech• of the PD dose is enhanced when the prescribing nique of the PD prescription. Instructions should physician has a sketchy knowledge of the princi• be repeated at least every 6 months, and patients ples of clearance studies in PD. It has recently should be monitored for signs of change in com• been recognized that nephrology fellowship cur• 86 pliance. Monitoring the output of creatinine in ricula lack emphasis on training in dialysis. To prescribe and deliver the proper dose of PD, nephrologists must ensure adequate education © 1997 by the National Kidney Foundation. Inc. and training in PD. 0272-6386/97/3003-0216$3.00/0 The use of computer modeling in PD may help

American Journalof Kidney Diseases, Vol 3D, No 3, Suppl 2 (September), 1997: pp S93-S94 S93 S94 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY achieve the prescribed dose by suggesting vari• principles of clearance and lack of proper teach• ous options to alter the PD dose. This approach ing technique. Staff responsible for patient edu• may assist in avoiding unrealistic PD dose sched• cation should be trained and competent in both ules for certain patients (see Guideline 18: Use the principles of clearance and the technique of of Empiric and Computer Modeling of PD Dose). patient instruction. The use of total creatinine appearance/output data in detecting noncompliance is important as RECOMMENDATIONS FOR RESEARCH discussed in Section II: Measures of Peritoneal Validate methods of assessing compliance. Dialysis Dose. Evaluate the association between patient un• Inadequate education may be a key factor in derstanding of PD techniques and compliance. the patient non-adherence to the prescribed dose Specifically, what is the role of inadequate pa• of therapy resulting in the above-mentioned tient knowledge in noncompliance? shortcuts. The American Association of Kidney Evaluate effect of staff s knowledge of clear• Patients suggests that PD patients are willing to ance principles and teaching techniques, and rep• increase the frequency and/or volumes of ex• etition frequency of patient instruction in proper changes, if necessary, and that explanations (edu• delivery of PD dose. cation) and participation in decision making are Is there a psychological profile which is pre• good incentives. Inadequate education may stem dictive of noncompliance? If so, what is the best from both poor educator understanding of the method to characterize this profile? VII. Clinical Outcome Goals for Adequate Peritoneal Dialysis

BACKGROUND GUIDELINE 22 Throughout these Guidelines, the Work Group Measurement of PD Technique Survival has focused on patient outcomes. Improving pa• (Opinion) tient outcomes is the primary objective of the PD technique survival, both dependent and in• DOQI (Dialysis Outcomes Quality Initiative). dependent of peritonitis, should be quantitated The Work Group realizes that definitions of goals serially in PD patients as an outcome measure. regarding patient outcomes are needed. As stated in the Introduction to these guidelines, the goals Rationale It is common for ESRD patients are integral to the definitions of adequate, opti• to change renal replacement therapy modalities mal, and effective dialysis. during the course of their treatment. Reasons for transfer include: complications of the therapy, GUIDELINE 21 inability to perform the therapy (lack of suitable access, no partner to do self care, medical contra• Measurement of PD Patient Survival indication), and patient requestllifestyle issues. (Opinion) Patient case mix, geographical location, and ex• Survival of PD patients should be quantitated perience with PD in complicated cases are factors serially as an outcome measure. _ affecting transfer. For some patients, for optimal Rationale Patient survival is an objective out• outcome, it may be medically appropriate to come that is dependent upon many variables, transfer from PD to HD; this does not imply fail• some controllable and some uncontrollable. Sub• ure of the therapy or the dialysis facility. optimal doses of delivered dialysis will adversely Peritonitis remains the primary cause of trans• affect patient survival in adults (data unavailable fer from PD. It is acknowledged that at times in children), as discussed in detail in Guideline peritonitis is the' 'precipitating" event for trans• 15: Weekly Dose of CAPD. A primary goal of fer, while the real underlying reason is patient ESRD therapy is to prolong life while minimiz• burnout, non-compliance, inadequate dialysis, a ing uremic symptoms. United States Renal Data request based on lifestyle, or an underlying exit Systems (USRDS) data from hemodialysis pa• site infection. Overall peritonitis rates can be in• tients have demonstrated an association between fluenced by the center. An association between malnutrition and frequency of peritonitis has low KtIV and increased incidence of death from 90 coronary artery disease, other cardiac disease, been reported. Inadequate dialysis may lead to cerebrovascular accidents, and other condi• inadequate dietary protein intake and malnutri• tions.87088 There is also evidence that underdial• tion as described in detail in Section IV: Assess• ysis adversely affects mortality in PD patients ment of Nutritional Status as it Relates to Perito• with ischemic cardiac disease or left ventricular neal Dialysis. The relationship between solute dysfunction.16089 Thus patient survival is a mea• clearance and frequency or/severity of peritonitis sure of renal replacement program effectiveness. has not been adequatelY studied. For example, it Case mix must be factored into survival analysis, is not clear if inadequate dialysis or malnutrition however. The USRDS case mix analysis is an directly predispose the patient to peritonitis. excellent starting point, addressing age, race, pri• However, peritonitis is an important outcome and mary renal disease, and presence or absence of its frequency and severity is an index of the over• diabetes. The USRDS is attempting to refine case all suitability of PD. It is, therefore, the opinion mix further by addressing other underlying co• of the Work Group that peritonitis should be morbidities. At dialysis centers with a small num• monitored. ber of patients, survival may need to be evaluated Inadequate dialysis is directly responsible over many years to obtain a reliable estimate. for at least 10% of transfer to HD.87 There is an association between PD technique failure and total solute clearance,16.91 and one possible © 1997 by the National Kidney Foundation, Inc. reason for poor technique survival rates may 0272-6386/97/3003-0217$3.00/0 be underlying inadequate dialysis. While the

AmericanJournal of Kidney Diseases, Vol 30, No 3, Suppl2 (September), 1997: pp S95-S100 595 S96 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

CANUSA study found a relationship between GUIDELINE 23 creatinine clearance and PD technique sur• Measurement of Hospitalizations (Opinion) vival, the investigators suspected this was more related to RRF than delivered dose of PD. In ESRD-related and ESRD-unrelated hospital• the CANUSA study, technique survival was izations (admissions/year, hospitalized days/ approximately 75% at 2 years in North year) in PD patients should be quantitated as an America. 16 However, peritonitis was not ana• outcome measure. lyzed as a variable in the multivariate analysis. Rationale Hospitalization is an indicator of No difference was found in technique survival the overall effectiveness of treatment of chronic between patients from Canada and the United conditions and therefore constitutes an important States. Average KtiV urea started at 2.38 and de• outcome for dialysis patients. The number of ad• creased to l.99 over 2 years. The risk of tech• missions per year and total number of hospital nique failure increased by 5% for every 5-L/ days per year are two separate, but related, serial wk decrease in creatinine clearance. These measures of outcome. Hospitalizations of PD pa• findings corroborated data from Tattersall et al tients can be related or unrelated to ESRD. An who found that patients with a lower KtiV urea association between low creatinine clearance and had a lower rate of technique surviva1.91 The increased overall hospitalization rate has been lower rate of technique survival was related to reported. 16 Based on this evidence, the Work underdialysis, not to peritonitis or hernia devel• Group recommends that cause, frequency, and opment. The nutritional parameter of nPCR has length of hospitalizations ofPD patients be moni• been shown to be predictive of CAPD tech• tored. Categorizing hospitalizations according to nique failure in a multivariable analysis.92 whether they are related to ESRD or not offers It is a common perception that patients trans• certain advantages for analysis and, therefore, it ferring from any renal replacement therapy are is the opinion of the Work Group that this type at increased risk for death in the immediate post• of stratified analysis should be performed. transfer period. However, indirect evidence from According to the USRDS, PD patients are hos• the USRDS does not support that perception.93 pitalized an average of 1.8 times per year. 94 The Early demise following transfer may be due to a CANUSA study found, by multi-factorial analy• variety of reasons, including the underlying sis, an association between prolonged hospital• cause of transfer, inadequate dialysis on PD, and ization and low creatinine clearance. 16 Some ad• malnutrition. Therefore, the Work Group recom• missions are specific to PD, ie, not seen with mends that technique survival be measured at other dialysis therapy, such as elective abdominal each PD facility. Admittedly, there may be cen• wall herniography. As larger instilled volumes ter-specific differences for rates of transfer and are administered to maintain target doses of dial• that at centers where there are small numbers of ysis, there is an increased risk of leaks and hernia PD patients, these rates may need to be evaluated formation, both of which can lead to hospitaliza• over many years. It is recommended that these tion. Admissions unrelated to ESRD are im• rates be evaluated both dependent and indepen• portant indicators of morbidity (cardiac disease, dent of peritonitis, although, in one study (CA• infections, etc.) in PD patients. Since hospitaliza• NUSA), there was no significant difference be• tion data are important outcome parameters for tween the two evaluations. 16 all dialysis patients and can reflect solute clear• PD technique survival is dependent upon ance, the Work Group recommends that they many factors including infections, patient mo• should be monitored. tivation, ultrafiltration (transport characteris• Although it is uncertain whether inadequate di• tics), and total solute clearance. Thus, PD tech• alysis is directly related to an increased risk of nique survival is not a simple outcome measure peritonitis and catheter infections, inadequate dial• for the adequacy of PD. Nonetheless, centers ysis is related to uremic symptoms such as nausea, should strive to achieve the goal of a 75% 2- vomiting, and gastrointestinal bleeding. It is ac• year technique survival rate (the rate noted in knowledged that some centers may treat all epi• CANUSA). Case mix must be factored into sodes of peritonitis in the hospital, while others survival statistics. only admit those with severe or refractory perito- CLINICAL OUTCOME GOALS FOR ADEQUATE PERITONEAL DIALYSIS S97

nitis. Therefore, in addition to tracking hospital• • General Health Questionnaire ization rates, centers should also monitor reasons • Simmons Self Esteem Scale for hospitalization (related vs. unrelated to ESRD, • Profile of Mood States and specific reason for admission). The use of • Multidimensional Health Locus of Control ICD-9 (International Classification of Diseases, 9th • Modality Specific Stresses Scale revision) or similar codes may be valuable in this • General Treatment Stress Scale process. The USRDS is attempting to categorize • Global Illness Stress on Self and Others, causes of hospitalizations as infectious, cardiovas• Global Adjustment to Illness Scale cular, dialysis access-related, and all other. • Quality of Life (QL 100 mm) Analogue Hospitalizations from causes unrelated to ESRD Scale may be related to inadequacy of PD. As discussed • Dialysis Relationship Quality Scale in Guideline 25: Measurement of PD Patient Sur• • Social Leisure Activities Index, Social Sup• vival, disease-specific (particularly cardiac) mor• port Satisfaction Scale tality is related to the dialysis dose for both HD • General Well Being Index and PD. Therefore, although studies of disease• • Index of General Affect, Overall Life Satis- specific hospitalizations and their relation to the faction dose of PD have not been reported, it is reasonable • Katz Activities of Daily Living to monitor this relationship. Each outcome mea• • Time Tradeoff Measures. sure should be adjusted as well as possible for Unfortunately, many of these instruments do not case mix. The USRDS is attempting to do so witlf have published data indicating that reliability (test• Standardized Hospitalization Rates (SHRs). retest, inter-rater), validity (content, construct, in• ternal consistency), and responsiveness-to-change 95 GUIDELINE 24 have been rigorously tested. For this reason, no particular instrument can be strongly recommended Measurement of Patient-Based Assessment over another. Furthermore, many instruments de• of Quality of Life (Opinion) veloped for research purposes may be burdensome Patient-based assessment of quality of life for patients or facilities, eg, require interviewer as• (QOL) in PD patients should be evaluated seri• sistance or have complicated scoring algorithms. Nevertheless, generic and disease-specific mea• ally as an outcome measure. 96 A patient-based quality of life instrument sures hold promise as useful clinical tools. should have both generic and disease/treatment• A popular generic measure used in peritoneal specific measures of health-related quality of life, dialysis patients is the Medical Outcomes Study and should be shown to be valid, reliable, and Short Form-36 (SF-36). Promising self-adminis• tered instruments used in peritoneal dialysis pa• responsive prior to use. Once such an instrument tients include the CHOICE Health Experience is available, it should be administered at initiation Questionnaire97,98 and the Kidney Disease Qual• of dialysis and at intervals determined to be ap• ity of Life (KDQOL).99 propriate by its validation studies. The Work Group recomillends that each facil• Rationale Quality of life (QOL) can be as• ity keep abreast of future developments regarding sessed with generic or disease-specific measures. these instruments. As experience increases and Many quality of life measures have been used in one or more instruments are clearly established dialysis patients. However, fewer measures have as useful in PD patients, standardized QOL mea• been used for peritoneal dialysis than for hemodi• surement should be integrated into the routine alysis patients.95 Measures used in peritoneal di• care and evaluation of patients, programs, and alysis patients and reported in the literature in• facilities. clude95 : • Medical Outcomes Study Short Form 36 GUIDELINE 25 (SF-36) Measurement of School Attendance, • Sickness Impact Profile (SIP) Growth, and Developmental Progress in • Index of Well Being, Index of Overall Life Pediatric PD Patients (Opinion) Satisfaction School attendance (in the absence of other • Index of Psychological Affect comorbidities precluding school attendance), S98 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

growth, and developmental progress should be in their model. 101 In this last study, the presence measured serially in pediatric PD patients. of comorbid conditions was associated with low Rationale The ability of pediatric PD patients serum albumin. lol Several cross-sectional studies to attend school is an important measure of the have identified a positive correlation between se• success of PD. rum albumin concentration and solute clear• Underdialysis may affect the cognitive devel• ance.49,102.103 However, urea and creatinine clear• opment and statural growth of children. How• ance were not identified as predictors of serum ever, the exact relationship between dialysis dose albumin by multivariate analysis.54,104 Age, com• and normal growth and development in children orbid factors (diabetes) and peritoneal solute is not clear. Nonetheless, the Work Group be• transport were the major predictors of serum al• lieves that cognitive development and statural bumin in these multi-factorial analyses. growth should be monitored serially in children Normal serum albumin concentrations vary and charted in relation to patient age. by laboratory methodology; hence local stan• dards should be used. If the serum albumin GUIDELINE 26 level is below normal for the laboratory, but is increasing, this suggests that the patient is Measurement of Albumin Concentration in anabolic and is increasing protein stores. Con• PD Patients (Opinion) versely, a low albumin or decreasing albumin A stable or rising serum albumin concentration level is likely to be associated with malnutri• that is greater than or equal to the lower limit of tion or decreasing protein stores. Although normal for each laboratory should be used as an there are no published data specifically ad• outcome goal. dressing this point, it is the Work Group's Rationale In PD patients, as with HD pa• opinion that a patient whose serum albumin tients, there is strong evidence to suggest that a has decreased 0.1 g/dLimonth from a baseline low serum albumin level is associated with an of 4.0 g/dL to 3.7 g/dL may be at higher risk increased risk of technique failure and than a patient with a stable serum albumin con• death. 16.48,100,101 Patients with the highest serum centration of 3.7 g/dL. albumin levels have the lowest risk of death. In Taken together, the data discussed above sug• the CANUSA study, a difference of 0.1 g/dL gest to the Work Group that: serum albumin concentration was associated with • Serum albumin concentration should be a 5% change in the risk of technique failure, a monitored on a regular basis and a stable or rising 5% change in days hospitalized, and a 6% change value is desirable. It should be measured at least in the risk of death. 16 Therefore, the Work Group every 4 months. recommends monitoring serum albumin concen• • Serum albumin levels should be evaluated tration in PD patients because of its association in the context of the patient's overall clinical with important outcomes. status including comorbid diseases, peritoneal Although the significance of serum albumin as transport type, delivered dose of PD, and quality• a predictor of outcomes in adults is undisputed, of-life issues. its relationship to overall nutrition and, to a larger • The highest albumin level possible should extent, to the levels of urea or creatinine clear• be the goal for each patient. ance is unclear. That albumin synthesis depends It is the Work Group's opinion that an optimal on dietary protein intake is well known. How• serum albumin level can be obtained by adequate ever, catabolic illness can reduce albumin syn• nutrition monitored frequently by the renal dieti• thesis, and probably increase albumin degrada• tian, prevention and treatment of catabolic ill• tion, even when dietary protein intake is not low. ness, and maintenance of KtNurea and creatinine Observations in PD patients have provided indi• clearance at or above the levels recommended in rect support for this effect of catabolic illness. Section V: Adequate Dose of Peritoneal Dialysis. Although serum albumin concentration is an im• In summary, low serum albumin is a strong portant predictor of outcome,48.100 it was not predictor of mortality and morbidity in PD pa• found to be significant in another study when tients. Therefore, serum albumin is an important comorbid conditions were entered as covariates outcome measure in PD patients and should be CLINICAL OUTCOME GOALS FOR ADEQUATE PERITONEAL DIALYSIS S99

monitored, although an association between se• are strongly correlated in cross-sectional stud• rum albumin and urea or creatinine clearance has ies.55,1~8 However, it has been suggested that this not been convincingly shown. Efforts to maintain is due in part to mathematical coupling of 53 serum albumin in the normal range should in• data. ,109 Three studies have investigated the ef• clude adequate nutrition, adequate clearances, fect of an increased KtIV urea on nutritional status and prevention and treatment of catabolic illness. (nPCR and serum albumin concentration) in a limited number of subjects.54,56,IlO While nPCR GUIDELINE 27 increased as KtIV urea increased, an increase in serum albumin concentration did not occur. A Measurement of Hematocrit in PD Patients reasonable conclusion from these data would be (Evidence) that in the absence of significant comorbidity, an Providers should strive to achieve a hematocrit increase in delivered dialysis dose should result level of 33% to 36% in 75% of PD patients. in a corresponding increase in nPNA. In patients Rationale See NKF-DOQI's Clinical Practice who show signs of malnutrition, their dialysis Guidelines for the Treatment of Anemia of prescription should be closely evaluated with Chronic Renal Failure. consideration to increasing their dose of dialysis if significantly below target. This is discussed in GUIDELINE 28 Guideline 15: Weekly Dose of CAPD. No study of PD patients has demonstrated that Measurement of Normalized PNA in PD Patients (Opinion) nPNA is an independent predictor of outcome when a multiple regression model is used. Corre• Providers should strive to achieve a normal• lation coefficients relating DPI to nPCR are on ized PNA (nPNA) of greater than or equal to 0.9 the order of 0.6. However, PD patients who are g/kg/day in PD patients. neither anabolic nor catabolic tend to demon• Rationale The role of PNA is discussed in strate higher correlation coefficients between Guideline 12: Assessment of Nutritional Status. nPNA and DPI.51 Also, there are little data to Maintenance of positive nitrogen balance and show a significant relationship between nPCR the prevention of underlying malnutrition is im• and serum albumin levels.54,104 Age, peritoneal portant because of the documented detrimental transport type, presence of diabetes, and other impact of hypoalbuminemia and low SGA scores comorbid diseases have a greater effect on albu• I6 on patient survival. ,74,loo,l04 From nitrogen bal- min than does nPCR. This subject is discussed ance studies, it has been estimated that PD pa• in further detail in Appendix B. tients need to ingest at least 1,2 g/kg/d of protein Despite these concerns, the Work Group rec• to maintain positive nitrogen balance. 105 This is ommends that PNA should be monitored. Low higher than the recommended daily protein in• nPNA values in nonanabolic PD patients indicate take for healthy individuals, but not surprising a low DPI regardless of the values of other nutri• due to the significant amount of protein known tional indices.52 One should strive to achieve an to be lost in dialysate. Despite these recommen• nPNA of at least 0.9 g/kg/d in adult PD patients, dations from balance studies, it has been reported PNA values at this level or higher are likely to that many patients are in positive nitrogen bal• be associated with neutral or positive nitrogen 49 51 ance with protein intakes of 0.9 to 1.0 g/kg/d. . balance in the absence of significant comorbidity Cross-sectional studies would suggest that in the or dialysate protein losses. absence of significant comorbid diseases, pa• While the recommended dietary allowance for tients with PO doses in the range of KtIV urea of normal children is known, there are no definitive 2.0 spontaneously ingest at least 0.9 g/kg/d of data regarding the real protein needs of children, protein.50,I06,107 Total solute clearance and nPCR* especially young children on dialysis. However, clinical practice suggests that the protein needs of children on PD are greater than the recommended * Studies that used the term peR are cited in this rationale. dietary allowance, in part related to dialysate pro• Since the original study authors used the term peR, this tein losses. Current recommended protein intakes rationale will use the term peR when specifically describing for children receiving PD, referenced for age are results from studies which used that term. However, the Work as follows III: Group favors the term PNA. S100 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

Age Protein Intake not been adequately studied. Specifically, does Birth to 3 years 2.5-3.0 g/kg/d inadequate PD dose contribute to peritonitis? 3 years to puberty 2-2.5 g/kg/d Studies to assess whether an increase in nPNA is associated with an increase in serum albumin Pubertal patients 2.0 g/kg/d levels, nutritional status, or improved survival Post-pubertal 1.5 g/kg/d would be valuable. Because dialysate protein losses may vary Studies to further define the relationship be• widely in children, individualized recommenda• tween nPCR and outcome are needed. For ex• tions for dietary protein intake may benefit from ample, the longitudinal follow-up of nutritional measurement of dialysate protein losses. An status (determined by a variety of methods) equivalent amount of protein to replace dialytic will be more influential in improving under• losses must be added to the recommended normal 112 standing of nutritional outcomes. Longitudinal daily allowance for normal children. studies should be emphasized over cross-sec• tional studies. RECOMMENDATIONS FOR RESEARCH The relationship between PD dose and out• come parameters in children needs definition. The relationship between solute clearance and Studies of nutritional interventions are lacking the frequency and/or severity of peritonitis has and are encouraged. VIII. Suitable Patients for Peritoneal Dialysis

GUIDELINE 29 patient~ benefit from increased vascular stability, which can be achieved with PD. Indications for PD (Opinion) Over a period of time, vascular accesses fail and revisions are no longer able to restore ade• quate blood flow. As a result, the patient receives Indications for PD include: inadequate hemodialysis and should be evaluated • Patients who prefer PD or will not do hemo• for PD. 1l3,116,126 dialysis (HD) Home hemodialysis requires an assistant. For • Patients who cannot tolerate HD (e.g., some patients with congestive or ischemic heart dis• patients who prefer dialysis at home, the lack ease, extensive vascular disease, or in whom vas• of a hemodialysis assistant may mandate PD. In addition, patients who have transportation prob• cular access is problematic, including the major• lems to a hemodialysis center or live a great ity of young children) distance from a center may prefer home PD.127 • Patients who prefer home dialysis but have no assistant for HD, or whose assistant cannot The decision to initiate PD rather than HD in be trained for home HD children is influenced by a variety of factors. Because of the difficulties in maintaining vascu• Rationale There is a rapid change in solute lar access in infants and small children, PD is transport as well as rapid shifting of volume within compartments during HD. Some patients usually the modality of choice when weight is <20 kg. Regular school attendance by children with severe cardiac disease may be better min• of all ages can best be achieved with a home aged on PD since these acute changes are dialysis procedure. PD is typically preferred over avOl'd e d . 113-117 PD h as b een propose d as a method HD. Finally, renal replacement therapy can also of managing refractory heart failure even in pa• best be provided by PD when the child lives a tients without renal failure. 118 long distance from a pediatric ESRD center. Advantages of PD in patients with cardiovas• cular disease include: better hemodynamic con• trol, less acute hypokalemia (or electrolyte shifts) GUIDELINE 30 which could result in arrhythmia, and better con• Absolute Contraindications for PD trol of anemia (important in patients with coro• (Opinion) nary artery disease). Although a comparison of PD to HD for patients with severe heart failure Absolute contraindications for PD include: has not been published, there are several reports • Documented loss of peritoneal function or of successful PD performance in subjects with extensive abdominal adhesions that limit dialy• severe heart failure.119-124 Tolerance of the proce• sate flow dure (PD), fluid management, prevention of ar• • In the absence of a suitable assistant, a pa• rhythmias, and patient survival were satisfactory tient who is physically or mentally incapable of in these reports. performing PD Extensive peripheral or central venous occlu• • Uncorrectable mechanical defects that pre• sive disease prohibits surgical placement of some vent effective PD or increase the risk of infection types of hemodialysis access. Manifestations of (eg, surgically irreparable hernia, omphalocele, severe ischemia, even gangrene, of the hands fol• gastroschisis, diaphragmatic hernia, and bladder low placement of vascular access in the same extrophy) wrist or forearm in a few patients with severe Rationale Documented Loss of Peritoneal peripheral vascular disease, particularly diabet• Function. PD efficiency relies on effective ics. 125 Marginal vascular beds are at risk for isch• peritoneal blood flow, dialysate flow, sufficient emia or reduced perfusion during hypotension, peritoneal surface area, and permeability to allow which is frequent in some HD patients. These adequate solute and fluid removal. Any compro• mise in these functions may result in inadequate peritoneal dialysis and thus the failure of pD.114.128 © 1997 by the National Kidney Foundation, Inc. It should not be assumed that children who 0272-638619713003-0218$3.0010 have previously undergone extensive abdominal

AmericanJournal of Kidney Diseases, Vol 30, No 3, Suppl 2 (September), 1997: pp S1 01-S1 04 S101 S102 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

surgery will not achieve successful PD. A trial subcutaneous tissues, pleural space, or genitalia of PD is warranted in such children and adequate can be painful and cause local problems. Leaking dose delivery must be documented. into the vagina or rectum increases the risk of Psycho-neurological Problems. The optimal contamination. Unsatisfactory drainage and performance of PD requires certain physical and clearance, as well as medical complications, such intellectual capabilities of the patient or care• as respiratory compromise in the case of dia• giver. With major loss of mechanical function or phragmatic leak, can occur as a result of such eye-hand coordination, PD becomes difficult to leakage. 114.117.127 perform. Patients or caregivers are responsible Body Size Limitations. Body size can be a rela• for problem identification and problem solving tive contraindication to PD when the patient is ei• during PD. If the patient is deemed psychologi• ther too small to tolerate the prescribed dialysate cally incompetent, these tasks and decisions may volume or too large to achieve adequate dialysis. not be reliably or safely executed.1l6.127 For patients with little or negligible RRF, there are Abdominal Mechanical Problems. The dial• definite size limitations for adults on CAPD with ysate in the abdomen must be accessible to the 4 daily exchanges. 132 However, even larger individ• vascular bed of the peritoneal membrane. Any uals can achieve acceptable clearances if they are mechanical problem that prevents this (eg, hernia treated with a combination of daily CAPD and sack, subcutaneous leak) will impair the effi• nocturnal automated PD. 133 In large individuals, in• ciency of PD. Intra-abdominal pressure increases crease in the exchange volume is more efficient with dialysate infusion and during the ultrafiltra• than increase in the number of daily exchanges. tion process, thereby exacerbating any structural However, the patient with the increased exchange defect such as hernia. Some of these abdominal volume may experience abdominal pain or discom• defects are not surgically correctable.1l6.117 fort, shortness of breath, or loss of appetite as a result of abdominal pressure. 134 GUIDELINE 31 Intolerance to PD Volumes Necessary to Achieve Adequate PD Dose. Intolerance to a Relative Contraindications for PD (Opinion) PD volume is generally not known until it is Relative contraindications for PD include: attempted. Frequent exchanges with small vol• • Fresh intra-abdominal foreign bodies (e.g., umes, as observed during automated PD, may 4-month wait after abdominal vascular prosthe• not be able to provide an adequate delivered dose ses, recent ventricular-peritoneal shunt) of PD. Raising volumes to the limit of tolerance • Peritoneal leaks may be problematic in patients with advanced • Body size limitations lung disease, or patients with recurrent hydrotho• • Intolerance to PD volumes necessary to rax. Infrequently, this may be applicable to some achieve adequate PD dose patients with polycystic kidney disease or severe • Inflammatory or ischemic bowel disease lumbo-sacral disk disease. • Abdominal wall or skin infection Inflammatory or Ischemic Bowel Disease. In• • Morbid obesity (in short individuals) flammatory or ischemic bowel disease or • Severe malnutrition frequent episodes of diverticulitis are relative • Frequent episodes of diverticulitis. contraindications to peritoneal dialysis. It is rea• Rationale Fresh Intra-abdominal Foreign sonable to assume that there may be increased Bodies. Newly implanted abdominal prosthe• risk for transmural contamination by enteric or• ses must be allowed sufficient time for healing ganisms in these circumstances. 127 to avoid leakage or possible dialysis-related peri• Abdominal Wall or Skin Infection. Abdomi• tonitis with potential spread to the prosthetic de• nal wall or skin infection can lead to contam• vice or material. The time required for healing ination of the catheter exit site, tunnel, and may vary from 6 to 16 weeks.129-131 The bacterial peritoneal cavity through touch and cross con• seeding of any vascular prosthesis during hemo• tamination. 116 The decision to use PD in patients dialysis is also a risk. The best type of dialysis with a colostomy or ileostomy must be individu• in this setting is unclear. alized, since successful application of PD has Peritoneal Leaks. Peritoneal leakage into been described in such patients. SUITABLE PATIENTS FOR PERITONEAL DIALYSIS S103

Morbid Obesity. Morbid obesity can pose tions for switching a patient from PD to HD are special dilemmas in peritoneal catheter place• based on the following considerations:

ment, the healing process, and in providing ade• Consistent Failure to Achieve Target KtlVurea quate dialysis. The possibility that increased ca• and Ccr. Consistent failure to achieve the target loric absorption from the dialysate could lead to total solute removal with proper PD prescription further weight gain should also be considered. management should lead to evaluation of compli• Severe Malnutrition. Wound healing is com• ance issues and deterrents to appropriate perfor• promised in severely malnourished patients. Self• mance of peritoneal dialysis exchanges. After all dialysis such as PD may not be suitable for many avenues have been explored, if social or physical severely malnourished patients because of inabil• issues cannot be overcome, transfer to HD may ity to comply with the dialysis regime. Further• be necessary as long as the same issues do not more, peritoneal protein losses may not be toler• deter appropriate therapy (e.g., adequate ultrafil• ated. tration, single pool delivered KtIV urea of 1.2 thrice Frequent Episodes of Diverticulitis. Di• weekly, etc.) on this modality.l28·m verticulitis during peritoneal dialysis often re• Inadequate Solute Transport or Fluid Re• sults in peritonitis. Peritoneal dialysis in patients moval. Peritoneal solute transport determined with frequent episodes of diverticulitis places by PET affects both solute and fluid removal these patients at higher risk for peritonitis. by PD. Obviously, peritoneal transport type is discovered after initiation of PD by the first PET. GUIDELINE 32 High transporters may have poor ultrafiltration and/or excessive protein losses (relative contrain• Indications for Switching from PD to HD dication). Excessive protein losses are those that (Opinion) exceed the patient's ability to compensate by an The decision to transfer a PD patient to HD increase in dietary protein consumption. How• should be based on clinical assessment, the pa• ever, peritoneal urea and creatinine clearances tient's ability to reach HD dose target levels, and tend to be adequate in high transporters. Many the patient's wishes. In particular, these patients high transporters with poor ultrafiltration can be should have vascular access addressed as advised effectively dialyzed with short dwell periods and by the NKF-DOQI Vascular Access Work daytime exchanges, but such a regimen may be• Group. come too burdensome for the patient's life• 114 Indications for switching from PD to HD in• style. ,128,135 Low transporters usually have ade• clude: quate ultrafiltration, but, when they are relatively • Consistent failure to achieve target KtIV urea large, may have inadequate peritoneal clearance and CCr when there are no medical, technical, or of creatinine, but not necessarily a decreased psycho-social contraindications to HD clearance of urea.61 • Inadequate solute transport or fluid removal. Excessive protein losses can occur if the pa• High transporters may have poor ultrafiltration tient's underlying disease iqcludes active nephro• and/or excessive protein losses (relative contrain• sis, if the patient is a high transporter, or if dication, obviously discovered after initiation frequent peritonitis occurs. The resulting malnu• and the first PET) trition will increase the patient's mortality and • Unmanageably severe hypertriglyceridemia morbidity. In some children who are actively ne• • Unacceptably frequent peritonitis or other phrotic, protein losses may be successfully re• PD-related complications placed by supplemental (e.g., nasogastric, gas• • Development of technical/mechanical prob• trostomy) tube feedings. lems There are medical complications that may de• • Severe malnutrition resistant to aggressive velop or have been present prior to initiation of management (relative) dialysis, but these may become apparent only Patients should be informed of the risks of after peritoneal equilibration testing and ade• staying on PD at a level of adequacy below that quacy studies. recommended by their physician. Inadequate solute transport documented by Rationale The above recommended indica- measures of KtIV urea and creatinine clearance S104 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

must be evaluated. If the maximum PD prescrip• Unmanageably Severe Hypertriglyceridemia. tion has been reached (increases in volumes and Unmanageably severe hypertriglyceridemia, re• frequency of exchanges including use of noctur• sulting from, or exacerbated by, the dextrose load nal cycling) or the procedure is no longer achiev• intrinsic to the dialysate, may increase the risk able due to lifestyle complications, hemodialysis for cardiovascular disease. as an alternative should be explored. 114 Development of Technical/Mechanical Prob• These guidelines have defined adequate sol• lems. Irreparable technical or mechanical de• ute transport with regard to urea and creatinine. fects, such as catheter malposition, resulting in However, the failure to adequately remove access failure. other solutes such as potassium may require Severe Malnutrition Resistant to Aggressive switching to another form of renal replacement Management (Relative). Due to the continuous therapy. protein loss associated with PD, malnourished Inadequate ultrafiltration is usually second• patients must be aggressively evaluated and ary to high transport characteristics or a me• treated. If treatment of malnutrition is not suc• chanical defect hampering catheter patency or cessful, transfer to HD is indicated. drainage. 136 In rare instances, inadequate ultra• filtration is associated with low peritoneal RECOMMENDATIONS FOR RESEARCH transport characteristics, probably due to a sig• nificant reduction in the area of the peritoneal The topic of suitability of patients for PD or membrane, or secondary to increased perito• HD has not been thoroughly investigated. Pro• neal lymphatic flow. 137 spective comparisons of PD and HD for specific Unacceptably Frequent Peritonitis. The def• ESRD patient categories (eg, those with severe inition of unacceptably frequent peritonitis has heart failure, those with advanced malnutrition, to be individually determined for each patient. those with large body size, etc.) are needed to Such considerations as the availability of hemo• define the subsets of ESRD patients which are dialysis facilities will inevitably playa role. most suitable or unsuitable for PD. IX. References

I. Committee to Advise the Public Health Service on Clin• 17. Rodby RA, Firanek CA, Cheng YG, Korbet SM: Re• ical Practice Guidelines, in Field, Ml, Lohr, KN (eds): Clini• producibIlity of studies of peritoneal dialysis adequacy. Kid• cal Practice Guidelines: Directions for a New Program. ney Int 50:267-271, 1996 Washington, DC, National Academy Press, 1990 18. Twardowski Zl, Nolph KD, Khanna R, Prowant BF, 2. American Medical Association/Specialty Society Prac• Ryan LP, Moore HL, Nielsen MP: Peritoneal equilibration tice Parameters Partnership Attributes to Guide the Develop• tests. Perit Dial Bull 7:138-147, 1987 ment and Evaluation of Practice Pararmeters/Guidelines, 19. Warady BA, Hossli S: Peritoneal membrane transport 1996, pp 1-10 function in children receiving long-term dialysis. 1 Am Soc 3. U.S. Renal Data System: The USRDS dialysis morbid• Nephrol 7:2385-2391, 1996 ity and mortality study (wave I), in National Institutes of 20. Rocco MY, 10rdan lR, Burkart 1M: Changes in perito• Health, National Institute of Diabetes and Digestive and Kid• neal transport during the first month of peritoneal dialysis. ney Diseases (ed): U.S. Renal Data Systems 1997 Annual Perit Dial Int 15:12-17, 1995 Data Report, chapter 4. Bethesda, MD, 1997, pp 49-67 21. Hiroshige K, Yuu K, Soejima M, Takasugi M, Kuro• 4. Kopple ID, Chumlea WC, Gassman 11, Hotlinger DL, iwa A: Rapid decline of residual renal function in patients Maroni Bl, Merrill D, Scherch LK, Schulman G, Zimmer on automated peritoneal dialysis. Perit Dial Int 16:307-315, GS: Relationship between GFR and nutritional status-Re• 1996 sults from the MDRD study. 1 Am Soc Nephrol 5:335-335, 22. Keen M, Lipps B, Gotch F: The measured creatinine 1994 (abstr) generation rate in CAPD suggests only 78% of prescribed 5. Pollock CA: Protein intake. 1 Am Soc Nephrol 8:777- dialysis is delivered. Adv Perit Dial 9:73-75, 1993 783, 1997 23. Goel SP, Nolph KD, Moore HL, Prowant BF, Khanna 6. Ikizler T A, Greene IH, Wingard RL, Parker RA, Hakim R, Twardowski Zl: A prospective study of the effect of non• RM: Spontaneous dietary protein intake during progression compliance on small solute removal in continuous ambulatory of chronic renal failure. 1 Am Soc Nephrol 6:1386-1:r91, peritoneal dialysis. ASAIO 1 41:M452-M456, 1995 1995 24. Gregory M, Holley lL: Assessing compliance with the 7. Hertel 1, Lightfoot B, Fincher ME, McConnell KR, Ca• dialysis prescription in children on peritoneal dialysis. Perit runa RJ: Correlation between urea reduction rates and serum DialInt 16:S69, 1996 (suppl 2) albumin levels in patients on hemodialysis. ASAIO 1 25. Johansson A-C, AUman P, Haraldsson B: Creatinine 41:M801-M804, 1995 generation rate and lean body mass: A critical analysis in 8. Lowrie EG, Lew NL: Death risk in hemodialysis pa• peritoneal dialysis patients. Kidney Int 51:855-859, 1997 tients: The predictive value of commonly measured variables 26. Mitch WE, Walser M: A proposed mechanism for and an evaluation of death rate differences between facilities. reduced creatinine excretion in severe chronic renal failure. Am J Kidney Dis 15:458-482, 1990 Nephron 21 :248-254, 1978 9. Hakim RM, Lazarus 1M: Initiation of dialysis. J Am 27. Cockroft DW, Gault MH: Prediction of creatinine Soc Nephrol 6: 1319-1328, 1995 clearance from serum creatinine. Nephron 16:31-41, 1976 10. Ratcliffe PI, Phillips RE, Oliver DE: Late referral for 28. Warren PI, Brandes lC: Compliances with the perito• maintenance dialysis. Br Med 1 288:441-443, 1984 neal dialysis prescription is poor. 1 Am Soc NephroI4:1627- II. Hood SA, Schillo B, Beane EG, Rozas Y, Sondheimer 1629, 1994 IH, Members of The Michigan Renal Plan Task Force of the 29. Nolph KD, Twardowski ZJ, Khanna R, Moore HL, Michigan Public Health Institute Okemos, MI: An analysis Prowant BF: Predicted and measured daily creatinine produc• of the adequacy of preparation for end-stage renal disease tion in CAPD: Identifying noncompliance. Perit Dial Int 15:22-25, 1995 care in Michigan. ASAIO 1 41:M422-M426, 1995 30. Tzamaloukas AH, Braun M, Malhotra D, Murata GH: 12. Jungers P, Zingraff 1, Page B, Albouze G, Estimated versus predicted creatinine generation as an indica• Hannedouche T, Man N: Detrimental effects of late referral tor of compliance with the prescribed dose of continuous in patients with chronic renal failure: A case control study. peritoneal dialysis. Int 1 Artif Organs 19:151-155, 1996 Kidney Int 43:S170-S173, 1993 (suppl 41) 31. Blake PG, Spanner E, McMurray S, Lindsay RM, Fer• 13. Ifudu 0, Dawood M, Homel P, Friedman EA: Excess guson E: Comparison of measured and predicted creatinine morbidity in patients starting uremia therapy without prior excretion is an unreliable index of compliance in PD patients. care by a nephrologist. Am 1 Kidney Dis 28:841-845, 1996 Perit Dial Int 16:147-153, 1996 14. Khan IH, Catto GRD, MacLeod AM: Chronic renal 32. Tzamaloukas AH: Can an excess of estimated over• failure: Factors influencing nephrology referral. Q 1 Med predicted creatinine generation be a discriminating test for 87:559-564, 1994 non-compliance in continuous ambulatory peritoneal dial• 15. Levey AS, Adler S, Caggiula A W, England BK, ysis? J Am Soc Nephrol 6:1519-1520, 1995 (letter) Greene T, Hunsicker LG, Kusek lW, Rogers NL, Teschan 33. Burkart 1M, Bleyer AI, 10rdan lR, Zeigler NC: An PE: Effects of the dietary protein restriction on the progres• elevated ratio of measured to predicted creatinine production sion of advanced renal disease in the modification of diet in in CAPD patients is not a sensitive predictor of noncompli• renal disease study. Am 1 Kidney Dis 27:652-663, 1996 ance with the dialysis prescription. Perit Dial Int 16: 142-146, 16. Churchill DN, Taylor DW, Keshaviah PR: Adequacy 1996 of dialysis and nutrition in continuous peritoneal dialysis: 34. Tzamaloukas AH: Pharmokinetic analysis of creati• Association with clinical outcomes. 1 Am Soc Nephrol 7: 198- nine generation discrepancy as an index of noncompliance 207, 1996 in CAPD. Adv Perit Dial 12:61-65, 1996

AmericanJournal of Kidney Diseases, Vol 30, No 3, Suppl 2 (September), 1997: pp S105-S108 S105 S106 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

35. Lo W-K, Prowant BF, Gamboa SB, Moore HL, Nolph bolic rate in patients on continuous peritoneal dialysis: A KD: How reproducible is daily creatinine recovery in chronic multivariate predictive model. ASAIO J 42:46-51, 1996 peritoneal dialysis. Perit Dial Int 14:286-288, 1994 53. Harty J, Faragher B, Venning M, Gokal R: Urea ki• 36. Watson PE, Watson !D, Batt RD: Total body water netic modelling exaggerates the relationship between nutri• volumes for adult males and females estimated from simple tion and dialysis in CAPD patients. (The hazards of cross• anthropometric measurements. Am J Clin Nutr 33:27-39, sectional analysis). Perit Dial Int 15:105-109, 1995 1980 54. Malhotra D, Tzamaloukas AH, Murata GH, Fox L, 37. Hume R, Weyers E: Relationship between total body Goldman RS, Avasthi PS: Serum albumin in continuous peri• water and surface area in normal and obese subjects. J Clin toneal dialysis: Its predictors and relationship to urea clear• Pathol 24:234-238, 1971 ance. Kidney Int 50: 1501-1507, 1996 38. Mellits ED, Cheek DB: The assessment of body water 55. Lindsay RM, Spanner E: A hypothesis: the protein and fatness from infancy to adulthood. Monographs Soc Res catabolic rate is dependent upon the type and amount of Child Dev, Serial 140,35:12-26, 1970 treatment in dialyzed uremic patients. Am J Kidney Dis 39. Du Bois D, Du Bois EF: A formula to estimate the 13:382-389, 1989 approximate surface area if height and weight be known. 56. Lindsay RM, Spanner E, Heidenheim RP, LeFebvre Arch Int Med 17:863-971, 1916 JM, Hodsman A, Baird J, Allison MEM: Which comes first, 40. Gehan E, George SL: Estimation of human body sur• KiN or PCR-chicken or egg? Kidney Int Suppl 42:S32- face area from height and weight. Cancer Chemother Rep S36, 1992 (suppl 38) Part I 54:225-235, 1970 57. Kopple JD, Jones MR, Keshaviah PR, Bergstrom J, 41. Haycock GB, Chir B, Schwartz GJ, Wisotsky DH: Lindsay RM, Moran J, Nolph KD, Teehan BP: A proposed Geometric method for measuring body surface area: A height• glossary for dialysis kinetics. Am J Kidney Dis 26:963-981, weight formula validated in infants, children, and adults. J 1995 Pediatr 93:62-66, 1978 58. Keshaviah PR, Nolph KD, Moore HL, Prowant B, 42. Bailey BJR, Briars G: Estimating the surface area of Emerson PF, Meyer M, Twardowski ZJ, Khanna R, Ponfer• the human body. Stat Med 15:1325-1332, 1996 rada L, Collins A: Lean body mass estimation by creatinine 43. Tzamaloukas AH, Saddler MS, Murphy G, Morgan kinetics. J Am Soc NephroI4:1475-1485, 1994 K, Goldman RS, Murata GH, Malhotra D: Volume of distri• 59. Forbes GB, Bruning GJ: Urinary creatinine excretion bution and fractional clearance of urea in amputees on contin• and lean body mass. Am J Clin Nutr 29:1359-1366, 1976 uous ambulatory peritoneal dialysis. Perit Dial Int 14:356- 60. Bhatia B, Moore H, Emerson P, Keshaviah P, Prowant 361, 1994 B, Nolph KD, Singh A: Lean body mass estimation by creati• 44. Tzamaloukas AH, Malhotra D: Creatinine clearance nine kinetics, bioimpedance, and dual energy x-ray absorpti• in amputees on CPD. Perit Dial Int 16:426-426, 1996 ometry in patients on continuous ambulatory peritoneal dial• 45. Tzamaloukas AH, Murata GH, Malhotra D, Fox L, ysis. ASAIO J 41:M442-M446, 1995 Goldman RS: Determinants of residual renal function in 61. Tzamaloukas AH, Murata GH, Malhotra D, Fox L, CAPD. Adv Perit Dial 1997 (in press) Goldman RS, Avasthi PS: Creatinine clearance in continuous 46. Lameire NH, Vanholder R, Veyt D, Lambert M, Rin• peritoneal dialysis: Dialysis dose required for a minimal ac• goir S: A longitudinal, five year survey of urea kinetic param• ceptable level. Perit Dial Int 16:41-47, 1996 eters in CAPD patients. Kidney Int 42:426-432, 1992 62. Tzamaloukas AH, Murata GH, Malhotra D, Fox L, 47. CANUSA Peritoneal Dialysis Study Group: Canada• Goldman RS, Avasthi PS: The minimal dose of dialysis re• quired for a target in continuous peritoneal dialysis. USA (CANUSA) multicentre study of peritoneal dialysis ade• KTN Clin Nephrol 44:316-321, 1995 quacy: Description of the study population and preliminary 63. Borah MF, Schoenfeld PY, Gotch FA, Sargent JA, results. Adv Perit Dial 8:88-92, 1992 Wolfson M, Humphreys MH: Nitrogen balance during inter• 48. Teehan BP, Schleifer CR, Brown J: Adequacy of con• mittent dialysis therapy of uremia. Kidney Int 14:491-500, tinuous ambulatory peritoneal dialysis: morbidity and mortal• 1978 ity in chronic peritoneal dialysis. Am J Kidney Dis 24:990- 64. Keshaviah PR, Nolph KD: Protein catabolic rate calcu• 1001, 1994 lations in CAPD patients. ASAIO Trans 37:M400-M402, 49. Nolph KD, Moore HL, Prowant B, Meyer M, Twar• 1991 dowski ZJ, Khanna R, Ponferrada L, Keshaviah P: Cross 65. Mendley SR, Majkowski NL: Protein intake and urea sectional assessment of weekly urea and creatinine clearances and nitrogen appearance in stable pediatric PD patients. J and indices of nutrition in continuous ambulatory peritoneal Am Soc Nephrol 7:1432-1432, 1996 (abstr) dialysis patients. Perit Dial Int 13:178-183,1993 66. Gotch FA: Kinetic modeling in hemodialysis, in Nis• 50. Lutes R, Holley JL, Perlmutter J, Piraino B: Correla• senson AR, Fine RN, Gentile DE (eds): Clinical Dialysis, tion of normalized protein catabolic rate to weekly creatinine chapter 8 (ed 3). East Norwalk, CT, Appleton & Lange, pp clearance and KTN in patients on peritoneal dialysis. Adv 156-189, 1995 Perit Dial 9:97-100, 1993 67. Gotch FA, Keen M: Kinetic modeling in peritoneal 51. Bergstrom J, Furst P, Alverstrand A, Lindholm B: dialysis, in Nissenson AR, Fine RN, Gentile DE (eds): Clini• Protein and energy intake, nitrogen balance and nitrogen cal Dialysis, chapter 16 (ed 3). Norwalk, CT, Appleton & losses in patients treated with continuous ambulatory perito• Lange, pp 343-375, 1995 neal dialysis. Kidney Int 44:1048-1057, 1993 68. Kohaut EC, Waldo FB, Benfield MR: The effect of 52. Murata GH, Tzamaloukas AH, Malhotra D, Saddler changes in dialysate volume on glucose and urea equilibra• MC, Fox L, Wood B, Morgan K, Goldman RS: Protein cata- tion. Perit Dial Int 14:236-239, 1994 REFERENCES S107

69. Robertson BC, luhasz NM, Walker PI, Raymond KH, 87. Burkart 1, Schreiber M, Tabor T, Korbet S, Stallard Taber TE, Adcock AA: A prescription model for peritoneal R, PD Collaborative Group: Prospective, multicenter evalua• dialysis. ASAIO 1 41:116-126, 1995 tion of patient transfer from PD to HD in 1995. Perit Dial 70. Gotch FA, Lipps Bl, Keen ML, Panlilio F, Coinvestigators Int 16:S66, 1996 (abstr; suppl 2) for the Fresenius Randomised Dialysis Prescriptions and Clinical 88. Bloembergen WE, Stannard DC, Port F, Wolfe RA, Outcome Study.(RDP/CO): Computerized urea kinetic modeling Pugh lA, 10nes CA, Greer lW, Golper TA, Held PI: Relation• (UKM) to prescribe and monitor delivered KtIV (pKtIV, dkTN) ship of dose of hemodialysis and cause-specific mortality. in peritoneal dialysis. Adv Perit Dial 12:43-45, 1996 Kidney Int 50:557-565, 1996 71. Vonesh EF, Burkart 1, McMurray SD, Williams PF: 89. Davies Sl, Bryan 1, Phillips L, Russell GI: The pre• Peritoneal dialysis in kinetic modeling: Validation in a dictive value of KTN and peritoneal solute transport in multicenter clinical study. Perit Dial Int 16:471-481, 1996 CAPD patients is dependent on the type of comorbidity pres• 72. Verrina E, Amici G, Trivelli A, Perfumo F, Gusmano ent. Perit Dial Int 16:S158-S162, 1996 (suppl 1) R: Clinical validation of PD Adequest software in children: 90. Young GA, Young JB, Hobson SM, Hildreth B, a pilot study. Perit Dial Int 17:591-591, 1997 Brownjohn AM, Parsons FM: Nutrition and delayed hyper• 73. Maiorca R, Brunori G, Zubani R, Cancarini GC, Manili sensitivity during continuous ambulatory peritoneal dialysis L, Camerini C, Movilli E, Pola A, d'Avolio G, Gelatti U: Pre• in relation to peritonitis. Nephron 43:177-186, 1986 dictive value of dialysis adequacy and nutritional indices for 91. TattersalllE, Doyle S, Greenwood RN, Farrington K: mortality and morbidity in CAPD and HD patients. A longitudi• Maintaining adequacy in CAPD by individualizing the dial• nal study. Nephrol Dial Transplant 10:2295-2305, 1995 ysis prescription. Nephrol Dial Transplant 9:749-752, 1994 74. Teehan BP, Schleifer CR, Brown 1: Urea kinetic mod• 92. Genestier S, Hedelin G, Schaffer P, Faller B: Prognos• eling is an appropriate assessment of adequacy. Semin Dial tic factors in CAPD patients: a retrospective study of a 10- 5:189-192, 1992 year period. Nephrol Dial Transplant 10:1905-1911, 1995 75. DepnerTA: Single-compartment model, in Prescribi~ 93. U.S.Renal Data System: U.S. Renal Data System 1991 Hemodialysis: A Guide to Urea Modeling, chapter 4. Nor• Annual Data Report. National Institutes of Health, National well, MA, Kluwer Academic Publishers, 1991, pp 81-88 Institute of Diabetes and Digestive and Kidney Diseases, 76. Twardowski ZJ: Clinical value of standardized equili• Bethesda, MD, 1991, pp E57-E61 bration tests in CAPD patients. Blood Purif 7:95-108, 1989 94. U.S. Renal Data System: Hospitalization, in National 77. Keshaviah P, Emerson PF, Vonesh EF, Brandes lC: Institutes of Health, National Institute of Diabetes and Diges• Relationship between body size, fill volume, and mass trans• tive and Kidney Diseases (ed): U.S. Renal Data Systems 1996 fer area coefficient in peritoneal dialysis. 1 Am Soc Nephrol Annual Data Report, Bethesda, MD, 1996 4:1820-1826, 1994 95. Cagney KA, Wu A W, Herbert RJ, Meyer KB, lenckes 78. Schoenfeld PY, Diaz-Buxo 1, Keen M, Gotch FA: The MW, Bass EB, Powe NR: Formal assessment of quality of effect of body position (P), surface area (BSA), fill volume life instruments for end-stage renal disease. Med Decision and intraperitoneal exchange volume (Vip) on the peritoneal Making 15:421-421 1995 transport constant. 1 Am Soc NephroI4:416-416, 1993 (abstr) 96. Meyer KB, Espindle DM, DeGiacomo 1M, lenulson 79. Blake PG, Abraham G, Sombolos K, Izatt S, Weissgar• CS, Kurtin PS, Davies AR: Monitoring dialysis patients' ten 1, Ayiomamitis A, Oreopoulos DG: Changes in peritoneal health status. Am I Kidney Dis 24:267-279, 1994 membrane transport rates in patients on long term CAPD. 97. Wu AW, Cagney KA, Fink NE, Meyer KB, Herbert Adv Perit Dial 5:3-7, 1989 RJ, Ienckes MW, Sadler IH, Powe NR: Development of a 80. Vonesh EF, Lysaght Ml, Moran 1, Farrell P: Kinetic quality life measure for renal dialysis patients. Quality Life modeling as a prescription aid in peritoneal dialysis. Blood Res 4:505-505, 1995 Purif 9:246-270, 1991 98. Powe NR, Klag MI, Sadler IH, Anderson GF, Bass 81. Smith L, Folden T, Youngblood B, Panlilio F, Gotch EB, Briggs W A, Fink NE, Levey AS, Levin NW, Meyer KB, F: Clinical evaluation of a peritoneal dialysis kinetic model• Rubin HR, Wu A W: Choices for healthy outcomes in caring ing set. Adv Perit Dial 12:46-48, 1996 for end-stage renal disease. Semin Dial 9:9-11, 1996 82. Randerson D, Farrell P: Mass transfer properties of 99. Hays RD, Kallichj JD, Mapes DL, Coons SI, Carter the human peritoneum. ASAIO 1 6:140-146, 1983 WB: Development of the kidney disease quality of life 83. Garred LJ, Canaud B, Farrell PC: A simple kinetic (KDQUOL) instrument. Quality Life Res 3:329-338, 1994 model for assesssing peritoneal mass transfer in chronic am• 100. Avram MM, Goldwasser P, Erroa M, Fein PA: Pre• bulatory peritoneal dialysis. ASAIO 1 6:131-137, 1983 dictors of survival in continuous ambulatory peritoneal dialysis 84. Owen WF lr, Lew NL, Yan Liu MS, Lowrie EG, patients: The importance of prealbumin and other nutritional Lazarus 1M: The urea reduction ratio and serum albumin and metabolic markers. Am I Kidney Dis 23:91-98, 1994 concentration as predictors of mortality in patients undergo• 101. Strnijk DG, Krediet RT, Koomen GCM, Boeschoten ing hemodialysis. N Engl 1 Med 329:1001-1006, 1993 EW, Arisz L: The effect of serum albumin at the start of 85. Gotch FA, Gentile DE, Keen M: The incident of patient continuous ambulatory peritoneal dialysis treatment on pa• cohort study design with uncontrolled dose may result in a tient survival. Perit Dial Int 14:121-126, 1994 substantial overestimation of mortality (M) as a function of 102. Brandes IC, Piering WF, Beres lA, Blumenthal SS, peritoneal dialysis (PD) dose. ASAIO Trans 42:102-102, 1996 Fritsche C: Clinical outcome of continuous ambulatory peri• 86. Kimmel PL, Bosch IP: Effectiveness of renal fellow• toneal dialysis predicted by urea and creatinine kinetics. I ship training for subsequent clinical practice. Am 1 Kidney Am Soc Nephrol 2:1430-1435, 1992 Dis 18:249-256, 1991 103. Heaf I: CAPD adequacy and dialysis morbidity: Det- S108 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY rimental effect of a high peritoneal equilibration rate. Ren Oreopoulos DG: Successful use of continuous ambulatory Fail 17:575-587, 1995 peritoneal dialysis in refractory heart failure. Perit Dial Bull 104. Blake PG, Flowerdew G, Blake RM, Oreopoulos DG: 5:127-130, 1985 Serum albumin in patients on continuous ambulatory perito• 121. McKinnie n, Bourgeois RJ, Husserl FE: Long-term neal dialysis: Predictors and correlations with outcomes. J therapy for heart failure with continuous ambulatory perito• Am Soc Nephrol 3:1501-1507, 1993 neal dialysis. Arch Intern Med 145:1128-1129, 1985 105. Blumenkrantz MJ, Kopple JD, Moran JK, Grodstein 122. Rubin J, Ball R: Continuous ambulatory peritoneal GP, Coburn JW: Nitrogen and urea metabolism during continu• dialysis as a treatment of severe congestive heart failure in ous ambulatory peritoneal dialysis. Kidney Int 20:78-82, 1981 the face of chronic renal failure. Report of eight cases. Arch 106. Ronco C, Bosch JP, Lew SQ, Feriani M, Chiaramonte Intern Med 146:1533-1535, 1986 S, Conz P, Brendolan A, La Greca G: Adequacy of continu• 123. Konig PS, Lhotta K, Kronenberg F, loannidid M, ous ambulatory peritoneal analysis: Comparison with other Herold M: CAPD: A successful treatment in patients suffer• dialysis techniques. Kidney Int Suppl 46:S 18-S24, 1994 ing from therapy-resistant congestive heart failure. Adv Perit (suppI48) Dial 7:97-101, 1991 107. Mactier RA, Perry M, Henderson IS: Relationships be• 124. Stegmayr B, Banga R, Lundberg L, Wikdahl AM, tween dialysis quantification and normalized protein catabolic Plum-Wirell M: PD treatment for severe congestive heart rate in peritoneal dialysis. Perit Dial Int 13:S508-S511, 1993 failure. Perit Dial Int 16:S231-S235, 1996 (suppI2) 125. Tzamaloukas AH, Murata GH, Harford AM, Sena P, Zager PG, Eisenberg B, Wood B, Simon D, Goldman RS, 108. Blake PG, Sombolos K, Abraham G, Weissgarten J, Kanig SP: Hand gangrene in diabetics on chronic dialysis. Pemberton R, Chu GL, Oreopoulos DG: Lack of correlation ASAIO Trans 37:638-643, 1991 between urea kinetic indices and clinical outcomes in CAPD 126. Coles GA: When should chronic peritoneal dialysis be patients. Kidney Int 39:700-706, 1991 recommended over hemodialysis? Semin Dial 2:213-214, 1989 109. Stein A, Walls J: The correlation between KtIV and 127. Zappacosta AR, Perras ST: The process of prescrib• protein catabolic rate-A self-fulfilling prophecy. Nephrol ing CAPD, in CAPD: Continuous ambulatory peritoneal dial• Dial Transplant 9:743-745, 1994 ysis, chapter 1. Philadelphia, PA, Lippincott, 1984, pp 4-23 110. Harty J, Boulton H, Heelis N, Uttley L, Venning 128. Korbet SM, Rodby RA: Peritoneal membrane failure: M, Gokal R: Limitations of kinetic models as predictors of Differential diagnosis, evaluation, and treatment. Semin Dial nutritional and dialysis adequacy in continuous ambulatory 7:128-137, 1994 peritoneal dialysis patients. Am J Nephrol 13:454-463, 1993 129. Gulanikar AC, Jindal KK, Hirsch DJ: Is chronic peri• 111. Salusky IB: Nutritional management of pediatric pa• toneal dialysis safe in patients with intrabdominal prosthetic tients on chronic dialysis, in Nissenson AR, Fine RN, Gentile vascular graphs? Nephrol Dial Transplant 6:215-217, 1991 DE (eds): Clinical Dialysis, chapter 23 (ed 3). Norwalk, CT, 130. Charytan C: Continuous ambulatory peritoneal dial• Appleton & Lange, 1995, pp 535-548 ysis after abdominal aortic graft surgery. Perit Dial Int 112. Salusky IB, Fine RN, Nelson P, Blumenkrantz MJ, 12:227-229, 1992 Kopple JD: Nutritional status of children undergoing continuous 131. Schmidt RJ, Cruz C, Dumler F: Effective continuous ambulatory peritoneal dialysis. J Clin Nutr 38:599-611, 1983 ambulatory peritoneal dialysis following abdominal aortic an• 113. Maiorca R, Cancarini GC, Brunori G, Camerini C, eurysm repair. Perit Dial Int 13:40-44, 1994 Manili L: Morbidity and mortality of CAPD and hemodialy• 132. Nolph KD, Jensen RA, Khanna R, Twardowski ZJ: sis. Kidney Int 43:S4-S15, 1993 Weight limitations for weekly urea clearances using various 114. Zawada ET Jr: Indications for dialysis, in Daugirdas exchange volumes in continuous ambulatory peritoneal dial• JT, Ing TS (eds): Handbook of Dialysis (ed 2). Toronto, ysis. Perit Dial Int 14:261-264, 1994 Canada, Little Brown, 1994, pp 6-7 133. Tzamaloukas AH, Dimitriadis A, Murata GH, Nico• 115. Khanna R, Nolph KD, Oreopoulos DG: Peritoneal lopoulou N, Malhotra D, Balaskas EV, Kakavas J, Antoniou dialysis in diabetics, in The Essentials of Peritoneal Dialysis, S, Dombros NV, Batzili E, Voudiklari S: Continuous perito• chapter 10. Boston, MA, Kluwer Academic Publishers, 1993, neal dialysis in heavyweight individuals: Urea and creatinine pp 99-100 clearances. Perit Dial Int 16:302-306, 1996 116. Pritchard SS: Treatment modality selection in 150 134. Beardsworth SF: The complications of peritoneal di• consecutive patients starting ESRD therapy. Perit Dial Int alysis, in Practical Peritoneal Dialysis, chapter 6. Bristol, 16:69-72, 1996 England, John Wright & Sons, 1984, pp 63-69 117. Charytan C, Spinowitz BS: Dialysate Leaks, in Nis• 135. Holley JL, Piraino B: Careful patient selection and senson AR, Fine RN (eds): Dialysis Therapy, chapter 15 (ed dialysis prescription are required for effective nightly inter• 2). Philadelphia, PA, Hanley and Belfus, 1993, pp 188-189 mittent peritoneal dialysis. Perit Dial Int 14:155-158, 1994 118. Mactier R: Non-renal indications for peritoneal dial• 136. Tzamaloukas AH, Saddler MC, Murata GH, Malho• ysis. Adv Perit Dial 8:141-144, 1992 tra D, Sena P, Simon D, Hawkins KL, Morgan K, Nevarez 119. Robson MD, Biro A, Knobel B, Schai C, Mordchia M, Wood, Ellege L, Gibel LJ: Symptomatic fluid retention in R: Peritoneal dialysis in refractory congestive heart failure. patients on continuous peritoneal dialysis. J Am Soc Nephrol Part II. Continuous ambulatory peritoneal dialysis (CAPD). 6:198-206, 1995 Perit Dial Bull 3:133-134, 1983 137. Mactier RA: Investigation and management of ultra• 120. Kim D, Khanna R, Wu G, Fountas P, Druck M, filtration failure in CAPD. Adv Perit Dial 7:57-62, 1991 x. Appendices Appendix A: Detailed Ration~le for Guideline 1

GUIDELINE I cially important when glomerular filtration rate (GFR) falls below 25 to 50 mL/min, at which When to Initiate Dialysis-KtlVurea time spontaneous decrease in dietary protein in• 3 6 Criterion (Opinion) take is commonly observed. - The blood urea nitrogen (BUN) and serum creatinine values Unless certain conditions are met, patients should not be used to monitor progression of should be advised to initiate some fonn of dial• renal failure, particularly in patients with diabe• ysis when the weekly renal KtIV urea (KrtIV urea) tes.2 BUN may be low secondary to low protein falls below 2.0. The conditions that may indicate intake and may not adequately reflect the degree dialysis is not yet necessary even though the of the renal functional impainnent. Serum creati• weekly KrtIV urea is less than 2.0 are: nine may be low due to decreased muscle mass 1. Stable or increased edema-free body as seen in some women, the elderly, and in mal• weight. Supportive objective parameters for ade• nourished patients. Hence, serum creatinine con• quate nutrition include a lean body mass >63%, centration may not adequately reflect the degree subjective global assessment score indicative of of the renal functional impairment. adequate nutrition (see Guideline 12: Nutritional The estimation of V (total body water) by any Status Assessment and Appendix B: Detailed 19- fonnulae has not been validated in children with tionale for Guideline 2) and a serum albumin renal failure. Thus, the use of KrtIV urea as an indi• concentration in excess of the lower limit for the cation for the initiation of PD is recommended lab, and stable or rising; and considering this caveat. Creatinine clearance as 2. nPNA ~0.8 g/kg/d (see Guideline 2: When a means of assessing RRF for purposes of initia• To Initiate Peritoneal Dialysis-nPNA Criteria tion of dialysis should be normalized to body and Appendix B: Detailed Rationale for Guide• surface area (BSA). line 2); and 1 7 10 An increasing body of evidence • - suggests 3. Complete absence of clinical signs or that KrtIV urea is a reliable predictor of outcome symptoms attributable to uremia. in PD and that weekly values in the range of A weekly KrtIV urea of 2.0 approximates a renal 2.0 provide adequate therapy (see Guideline 15: urea clearance of 7 mL/minand a renal creatinine Weekly Dose of CAPD). Although the CANUSA clearance that varies between 9 to 14 mL/mini data indicated a linear decrease in modeled mor• 1.73 m2. Urea clearance should be nonnalized tality rate with increasing KprtIV up to 2.3, there to total body water (V) and creatinine clearance is some uncertainty about the significance of the should be expressed per 1.73 m2 of body surface high KprtIV levels achieved in this study.ll area. The GFR, which is estimated by the arith• It has always been a paradox that nephrologists metic mean of the urea and creatinine clearances, have insisted on optimal/therapy once patients will be approximately 10.5 mL/minl1.73 m2 are started on dialysis but have accepted much when the KrtlVurea is about 2.0. lower levels of renal function, defined as Krtl Rationale In patients with chronic renal dis• ease, progression of renal failure should be moni• Vurea , during the pre-dialysis phase of patient tored by following total weekly renal urea nitro• management. For example, while we recognize gen clearance (Krturea) nonnalized to urea volume that a weekly KprtIV urea of 2.0 or higher is associ• of distribution (V), i.e., KrtlVurea.l.2 This does not ated with improved outcome on PD, dialysis is usually not initiated until weekly KrtIVurea is in imply that a weekly collection of urine is neces• 2 sary. A daily collection multiplied by seven the range of 0.71 to 1.3. ,12 It is possible that the yields a reasonable approximation of weekly consequences of delaying initiation of PD may clearance. The knowledge of KrtIV urea is espe- be analogous to the experience of the National Cooperative Dialysis Study, wherein the mortal• ity rate in the year after the study ended was © 1997 by the National Kidney Foundation, 1nc. more than twice as high in those randomized to 0272-6386/97/3003-0219$3.00/0 the low dose dialysis protocols, even though they

AmericanJournal of Kidney Diseases, Vol 30, No 3, Suppl 2 (September), 1997: pp S109-S133 S109 S110 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY were returned to standard dialysis after complet• 4.00- ..... 1' ..... ;..... ~ ..... ~ ..... T" .... :.. ···r···-1-·····r······i·····T···-l ing 24 weeks in the high BUN (low clearance) arm of the study.13 The Work Group feels that the data of Ikizler i••• et aV McCusker et al,12 and Pollock et al6 ~ ::••••• ;;:•• I ••••• I]] ••••• !•• ·.l.··: ••• ·; strongly demonstrate the linkage between de• >. 2.00 : : , creasing renal function and worsening nutritional ! 1.50' ..... j...... l····-1-····-1-············:······j······[······i·····j······j status. In the CANUSA study, multiple estimates 1 00- ..... ;...... ;....,..... ;...... ;..... +----+----+--... ,.. -.. +·····;·-----i of nutritional status were associated with two es• . ::::::Started ~Inctementar' PO timates of RRF.12 Patients who started PO at 0.50' -··--j·--···j-···+----j------j------t--·--t-·---j····t-----i-----j------l lower levels of RRF had a worse nutritional sta• 0.00 ' , ' , , . " '" tus than those who started at a higher level of o 1 2 3 4 5 6 7 8 9 10 11 12 RRF. CANUSA further demonstrated an associa• Elapsed lime, Months tion between the relative risk of death and worse Fig 11-1. Solute removal and initiation of PD. Exam• baseline serum albumin concentration, worse ple of tracking solute clearance measurements for a time-dependent SGA, and worse time-dependent single patient over a 12-month period. The point at which incremental PO was initiated is indicated by the percent lean body mass. 1 While an association arrow. between risk of death and nPCR could not be demonstrated in the multivariate analysis, several univariate analyses did demonstrate an associa• total solute removal utilizing the incremental ini• tion of individual estimates of baseline nutri• tiation approach, more intense scrutiny of RRF tional status with survival. is necessary. For initiation with full dose PO, These data are consistent with the observations less intense scrutiny of RRF is indicated. This is of Bonomini et al who found that patients starting discussed in Guideline 3: Frequency of Delivered dialysis with a residual renal creatinine clearance PO Dose and Total Solute Clearance Measure• of <5 mL/min had a worse long-term outcome ment Within Six Months of Initiation and Guide• than patients starting incremental hemodialysis line 5: Frequency of Measurement of KtIV ure., with a mean residual renal creatinine clearance Total CCr> PNA, and Total Creatinine Appear• of 11 mUmin. 14 ance, which address frequency of measurements. From these observational data, it seems rea• In the CANUSA study, 1 the weekly CCr equiv• 2 sonable to draw the following conclusions: alent to a KprtlVurea of 2.0 was 70 UwklI.73 m • Once KrtIV urea falls below 2.0 per week or cre• As will be clear later in this discussion, a CCr atinine clearance falls into the range of 9 to 14 level this high is indicative of residual renal func• 2 mUmin/I.73 m , initiation of dialysis or trans• tion, which was clearly present in the CANUSA plantationl5.I6 should be strongly advised. The patients at initiation of PD. patient should be considered to be at increasing CAPO is the only continuous chronic renal risk with any further decreases in KrtIV urea in the replacement therapy with which to quantitatively absence of renal replacement therapy interven• compare continuous residual renal solute clear• tion. If dialysis is not instituted when KrtIV urea ance. The Work Group strongly supports the falls below 2.0, it is mandatory to document: (1) opinion that the outcome data for a weekly KtI nPNA ;;:,:0.8 g/kg/d (see Guideline 2 and Appen• V ure• of ;;:,:2.0 are so compelling that using the dix B), (2) stable or increased body weight with• same figure for initiation of dialysis justifies the out edema, and (3) a negative inquiry for clinical unknown but presumably small risks of per• signs or symptoms of uremia. When PO is initi• forming peritoneal dialysis. Those risks include ated, the KptIV urea could be increased incremen• infections and the possibility that increasing the tally such that the combined value of KrtIV urea + length of time on PO contributes to eventual pa• KptIV urea does not fall below the target level of tient "burn-out." If a patient is suspected to be 2.0 (see Figure II-I and Guideline 15: Weekly at high risk for these complications, PO may not Dose of PO). Alternatively, the initiation of "full be the best choice for renal replacement therapy. dose" PO may be offered. Since residual renal The Work Group recognizes that the patient function (RRF) is such a crucial component of will playa major role in accepting the initiation APPENDICES S111

of dialysis based on a certain' 'laboratory value." pool) coefficient previously developed I? is 5.0, there• It is the responsibility of the care providers to fore, make clear to the patient the rationale for initiat• ing dialysis when the above conditions become 1.0 = 5.0* Kr ureaN or (7) applicable. Reasons to justify a delay in initiating Kr urea = 0.20*V (8) dialysis are listed above. These reasons should be documented, if present. Since Equations 4, 6, and 8 are identical, it is The Work Group also recognizes that for many apparent that PD, biw HD, and tiw HD should all clinicians, initiating dialysis based on KtIV urea is be started when Kr urea = 0.20*V. For an average a new concept. Therefore, we have attempted patient with V = 35 L, this defines a level of to equate this to the traditional measure of urea Krurea = 7.0 mL/rnin. There are constraints on the clearance, creatinine clearance, and GFR (esti• lower level of Kr urea for biw HD. I? As Figure II-I mated by the arithmetic mean of urea and creati• suggests, treatment could be started incrementally nine clearance). once weekly KrtlVurea falls below 2.0. Typically, What follows is an explicit quantitative ap• this would involve a single overnight exchange, proach to the concept of renal urea clearance intended to restore KtIV urea to 2.0 per week. Ultra• (Kr urea, mL/min). We recommend that adequate filtration would not be needed since at this level PD be considered to require: of KrtIV urea urine volumes are usually adequate. Levels of Residual Renal Creatinine Clear• KprtIV = 2.0 per week where Kprt is total weekly ance, C Cn mUmin at Which Dialysis Should Be peritoneal plus renal urea clearance. Guideline Initiated. There are no Krcr criteria for HD, and 15: Weekly Dose of CAPD explains the ration• the criteria defining the contribution of residual ale for recommending that KprtIV be 2.0 per renal function (Kr Cr) to therapy are different from week. (1) the criteria defining the contribution of peritoneal For the purpose of this discussion, Kr urea is creatinine clearance (Kpcr) to the dose oftherapy. considered equivalent to Kp urea and, therefore, The problem arises from the argument that tubu• lar secretion of creatinine should be subtracted KprtIV = 1.44*7*Kr ureaN, where 1.44 converts from the total renal creatinine clearance, and re• mL/min to Lid, when KrtIV is 2.0. (2) nal function with respect to creatinine clearance is best expressed as "GFR" as developed below. 2.0 = 1.44*7*Kr ureaN and (3) Therefore, the definitions used with respect to Kr Cr will each be considered separately and re• Krurea = 0.20*V (4) lated to the level of KtIV urea for which Kr urea and Equation 4 shows that Kr urea must equal 0.2 Kp urea are considered simply additive. times V when KrtIV = 2.0. It is of interest to In all instances, the total weekly CCr is normal• compare this criterion to those developed indepen• ized to 1.73 m2 of BSA. In order to compare dentlyl? for hemodialysis (HD). For twice weekly this to the KprtlVurea, BSA must be normalized (biw) HD, a coefficient has been developed I? from relative to V, e.g., based on the Hume equations urea kinetic modeling to convert Kr urea, mUrnin, and taking the mean of the genders, 1.73 m2 = to L of equivalent urea clearance during each 35 L.19 It is also reasonable to extend this as a twice weekly dialysis. The twice weekly adequate linear relationship over the domain of patient size level of KtIV urea 18 in HD is 1.85 (double pool) and although it should also be considered a gender• the coefficient to convert Kr urea to equivalent urea dependent relationship. clearance during dialysis is 9.0 with units of LI Consideration of Renal Contribution to Dose treatmentlmL Kr urea. Therefore, Expressed as CrC,.. As noted above, in all cases the dose of total creatinine clearance is expressed 1.85 = 9.0*Kr ureaN or (5) as Llwkl1.73 m2 of BSA. The following will nor• malize total creatinine clearance to 1.73 m2 Kr urea = 0.20*V (6) (nBSA) and KtlVurea to a standard V = 35 L corre• Equation 6 for biw HD is identical to Equation sponding to 1.73 m2 of BSA in order to develop 4 for PD. For thrice weekly (tiw) HD the (double constants relating the creatinine and urea-based S112 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

200 dosage parameters. To the extent that BSA and V (1) K,,)/nBSA =70(K"W) increase and decrease at the same ratio (reason• (2) eK""VnBSA = 52(K"W) (1) ably valid), the constants developed are generaliz• (3) K,c,VnBSA = 28(K"W) E 150 able, and therefore, the relation between Kr Cr and C'l I': (2) nBSA can be expressed as follows: -.~ -. 100 ~ = >- [Kr crtlnBSA] 1.44*7*Kr Cr 32 CIl 10.1 *Kr Cr (9) ~ 50 Equation 9 simply describes the total weekly liters of renal creatinine clearance as O~~~~~~~~~~~~~~~ a function of Kr Cr and our normalized BSA of 0.00 0.50 1.00 1.50 2.00 2.50 2 Weekly K"t/V 1.73 m • For the KtlVurea normalized to V = 35, there• Fig 11-2. Dose of PD with respect to weekly creati• fore, nine clearance relative to weekly K.,rtIV varies widely depending on the definition of renal creatinine clear• ance.

= 0.29*Kr urea and, therefore, (10)

(11) [eKr Cr tlnBSA] = 5*6.94 (KrtIV) Assuming + 5*3.47 (KrtIV) = 52 (KrtIV) (17)

Kr Cr = 2 *Kr urea, we can substitute Equation 11 Equation 17 shows that if the renal contribu• into Equation 12 to show (12) tion to PD therapy is defined as GFR, which is equivalent to "effective" creatinine clearance or eKr Cr, the total weekly eKr Cr required relative to K 2*3.47*K tlnV 6.94*KrtlnVurea (13) 2 rcr = r = KrtlVurea is 52 L/wkl1.73 m per unit of KrtlVurea. Substituting now Equation 13 into Equation 9 It can also be noted that in all instances these yields the following equation: equations also relate to KprtIV urea and KptIV urea since we have defined Kr = Kp. [Krcr tlnBSA] = 10.1 [6.94 (KrtlnV)] Consideration of Peritoneal Creatinine Clear• = 70 (Krtln V) (14) ance to PD Dose. In this case, the dose must be expressed directly in terms of peritoneal creat• Equation 14 shows that if we define the renal inine clearance (Kpcr) and by definition, contribution to PD therapy by Kr Cr, the level of weekly Kr Cr per 1.73 m2 must be 70 times the [KpcrtlnBSA] = 1.44*7*Kpcr = 10.1 Kpcr (18) 2 KrtlnVurea so at KrtlnVurea of 2.0, Kr Cr tl1.73 m The average relationship between Kpcr and is 140. For the average patient with V = 35 L Kp urea in CAPD is 2 and BSA = 1.73 m , the required level of Kr Cr = 14 rnL/min. KpCr = 0.8*Kp urea (19) Consideration of Renal Contribution to Dose Since Kp urea = Kr urea and Kp ureatln V = Kr ureatl Expressed as "GFR." In this case, the effec• nY, Equations 18, 19, and 11 can be combined tive renal creatinine clearance, eKr Co is defined to derive as: [KpcrtlnBSA] = 28 [Kp urea tlnV] eKr Cr = GFR = [Kr Cr + Kr urea] 12 (15) = 28 [Kr urea tin V] (20) Therefore, Equations 14, 17, and 20 show that the Creati• [eKrcr tlnBSA] = 1.44*7 (Krcr + Krurea)12 nine Dose Equivalency with respect to the single urea Kprt N criterion will vary widely depending = 5.0*Kr Cr + 5.0*Kr urea (16) on how RRF is defined. The relationships are plot• Substituting from Equations 11 and 13, yields ted in Figure 11-2, Dose of PD With Respect to APPENDICES S113

Weekly Creatinine Clearance Relative to Weekly Another way to view the creatinine clearance KprtlV, where the weekly creatinine clearance re• at which to initiate dialysis is to extrapolate back• quired per 1.73 m2 corresponding to a weekly Kp,t I ward from the CAPO target of 60 Llweekll.73 V urea of 2.0 ranges from 140 to 56 L, depending m2 (see Guideline 15: Weekly Dose of PD.) This on the definitions used. There is no problem in approximates a purely filtered only creatinine either the case of pure residual renal function with clearance of 6 mLimin. Since at this level of no PD therapy or the case of pure peritoneal dial• residual renal function much of the creatinine ysis with no residual renal function present. The appearing in the final urine is from tubular secre• problem arises when one attempts to sum residual tion, 60 Llwkl1.73 m2 approximates a total resid• renal creatinine clearance with peritoneal creati• ual renal creatinine clearance of 9 to 14 mL/min. nine clearance. In this common circumstance, the Peritoneal Dialysis and Residual Renal Func• dialysis dose relationship will be bounded by re• tion Equivalency. Quantitative replacement of gression lines 2 and 3 in Figure 11-2. renal urea clearance by peritoneal clearance is For line 1, which defines creatinine clearance based on the assumption that the two clearance as the uncorrected (for secretion) creatinine parameters confer equal clinical benefit with re• clearance, the creatinine clearance that is equiva• spect to control of uremic morbidity. Thus, we 2 lent to a KtlVurea of 2.0 is 140 Llwkl1.73 m • can write the relationship Line 2 defines creatinine clearance as the mean KprtlV urea = KptIV urea + KrtIV urea' where KptIV urea of urea and creatinine clearance, and the equiva• is total daily or weekly peritoneal urea clearance lence to a KtlVurea of 2.0 is 104 Llwkl1.73 ~. normalized to V; KrtlV urea is total daily or weekly Line 3 represents all clearance from PO (com• renal urea clearance normalized to V. (21) plete absence of residual renal function). Under Solution of Equation 21 for KptlV with the this condition a KtlVurea of 2.0 is equivalent to a urea 2 assumption that adequate weekly KprtIV urea is 2.0 creatinine clearance of 56 Llwkl1.73 m • This creates an irreconcilable ambiguity with results in respect to the creatinine and urea dosage criteria KptIV urea = 2.0 - KrtlV urea (22) for defining optimal dialysis and the study of outcome as a function of dose. Because the prac• Equation 22 provides a quantitative guideline tice of PO has used both CCr and KtIV to quantify for replacing residual renal urea clearance by delivered dose and there is a large body of litera• peritoneal clearance such that the sum of weekly ture describing outcomes related to Cc" the Work KptIV urea and KrtIV urea remains 2.0. Group recommends continuing to use both mea• Hemodialysis and Residual Renal Function sures (see Guideline 4: Measures of PO Dose Equivalency. Compared to CAPO it is more and Total Solute Clearance.) However, in view complex to calculate incremental doses of HD of this ambiguity, the Work Group recommends such that the sum of intermittent dialyzer clear• that if only one measure is to be utilized, use ance (KdtlVurea) and continuous KrtlVurea remain KprtlVurea rather than Kr crt (see Guideline 15: constant at a level equiv51lent to a weekly Krtl Weekly Dose of CAPO.) Nonetheless, creatinine Vurea of 2.0. However, .the dose and frequency of kinetics as discussed in Guidelines 4, 6, and 17 HD which provide therapy equivalent to continu• are useful for estimating edema-free, fat-free ous KptlV urea can be calculated using the funda• body mass, compliance with dialysis prescrip• mental assumption underlying CAPO therapy: tion, and some programs may prefer it for quanti• the level of continuous KptlV urea required for fication of delivered dose of PD. Thus, total cre• treatment which is clinically equivalent to inter• atinine excretion is valuable. mittent HD is that KptlV urea which results in a Finally, it is worth emphasizing that the basic steady state BUN equal to the average predialysis relationship between the level of residual Cr Cr to BUN with any specific intermittent HD treatment 20 23 Kr urea for initiation of dialysis will be the same schedule at the same nPCR. - From this basic for all of these creatinine dosage criteria. They assumption and the urea kinetic model 19 the dose are related to KprtIV since Kr Cr = 2 Kr urea and and frequency of HD required for incremental all of the expressions ultimately reduce to this replacement of KrtlVurea as it falls below 2.0 can relationship. be readily calculated as depicted in Figure 11-3. S114 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

I_ N=3 0 N=2 0 N=11 equivalent to 2.0. Therefore, for a weekly Krtl 2.00 Vurea <0.5, more than twice weekly HD will be necessary. Finally, in the case of N = 3, eK.!tI Vurea increases linearly to 1.05 as KrtlVurea falls to zero. The eK.!tIV urea of 1.05 corresponds to single pool K.!tlVurea values of 1.46 and 1.20 at treatment times of 2.0 and 4.0 hours, respec• tively. There is emerging evidence that RRF is better preserved in patients undergoing HD with the use of more biocompatible membranes. 25

APPENDIX A REFERENCES Fig 11-3. Equivalent total dialysis doses for incre• 1. Churchill DN, Taylor DW, Keshaviah PR: Adequacy mental replacement of K,.Wurea calculated with the as• of dialysis and nutrition in continuous peritoneal dialysis: sumptions that (1) K" Kp , and ~ are clinically equiva• Association with clinical outcomes. J Am Soc Nephrol 7:198- lent clearance terms and (2) the intermittent dialysis 207, 1996 dose schedule is equivalent to continuos dialysis 2. Tattersall JE: Is continuous ambulatory peritoneal dial• when average pre-dialysis BUN equals steady state ysis an adequate long-term therapy for end-stage renal-dis• BUN with continuous therapy at equal npeR. Note that ease? Semin Dial 8:72-76. 1995 e~Wure. equals the equilibrated (double-pool), deliv• ered, and normalized hemodialysis dose. 3. Kopple JD, Chum1ea WC, Gassman JJ, Hotlinger DL, Maroni BJ, Merrill D, Scherch LK, Schulman G, Zimmer GS: Relationship betweeen GFR and nutritional status - re• sults from the MDRD study. J Am Soc Nephro1 5:335-335, The dose of intermittent HD is expressed in 1994 (abstr) 4. lkizler TA, Greene JH, Wingard RL, Parker RA, Hakim Figure 11-3 as eK.!tlVurea which is the equilibrated, RM: Spontaneous dietary protein intake during progression delivered and normalized hemodialysis dose (see of chronic renal failure. J Am Soc Nephrol 6:1386-1391, the NKF-DOQI Clinical Practice Guidelines for 1995 Hemodialysis Adequacy for further discussion of 5. Hakim RM, Lazarus JM: Initiation of dialysis. J Am eK.!tlVurea). The equilibrated measure is utilized Soc Nephrol 6:1319-1328, 1995 6. Pollock CA: Protein intake. J Am Soc Nephrol 8:777- here because in peritoneal dialysis, transcellular 783, 1997 urea equilibration is achieved, and therefore, it 7. Teehan BP, Schleifer CR, Brown JM, Sigler MH, Rai• makes conceptual sense to think in terms of mondo J: Urea kinetic analysis and clinical outcome on equilibrated values. From preliminary data of the CAPD. A five year longitudinal study. Adv Perit Dial 6:181- HEMO study, eK.!tIV is approximately 0.21 185, 1990 8. Maiorca R, Cancarini G, Brunori G, Zubani R, Camerini lower than that computed from immediate post• C, Manili L, Campanini M, Mombelloni S: Which treatment HD BUN sampling, using single-pool, variable for which patient in the future? Possible modifications in volume kinetic modeling.24 The dashed line de• CAPD. Nephrol Dial Transplant 10:20-26, 1995 picts incremental increase in daily KptIV urea as 9. Tattersall JE, Doyle S, Greenwood RN, Farrington K: KrtIV urea falls in accordance with Equation 22. Kinetic modelling and underdialysis in CAPD patients. Nephrol Dial Transplant 8:535-538, 1993 Model solutions are shown for once, twice and 10. Maiorca R, Brunori G, Zubani R, Cancarini GC, Man• thrice weekly hemodialysis (N = 1, 2, and 3, iIi L, Camerini C, Movilli E, Pola A, d'Avolio G, Gelatti U: respectively). The model solutions are limited to Predictive value of dialysis adequacy and nutritional indices eK.!tlVurea ::;;2.0 since it is unrealistic to prescribe for mortality and morbidity in CAPD and HD patients. A eK.!tlVurea >2.0. Such a dose would correspond longitudinal study. Nephrol Dial Transplant 10:2295-2305, 1995 to single pool K.!tlVurea values of 2.8 and 2.3 11. Gotch FA, Gentile DE, Keen M: The incident of pa• with treatment times (t) of 2.0 and 4.0 hours, tient cohort study design with uncontrolled dose may result respectively. It can be seen that when N = 1, in a substantial overestimation of mortality (M) as a function eK.!tlVurea = 2.0 when KrtIVurea = 1.6. Thus, the of peritoneal dialysis (PD) dose. ASAIO Trans 42:102-102, option for once weekly hemodialysis is limited 1996 12. McCusker PM, Teehan BP, Thorpe K, Keshaviah P, to a weekly KrtIV urea? 1.6. If KrtIV urea = 0.5 and Churchill D: How much peritoneal dialysis is required for N = 2, the eK.!tIV urea for each HD treatment must the maintenance of a good nutritional state? Kidney Int be 2.0 to achieve a weekly continuous KtIV urea 50:S56-S61, 1996 (suppl 56) APPENDICES S115

13. Hakim R: Assessing the adequacy of dialysis. Kidney anthropometric body water in patients on CPD. Perit Dial Int Int 37:822-832, 1990 15:284-285, 1995 (letter) 14. Bonomini V, Feletti C, Scolari MP, Stefoni S: Benefits 20. Popovich RP, Moncrief JW: Kinetic modeling of peri• of early initiation of dialysis. Kidney Int 28:S57-S59, 1985 toneal transport. Contrib Nephrol 17:59-72, 1979 15. Roake JA, Cahill AP, Gray CM, Gray DRW, Morris 21. Teehan BP, Schleifer CR, Sigler MH, Gilgore GS: A PJ: Preemptive cadaveric renal transplantation-Clinical out• quantitative approach to the CAPD prescription. Perit Dial come. Transplantation 62: 1411-1416, 1996 Bull 5:152-156, 1985 16. Katz SM, Kerman RH, Golden DG, Grevel J, Camel 22. Keshaviah PR, Nolph KD, Prowant B, Moore H, Pon• S, Lewis RM, Van Buren CT, Kahan BD: Preemptive trans• ferrada L, Van Stone J, Twardowski ZJ, Khanna R: Defining plantation-An analysis of benefits and hazards in 85 cases. adequacy of CAPD with urea kinetics. Adv Perit Dial 6: 173- Transplantation 51:353-355, 1991 177, 1990 23. Gotch FA: Prescription criteria in peritoneal dial• 17. Gotch FA, Keen M: Kinetic modeling in peritoneal ysis. Perit Dial Int 14:S83-S87, 1994 dialysis, in Nissenson AR, Fine RN, Gentile DE (eds): Clini• 24. Eknoyan G, Levey AS, Beck GJ, Agodoa L Y, Daugir• cal Dialysis (ed 3). Norwalk, CT, Appleton & Lange, 1995, das JT, Kusek JW, Levin NW, Schulman G: The hemodialy• pp 343-375 sis (HEMO) study: Rationale for selection of interventions. 18. Wish JB, Webb RL, Levin AS, Port FK, Held PJ: Semin Dial 9:24-33, 1996 Medical appropriateness of initiating chronic dialysis in the 25. McCarty JT, Jenson BM, Squillace BS, Williams A W: U.S.: Results of the revised medical evidence report pilot Improved preservation of residual renal function in chronic study. J Am Soc Nephrol 5:345 1994 (abstr) hemodialysis patients using polysulfone dialyzers. Am J Kid• 19. Tzamaloukas AH, Murata GH: Body surface area and ney Dis 29:576-583, 1997

Appendix B: Detailed Rationale for Guideline 2

GUIDELINE 2 and these patients must be monitored very closely. This is particularly relevant since the When to Initiate Dialysis-nPNA Criterion USRDS reports that 45% of incident PO patients (Opinion) in 1996 had never visited with a dietitian prior It is recognized that an adequately supple• to dialysis. 3 mented, tightly monitored, low protein diet may However, often there is a spontaneous reduc• slow the progression of renal failure in certain tion in dietary protein intake as renal failure prog• circumstances. When properly performed, such resses. The USROS reports that at least 40% of diets do not result in loss of lean body mass or incident PO patients in 1996 began dialysis after other manifestations of malnutrition. However, months of nausea/vomiting.3 Estimating PNA progressive renal failure is associated with spon• from renal urea nitrogen kinetics will provide taneous anorexia and malnutrition. Under these an estimate of protein intake. Malnutrition often circumstances and in the absence of comorbid follows prolonged decreases in protein intake, causes of anorexia, and after unsuccessful inter• unless adequate energy intake is added to com• vention by a registered dietitian, dialysis should pensate. PNA measurements can be used to assist be started in adult patients when nPNA spontane• in documenting malnutrition. Malnutrition pres• ously falls below 0.8 g/kg/d. ent at the initiation of dialysis has an adverse Rationale The extrapolation of data from di• effect on survival. Thus, monitoring nutritional alysis patients to pre-dialysis patients and vice indices in pre-dialysis patients provides a means versa may be appropriate, if cautiously applied. of early recognition so that appropriate interven• The Work Group attempts in this rationale to tional measures may prevent further deterioration address the pre-dialysis patient popUlation. It is of nutrition. The initiation of dialysis is one such recognized that an adequately supplemented, appropriate measure, when causes of malnutri• tightly monitored, low protein diet may slow the tion other than uremia have been ruled out. progression of renal failure in certain circum• The level of spontaneous protein intake (as stances 1 and may delay the need for dialysis. reflected by urea nitrogen appearance) prior to Patients being treated by such diets do not dem• the initiation of dialysis has been shown in some onstrate findings of malnutrition or loss of lean studies under certain circumstances to be pre• 4 8 body mass. Clearly though, energy intake must dictive of outcome. - A spontaneous decrease in be increased to perhaps as high as 40 kcal/kg/d,2 protein intake occurs once the GFR falls below S116 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

Fig 11-4. Algorithm for iniating dialysis in patients with chronic renal failure.

9 50 mLlmin ,1O and may fall below 0,7 g/kg/d Rationale for Guideline 12). This technique, ini• when the GFR is <25 mL/min,4 In addition, it tially described in surgical patients, has been val• has been shown that serum albumin concentra• idated in ESRD patients. IS Four items (weight 7 tion ,11-15 and the subjective global assessment loss during a predetermined interval, anorexia, (SGA) of nutritional status at initiation of PD are subcutaneous fat, and muscle mass) are scored both predictive of subsequent patient outcome.6 on a 7-point Likert scale. Likert scales are re• The method of SGA of nutritional status used in sponses to a question in which the answers have the CANUSA Study was a modification of the multiple discrete response options and direction• l6 original method ,17 (see Appendix F: Detailed ality. Values as high as 6 or 7 reflect normal APPENDICES S117

nutrition and values as low as 1 or 2 reflect severe a broader view of nutritional status should be malnutrition. Intermediate values such as 3, 4, or neglecte~. On the contrary, the presence of symp• 5 reflect moderate to mild malnutrition. Further toms, such as the following, should influence the confirmation of the relation between RRF and clinical judgment regarding the need for dialysis: s nutritional status has been presented. Using uni• • anorexia variate analysis, a lower level of renal function • weight loss at the time of initiation of PD (measured by Krtl • nausea V urea, Cer , or urea clearance) was associated with • vomiting worse nutritional status (serum albumin concen• • falling caloric or protein intake by history tration, lean body mass, SGA, and nPNA), • decreased fat-free, edema-free (lean) body which, in turn, was predictive of a poor clinical mass by creatinine kinetics (see Guideline 13: outcome on PD over the subsequent 24 months. Determining Fat-Free, Edema-Free Body Mass) While an optimal level of protein intake for or other measures PD patients has not been rigorously defined, most • declining subjective global assessment of would agree that neutral nitrogen balance could nutritional status (see Appendix F) be maintained at an intake of 1.0 to 1.2 g/kg/d, • decreasing serum albumin concentration, and that negative nitrogen balance is likely at etc. levels of protein intake below 0.8 g/kg/d. 19 There These topics are discussed from a slightly dif• is a marked difference between nPNA and ferent perspective in Guideline 28: Measurement weekly KprtIVurea in patients with continuorrs of Normalized PNA in PD Patients. A detailed clearance (CAPD and pre-dialysis chronic renal review of nutrition in ESRD and pre-dialysis pa• failure patients) versus patients on chronic hemo• tients has recently been presented.23 An algo• dialysis. 20 In the former groups, KprtIV values of rithm outlining this approach is attached (see Fig• 2.0 or more were required to achieve a nPNA ure 11-4). value of 0.9 g/kg/d or greater. Therefore, the Work Group recommends that APPENDIX B REFERENCES patients with progressive renal failure should 1. Levey AS, Adler S, Caggiula A W, England BK, Greene have quarterly estimates of nPNA (from 24-hour T, Hunsicker LG, Kusek JW, Rogers NL, Teschan PE: Effects of the dietary protein restriction on the progression of ad• urine collections) once GFR falls below 25 mLi vanced renal disease in the modification of diet in renal dis• min. When nPNA is below 0.8 g/kg/d, vigorous ease study. Am 1 Kidney Dis 27:652-663, 1996 nutritional intervention should be undertaken. 2. Rigalleau V, Combe C, Blanchetier V, Aubertin 1, Some patients may be specifically advised to Aparicio M, Gin H: Low protein diet in uremia: Effects on maintain protein intake below 0.8 g/kg/d by de• glucose metabolism and energy production rate. Kidney Int 51:1222-1227, 1997 sign, with the close involvement of a registered 3. U.S. Renal Data Systems: The USRDS dialysis morbid• dietitian, a caloric intake of at least 35 kcal/kg/ ity and mortality study (wave 1), in National Institutes of d,2l and careful follow-up with the nephrologist. Health, National Institute of Diabetes and Digestive and Kid• However, a spontaneous reduction of protein ney Diseases (ed): U.S. Renal Data Systems 1997 Annual intake to below 0.8 g/kg/d is suggestive of ure• Data Report, chapter 4. Bethesda, MD, 1997, pp 49-67 4. !killer TA, Greene JH, Wingard RL, Parker RA, Hakim mia. Comorbid conditions which may cause an• RM: Spontaneous dietary protein intake during progression of orexia such as gastroparesis, infection, acidosis chronic renal failure. 1 Am Soc Nephrol 6:1386-1391, 1995 and depression should be identified and 5. Hakim RM, Lazarus 1M: Initiation of dialysis. 1 Am treated.22 A registered dietitian must be involved Soc Nephrol 6:1319-1328, 1995 by this stage. It is appropriate to utilize the reg• 6. Churchill DN, Taylor DW, Keshaviah PR: Adequacy of dialysis and nutrition in continuous peritoneal dialysis: Associa• istered dietitian at the dialysis facility to whom tion with clinical outcomes. 1 Am Soc Nephrol 7:198-207, 1996 the patient will be referred. However, if these 7. Tattersall IE: Is continuous ambulatory peritoneal dial• measures fail to restore nPNA to 0.8 g/kg or ysis an adequate long-term therapy for end-stage renal-dis• higher, dialysis should be initiated and increased ease? Semin Dial 8:72-76, 1995 incrementally until a satisfactory level of nPNA 8. McCusker FM, Teehan BP, Thorpe K, Keshaviah P, Churchill D: How much peritoneal dialysis is required for is achieved. the maintenance of a good nutritional state? Kidney Int The recommendation to use an nPNA criterion 50:S56-S61, 1996 for the initiation of dialysis does not imply that 9. Kopple JD, Chumlea WC, Gassman JJ, Hollinger DL, S118 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

Maroni BJ, Merrill D, Scherch LK, Schulman G, Zimmer GS: Whittaker S, Mendelson RA, Jeejeebhoy KN: What is subjec• Relationship betweeen GFR and nutritional status-Results tive global assessment of nutritional status? J Parenter Enteral from the MDRD study. J Am Soc Nephrol 5:335-335, 1994 Nutr 11 :8-13, 1987 (abstr) 17. Baker JP, Detsky AS, Wesson DE, Wolman SL, Stew• 10. Pollock CA: Protein intake. J Am Soc Nephrol 8:777- art S, Whitewell J, Langer B, Jeejeebhoy K: Nutritional as• 783, 1997 sessment. A comparison of clinical judgement and objective 11. Teehan BP, Schleifer CR, Brown JM, Sigler MH, Rai• measurements. N Engl J Med 306:969-972, 1982 mondo J: Urea kinetic analysis and clinical outcome on CAPD. 18. Enia G, Sicuso C, Alati G, Zoccali C: Subjective A five year longitudinal study. Adv Perit Dial 6:181-185, 1990 global assessment of nutrition in dialysis patients. Nephrol 12. Nolph KD, Moore HL, Prowant B, Meyer M, Twar• Dial Transplant 8:1094-1098, 1993 dowski ZJ, Khanna R, Ponferrada L, Keshaviah P: Cross 19. Blumenkrantz MJ, Kopple JD, Moran JK, Coburn JW: sectional assessment of weekly urea and creatinine clearances Metabolic balance studies and dietary protein requirements and indices of nutrition in continuous ambulatory peritoneal in patients undergoing continuous ambulatory peritoneal dial• dialysis patients. Perit Dial Int 13: 178-183, 1993 ysis. Kidney lnt 21 :849-861, 1982 13. Blake PG, Flowerdew G, Blake RM, Oreopoulos DG: 20. Mehrotra R, Saran R, Moore HL, Nolph KD: Toward Serum albumin in patients on continuous ambulatory perito• targets for initiation of dialysis: the relationship of protein neal dialysis: Predictors and correlations with outcomes. J intake (PI) to KTN ,," in predialysis patients. J Am Soc Am Soc Nephrol 3:1501-1507, 1993 u 14. Spiegel DM, Breyer JA: Serum albumin: A predictor Nephrol 7:1521, 1996 (abstr) of long-term outcome in peritoneal dialysis patients. Am J 21. Kopple JD, Monteon FJ, Shaib JK: Effect of energy Kidney Dis 23:283-285, 1994 intake on nitrogen metabolism in nondialyzed patients with 15. Struijk DG, Krediet RT, Koomen GCM, Boeschoten chronic renal failure. Kid Int 29:734-742, 1986 EW, Arisz L: The effect of serum albumin at the start of 22. Bergstrom J: Anorexia in dialysis patients. Semin Dial continuous ambulatory peritoneal dialysis treatment on pa• 16:222-229, 1996 tient survival. Perit Dial Int 14:121-126, 1994 23. lkizler TA, Hakim RM: Nutrition in end-stage renal 16. Detsky AS, McLaughlin JR, Baker JP, Johnston N, disease. Kidney Int 50:343-357, 1996

Appendix C: Detailed Rationale for Guideline 6

GUIDELINE 6 over time with both HO and PO. Several studies have compared the rate of decline of RRF with Assessing Residual Renal Function 2 S the two dialytic modalities - and demonstrated (Evidence) that RRF is preserved better in patients undergoing Residual renal function (RRF), which can pro• PO therapies compared to HD. In a study of 25 vide a significant component of total solute and CAPO patients and 25 HD patients, the rate of de• water removal, should be assessed by measuring cline of creatinine clearance was significantly slower the renal component of KtlVurea (KrtIVurea) and over the first 18 months of dialysis in the CAPO estimating the patient's glomerular filtration rate patients.2 The PO patients started dialysis with an (GFR) by calculating the mean of urea and creati• uncorrected CCr of 4.4 mUmin, and after 18 months nine clearance. it was 4.0 mUmin. In the HD patients, CCr at initia• Rationale The contribution of RRF to total tion was 4.3 mUmin, and after 18 months it was solute and water clearance is significant (30% to 1.3 mUmin (P <0.01 compared to PO). 50%), especially during the first few years of Similar differences were observed in patients dialysis therapy. Assessment of RRF is important with diabetes. S In another study comparing the for several reasons. A substantial fraction (30%) urine output and CCo the urine output dropped of the total renal replacement therapy may be significantly in HO patients at the end of one provided by RRF when a patient begins PO. I year compared to 3 years in CAPO patients? The After 2 years of PO, the RRF may still contribute mean annual decline of CCr was identical in HO about 15% of the total KtlVurea . Since the RRF and CAPO for patients with primary glomerulo• contribution will be added to that of PO, it will pathy. However, in the groups with nephro• be measured in the same units and for the same sclerosis and interstitial nephropathy, the rate of >alutes. decline of CCr was significantly slower in CAPO Preservation of RRF may be of particular im• compared to HO patients. In another retrospec• portance to the effectiveness of long term PO tive study of 4 years duration which compared :herapy. There is a progressive decline of RRF 55 CAPO patients to 57 HO patients, the rate of APPENDICES S119

decline in the HD group was twice that of the V urea in both initiation of dialysis and for follow• CAPD group.4 This difference persisted after ad• ing RRF changes over time. justment for age, gender, hypertensive status, and KrtIV~rea as a measure of RRF is recommended the use of ACE inhibitors. Children have a better because total KtlVurea is associated in a clinically preservation of urinary volume, but not GFR, in important and statistically significant way with 11O those receiving PD.7 patient survival . ,1l (see Guideline 15: Weekly Despite their limitations, these studies gener• Dose of CAPD). The peritoneal clearance of cre• ally demonstrate a slower rate of decline of RRF atinine is about 80% of the urea clearance, while in patients on PD compared to HD. They also at end stage the renal clearance of creatinine is demonstrate that the rate of the decline varies about 1.5 to 2 times that of urea. Perhaps as a from patient to patient. The faster rate of decline consequence of this physiological phenomenon ofRRF in HD is speculated to be due to repetitive or for other reasons, there is a discrepancy be• hypotensive episodes, possibly complement acti• tween total KtlVurea and total CCr normalized to 2 vation and cytokine release, and the possibility 1.73 m BSA (see paragraphs below). In the case that the more efficient HD may remove GFR of discrepancy, KtIV urea is preferentially recom• stimulatory factors. mended to determine PD adequacy, because it is Several methods for measuring the CCr compo• more predictable and reproducible, and is inde• nent of RRF are available. These include the un• pendent of the confounding effects of renal secre• corrected CCr, a flat percentage of uncorrected tion of creatinine. A retrospective study of PD CCr as an estimate of GFR, or the average-of adequacy demonstrated an association between creatinine and urea clearance also as an estimate KtIV urea and outcomes. 12 of GFR. This emphasis on using KrtIV urea is not in• tended to detract from the utility of CCr. In terms While each method has its particular merits, 2 the Work Group recommends using the arithme• of validity, total CCr normalized to 1.73 m BSA is predictive of patient survival, technique sur• tic mean of creatinine and urea clearances to de• vival, and hospitalization. 1 The creatinine gener• termine the RRF component to CCr and as an ation rate is useful for assessment of nutritional estimate of GFR. Therefore, the CCr component status, in particular, in measuring fat-free, of RRF will subsequently refer to residual renal edema-free body mass. Total CCr may also be CCo corrected for secretion by taking the arith• useful for assessment of compliance. metic mean of urea and creatinine clearances. CCr as an index of PD adequacy is associated This method was selected for several reasons. statistically with both morbidity and mortality 1 First, it was used in some of the major outcome and correlates with urea clearance. 13 Discrepan• studies used in establishing these guidelines (see cies between the two clearances may be found Guideline 15, Weekly Dose of CAPD.) Second, in approximately 20% of PD subjects. 14,15 The the corrected CCr correlates better with KtIV urea main reasons for the discrepancies are the pres• 9 than the uncorrected CCr. Third, it makes con• ence of substantial RRF, which tends to cause ceptual sense because the peritoneal transport of disproportionately high Ccr values, and low peri• creatinine is by diffusion and convection, not se• toneal solute transport type, which tends to cause cretion. The correction process addresses this. disproportionately low CCr values. 14.15 CCr values However, the Work Group recommends using corresponding to a weekly KtIV urea of 2.0 differ urea clearance, normalized to total body water, between CAPD subjects with and without RRF i.e., KrtlVurea, as the key measure to follow seri• (see Figure X-2, referenced previously in Appen• ally to determine whether urine collections need dix A: Detailed Rationale for Guideline 1). In to continue (see Guideline 11: Dialysate and patients with RRF, the mean CCr corresponding Urine Collections.) This is termed the renal KtJ to a KtlVurea of 2.0 weekly is between 60.513 and 15 2 V urea or KrtIV urea' This recommendation was 67.6 L/wkl1.73 m • In anuric CAPD subjects, made to simplify the concept of a residual renal the mean CCr corresponding to a KtIV urea of 2.0 2 15 component to the total renal replacement dose. weekly is 52.1 L/wkl1. 73 m • KtIV urea is believed to be the more valuable mea• In the case of a discrepancy, the Work Group sure of renal replacement therapy and the Work recommends the use of KtIV urea as an immediate Group carried this thinking through to using KrtJ guide of dialysis adequacy because it directly S120 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

relates to protein metabolism. However, if there Residual renal function in children on haemodialysis and is a discrepancy between CCr and KtIV ure., the peritoneal dialysis therapy. Pediatr Nephrol 8:579-583, 1994 8. Rottembourg J: Residual renal function and recovery patient must be observed closely because initially of renal function in patients treated by CAPO. Kidney Int it may not be clear why the discrepancy exists 43:SI06-SIIO, 1993 and the reason may be important. This is dis• 9. Bhatia B, Moore HL, Nolph KD: Modification of creati• cussed in Guidelines 1, 7, and 15. nine clearance by estimation of residual creatinine and urea clearance in CAPO patients. Adv Perit Dial 11:101-105, 1995 APPENDIX C REFERENCES 10. Maiorca R, Brunori G, Zubani R, Cancarini GC, Ma• nili L, Camerini C, Movilli E, Pola A, d'Avolio G, Gelatti 1. Churchill DN, Taylor DW, Keshaviah PR: Adequacy U: Predictive value of dialysis adequacy and nutritional indi• of dialysis and nutrition in continuous peritoneal dialysis: ces for mortality and morbidity in CAPD and HD patients. Association with clinical outcomes. J Am Soc NephroI7:198- A longitudinal study. Nephrol Dial Transplant 10:2295-2305, 207, 1996 1995 2. Rottembourg J, Issad B, Gallego JL, Degoulet P, Aime 11. Teehan BP, Schleifer CR, Brown JM, Sigler MH, Rai• F, Gueffaf B, Legrain M: Evolution of residual renal function mondo J: Urea kinetic analysis and clinical outcome on in patients undergoing maintenance haemodialysis or contin• CAPD. A five year longitudinal study. Adv Perit Dial 6:181- uous ambulatory peritoneal dialysis. Proc Eur Dial Transplant 185, 1990 Assoc 19:397-403, 1982 12. Selgas R, Bajo MA, Fernandez-Reyes MJ, Bosque E, 3. Cancarini GC, Brunori G, Camerini C, Brass S, Manili Lopez-RevueJta K, Jimenez C, Borrego F, De Alvaro F: An L, Maiorca R: Renal function recovery and maintenance of analysis of adequacy of dialysis in a selected population on residual diuresis in CAPD and hemodialysis. Perit Dial Bull CAPD for over 3 years: The influence of urea and creatinine 6:77-79, 1986 kinetics. Nephrol Dial Transplant 8:1244-1253, 1993 4. Lysaght MJ, Vonesh EF, Gotch F, Ibels L, Keen M, 13. Nolph KD, Twardowski ZJ, Keshaviah PR: Weekly Lindholm B, Nolph KD, Pollock CA, Prowant B, Farrell PC: clearances of urea and creatinine on CAPD and NIPD. Perit The influence of dialysis treatment modality on the decline DialInt 12:298-303, 1992 of remaining renal function. ASAIO Trans 37:598-604, 1996 14. Chen HH, Shetty A, Afthentopoulos IE, Oreopoulos 5. Hallett M, Owen J, Becker G, Stewart J, Farrell PC: DG: Discrepancy between weekly KT/V and weekly creat• Maintenance of residual renal function: CAPD versus HD. inine clearance in patients on CAPD. Adv Perit Dial II :83- Perit Dial Int 12:S124, 1992 (abstr; suppl I) 87, 1995 6. Lutes R, Perlmutter J, Holley JL, Bernardini J, Piraino 15. Tzamaloukas AH, Murata GH, Malhotra 0, Fox L, B: Loss of residual renal function in patients on peritoneal Goldman RS, Avasthi PS: Creatinine clearance in continuous dialysis. Adv Perit Dial 9:165-168, 1993 peritoneal dialysis: Dialysis dose required for a minimal ac• 7. Feber J, Scharer K, Schaefer F, Mikova M, Janda J: ceptable level. Perit Dial Int 16:41-47, 1996

Appendix D: Detailed Rationale for Guideline 8

GUIDELINE 8 riod. For CCPD patients, the serum sample should be obtained at the midpoint of the daytime Reproducibility of Measurement (Opinion) dwell(s). Rationale To be clinically useful, measure• Accurate measurement of total KtIV urea and to• ment of PD dose must be performed in a valid tal creatinine clearance (CCr) requires collection and reproducible fashion. The following factors and analysis of urine, dialysate and serum in a influence the validity and reliability of KtIV urea way that yields reproducible and valid results. and total Ccr as measures of PD dose. Dialysate creatinine concentration must be cor• Dialysate glucose. Dialysate creatinine con• rected for the presence of glucose in some assays. centration should be corrected for the presence Peritonitis precludes reliable measurement of de• of glucose, which interferes with some creatinine livered PD dose for up to a month. Compliance measurement methodologies. 1 Each facility must with complete collections is mandatory. For pa• determine this by specifically inquiring of its lab• tients who void :2:3 times per day, a 24-hour oratory whether the creatinine assay used by that urine collection is sufficient. For patients who lab is altered by high glucose concentrations. void less frequently, a 48-hour collection is rec• Each laboratory should establish its own correc• ommended. For CAPD patients, the serum sam• tion factor and should reestablish the correction ple can be obtained at any convenient time. For factor if the laboratory's methodology changes. NIPD patients, the serum sample should be ob• The Work Group does not recommend using cor• tained at the midpoint of the daytime empty pe- rection factors from the literature. APPENDICES S121

Peritonitis. Peritoneal solute transport in• The aliquots are measured by syringes; usually creases during peritonitis and usually recovers a 5-mL syringe is accurate enough. The aliquots some time after resolution of peritonitis, with a can be ·mixed in the same container, because the reported recovery time between 3 days2 and 1 sampling proportion from each original bag is month. 3 constant at 111000. The total effluent is recorded, Patient compliance with the dialysis prescrip• and that figure plus the small container with all tion. Following creatinine appearance in dialy• the collected aliquots are brought to the dialysis sate and urine longitudinally is an objective center. Some dialysate manufacturers have de• method to evaluate the degree of compliance (see veloped special aliquoting exchange bags that 7 Guideline 7: PO Oose Troubleshooting). Clinical separate an aliquot as part of the exchange. tools for evaluating compliance are in the process The collection of effluent dialysate from auto• of development.4 mated PO is conceptually similar to that de• Variability of residual renal function (RRF). scribed above. The effluent drained via a cycler Oay-to-day total clearances can vary greatly in is quantified automatically by the cycler and gen• PO. The major portion of this variance is caused erally pools in one collection container, or if in by changes in measured RRF,5 although creati• several containers, free mixing is possible. Since the effluent volume is known and the containers nine generation may vary in apparently stable PO are allowed to mix freely, a sample of any reason• patients.6 able volume (e.g., 10 mL) can be brought to the Completeness of urine collection. To avoid dialysis unit. If several containers are used with sampling errors, urine should be collected aver equal filling, an equal volume aliquot from each 48 hours in patients who void infrequently «3 container can be pooled. The total effluent vol• times in 24 hours). A 24-hour urine sample can ume must be known and recorded. If the effluent be used for all other patients. The urine collection bags are not freely mixing, then a sample from should be performed on the same day as the dial• each bag is required, as well as the exact volume ysate collection. In children, the urine collection of the container from which the sample was period may be reduced to a minimum of 12 hours. drawn. One cannot extrapolate from one con• Dialysate and urine collection for PD ade• tainer (bag) to the next. quacy studies. Two methods of dialysate sam• No matter what method is used for dialysate, pling predominate. In the first method, for a complete and accurately timed urine collec• CAPO, all effluent bags in a 24-hour period are tion is necessary. The urine volume is more brought to the center. While this' 'batch" method easily managed since it is much smaller than is simple in concept, it is difficult to carry out the dialysate volume. The longer the collection because it means transporting all the dialysate interval, the more reliable are the collections, bags, which are heavy and bulky. assuming patient compliance. A timed collec• The second method is referred to as the "ali• tion of 12 to 48 hours is recommended, de• quot" method. In this approach, each bag of ef• pending on how frequently the patient voids. fluent dialysate is shaken vigorously for a few Polyuric patients, particularly children, or pa• seconds, then is emptied into a measuring con• tients with a short· attention span, may void tainer accurate to an error of <50 mL per frequently enough that a supervised 12-hour 2000 mL. The volume for that bag is recorded collection is accurate. As for any urine collec• in mL and the decimal point is moved three tion, the bladder should be emptied and the places to the left. The resulting figure is the num• urine discarded moments before the start of the ber of mL which must be drawn from the dialy• collection period. Then, moments before the sate effluent in the measuring container and end of the collection period, the patient empties placed in the laboratory red top test tube, pro• the bladder, and this urine, plus all that has vided by the dialysis center. For example, if the been collected in the interval, completes the effluent volume for the CAPO bag is 2450 mL, collection. The patient should try to delay the moving the decimal point three places to the left final voiding until just before the interval ends. means that 2.45 mL of this fluid is put in the test Three or more bladder voidings generally are tube (or other small collection container). Each necessary for urine collections. For patients bag for the 24-hour interval is handled this way. who make little urine and hence void infre- S122 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

quently, 48-hour collections may be more in• APPENDIX D REFERENCES formative. 1. Farrell SC, Bailey MP: Measurement of creatinine in Serum samples. In CAPD, serum concentra• peritoneal dialysis fluid. Ann Clin Biochem 28:624-625, 1991 2. Raja RM, Kramer MS, Rosenbaum JL, Bolisay C, Krug tions of urea and creatinine are relatively con• M: Contrasting changes in solute transport and ultrafiltration stant, and thus blood samples can be drawn at with peritonitis in CAPO patients. Trans Am Soc Artif Intern any convenient time for clearance determina• Organs 27:68-70, 1981 tions. For the asymmetric therapies (NIPD and 3. Krediet RT, Zuyderhoudt FMJ, Boeschoten EW, Arisz L: Alterations in the peritoneal transport of water and solutes CCPD), blood concentrations are lowest at the during peritonitis in continous ambulatory peritoneal dialysis end of the cycling period and highest prior to the patients. Eur J Clin Invest 17:43-52, 1987 next cycling period. Theoretically, the best time 4. Bernardini J: Predicting compliance in home peritoneal to draw these blood samples is half way between dialysis (PO) patients. ASAIO J 42:102, 1996 (abstr) the lowest and highest times. For NIPD patients, 5. Rodby RA, Firanek CA, Cheng YG, Korbet SM: Repro• ducibility of studies of peritoneal dialysis adequacy. Kidney the serum sample should be obtained at the mid• Int 50:267-271, 1996 point of the daytime empty period. For CCPD 6. Johansson A-C, Altman P, Haraldsson B: Creatinine patients, the serum sample should be obtained at generation rate and lean body mass: A critical analysis in the midpoint of the daytime dwell. For most peritoneal dialysis patients. Kidney Int 51:855-859, 1997 7. Smith L, Folden T, Youngblood B, Panlilio F, Gotch NIPD and CCPD patients, this time point conve• F: Clinical evaluation of a peritoneal dialysis kinetic model• niently occurs in the early afternoon. ing set. Adv Perit Dial 12:46-48, 1996

Appendix E: Detailed Rationale for Guideline 9

GUIDELINE 9 by either the DuBois and DuBois method,4 the Gehan and George method,S or the Haycock Estimating Total Body Water and Body method6 using actual body weight. Surface Area (Opinion) For all formulae, Wt is in kg and Ht is in cm: V (total body water) should be estimated by DuBois and DuBois method: 2 either the Watson! or Hume2 method in adults BSA (m ) = 71.84*Wt0425*Hto.725 using actual body weight, and by the Mellits• Gehan and George method: 2 Cheek method3 in children using actual body BSA (m ) = 0.0235*Wt°.5!456*Hto.42246 weight. Haycock method: 2 Watson method!: BSA (m ) = 0.024265* WtO.5378*Ht03964 For Men: V (liters) = 2.447 + 0.3362*Wt (kg) Rationale The practical methods described in + 0.1074*Ht (cm) -0.09516*Age (yrs) the literature to estimate V include a fixed frac• For Women: V = -2.097 + 0.2466*Wt + tion of body weight (0.60 in males and 0.55 in 0.1069*Ht females, or 0.58 in all subjects) and anthropomet• Hume method2: ric formulae based on sex, age, height, and For Men: V = -14.012934 + 0.296785*Wt + weight. !-3 The fixed fraction method is inaccu• 0.192786*Ht rate, as it overestimates total body water even in 7 For Women: V = -35.270121 + 0.183809*Wt overhydrated PD subjects. Therefore, the Work + 0.344547*Ht Group recommends that this method not be used. 2 Mellits-Cheek method for children3: Both the Watson! and Hume formulae were For Boys: V (liters) = -1.927 + 0.465*Wt (kg) derived by comparing anthropometric measure• + 0.045*Ht (cm), when Ht :5132.7 cm ments to measurements of body water by indica• V = -21.993 + 0.406*Wt + 0.209*Ht, when tor dilution techniques. An advantage of these height is ~ 132.7 cm formulae is that they were derived in populations For Girls: V = 0.076 + 0.507*Wt + O.013*Ht, which included obese subjects and, therefore, can when height is :5110.8 cm account for obesity. Estimates from the Watson V = -10.313 + 0.252*Wt + 0.154*Ht, when and Hume formulae are, in general, close to iso• 7 height is ~ 110.8 cm topic body water measurements in PD patients. Body surface area, BSA, should be estimated The error of the estimates based on these formu- APPENDICES S123

lae can increase in subjects with abnormalities inadequate. Therefore, it is recommended in such in body water (hydration), because such subjects individ~als to provide a dose of PD which will were systematically excluded from the studies result in adequate KprtIV urea when they reach their used to derive the formulae. 8 Indeed, both the desired weight without changing the dialysis pre• Watson and Hume formulae tend to consistently scription. For malnourished patients defined by underestimate isotopic body water in both lean SGA or Table 11-3 below, provide a PD dose to and obese PD patients with overhydration.7 The achieve a weekly KtIV urea of 2.0 for the volume formulae are recommended as a reasonable ap• of the patient at desired weight. The calculation proximation with systematic error, but acceptable of the target KprtIV urea in malnourished CAPD based on ease of determination. A proposed cor• subjects equivalent to a weekly KprtIV urea of 2.0 rection of the formulae for edema requires care• at desired weight is as follows: If Vactual is body ful assessment of the dry weight. The correction water obtained from the Watson or Hume formu• considers actual weight at the edematous state as lae using the actual weight and V desired is body the sum of two weights: the dry weight plus the water obtained by the same formulae using the weight gain secondary to edema. V is the body desired weight, then for malnourished subjects water at dry weight obtained from the Watson Vactual

Table 11-3. Median (50th Percentile) Body Weight (kg)11

Age Range (yr) Males Females

18.0-24.9 Frame index <38.4 38.4-41.6 >41.6 <35.2 35.2-38.6 >38.6 Median body weight (kg) 68.3 71.5 74.7 55.1 58.1 62.9 25.0-29.9 Frame index <38.6 38.6-41.8 >41.8 <35.7 35.7-38.7 >38.7 Median body weight (kg) 71.8 75.9 82.2 55.6 58.6 68.7 30.0-34.9 Frame index <38.6 38.6-42.1 >42.1 <35.7 35.7-39.0 >39.0 Median body weight (kg) 74.6 72.8 85.4 57.6 60.7 72.7 35.0-39.9 Frame index <39.1 39.1-42.4 >42.4 <36.2 36.2-39.8 >39.8 Median body weight (kg) 75.9 80.4 84.1 59.5 61.8 76.7 40.0-44.9 Frame index <39.3 39.3-42.5 >42.5 <36.7 36.7-40.2 >40.2 Median body weight (kg) 76.1 79.3 84.9 59.1 62.8 77.1 45.0-49.9 Frame index <39.6 39.6-43.0 >43.0 <37.2 37.2-40.7 >40.7 Median body weight (kg) 76.2 79.8 84.0 60.3 63.4 76.8 50.0-54.9 Frame index <39.9 39.9-43.3 >43.3 <37.2 37.2-41.6 >41.6 Median body weight (kg) 74.7 78.3 83.1 60.3 64.4 77.7 55.0-59.9 Frame index <40.2 40.2-43.8 >43.8 <37.8 37.8-41.9 >41.9 Median body weight (kg) 74.8 77.9 84.5 59.9 66.3 77.6 60.0-64.9 Frame index <40.2 40.2-43.6 >43.6 <38.2 38.2-41.8 >41.8 Median body weight (kg) 73.4 76.3 80.7 60.9 64.5 76.8 65.0-69.9 Frame index <40.2 40.2-43.6 >43.6 <38.2 38.2-41.8 >41.8 Median body weight (kg) 70.3 74.5 78.9 60.2 64.9 74.5 70.0-74.9 Frame index <40.2 40.2-43.6 >43.6 <38.2 38.2-41.8 >41.8 Median body weight (kg) 70.1 72.6 76.7 60.2 62.9 74.5

Obtained from the NHANES Data as reported by Frishanco AR: Anthropometric standards for the assessment of growth and nutritional status. Ann Arbor, MI, University of Michigan Press, 1990. Frame Index has no units and is calculated as follows: [Elbow Breadth (mm)/Height (cm)] x 100. Reprinted with permission from W.B. Saunders Company, American Journal of Kidney Diseases 26:963-981, 1995. ranges based on the input parameters of the pa• a way that restores the relationship between body tient in question. water content and degree of obesity in three Alterations Caused by Amputation. The an• steps: thropometric formulae for total body water calcu• Step A: The fraction of body weight lost to late that as obesity develops and body weight amputation (fw) is obtained from a normograml4 increases, V also increases, but body water con• (see Table 11-4). The hypothetical non-amputated tent (the ratio V/weight) decreases. This is con• weight at the same body composition would be sistent with the known fact that water content of equal to actual weight/(1 - fw). fat tissue is low. Calculations of V in amputees Step B: V at the hypothetical non-amputated by uncorrected anthropometric formulae (using weight (V non-amputated) is calculated from the W at• the actual post-amputation weight and height in son formula. Body water content is Vnon-amputatedl the calculations) distorts the relationship between Weightnon-amputated . V and weight. In this case, body water content Step C: Actual V is calculated by multiplying is not consistent with the degree of obesity.12.13 the actual post-amputation weight by Vnon-amputatectf The anthropometric formulae can be corrected in Weightnon-amputated' This correction makes the as- APPENDICES S125

Table 11-4. Fraction of Weight and BSA approximate surface area if height and weight be known. Corresponding to Amputated Limbs Nutrition 5:303-311, 1989 5. Gel!an E. George SL: Estimation of human body surface Body Part % Loss % of BSA area from height and weight. Cancer Chemother Reports Part Amputated in Weight" to Subtract16 I 54:225-235, 1970 6. Haycock GB, Chir B, Schwartz GJ, Wisotsky DH: Geo• Foot 1.8 3.5 metric method for measuring body surface area: A height• Leg below knee 6.5 10.0 weight formula validated in infants, children, and adults. J Leg above knee 8.0 12.5 Pediatr 93:62-66, 1978 Leg at hip 18.5 18.0 7. Wong KC, Xiong OW, Kerr PG, Borovnicar DJ, Stroud Hand 0.8 2.5 DB, Atkins RC, Strauss BJG: KtIV in CAPD by different estimations of V. Kidney Int 48:563-569, 1995 Arm at elbow 3.1 6.0 8. Tzamaloukas AH: Effect of edema on urea kinetic stud• Arm at shoulder 6.6 10.0 ies in peritoneal dialysis patients. Perit Dial Int 14:398-401, 1994 9. Tzamaloukas AH, Dombros NV, Murata GH, Nicolo• sumption that amputation per se will not change poulou N, Dimitriadis A, Kakavas J, Malhotra D, Antoniou S, Balaskas EV, Voudiklari S: Fractional clearance estimates body water content. 13 using two anthropometric formulae in continuous peritoneal The calculations for body surface area (BSA) dialysis: Gender, height and body composition differences. in amputees should be also corrected, because of Perit Dial Int 16:135-141, 1996 inconsistent results obtained with the uncorrected 10. Jones MR: Etiology of severe malnutrition: Results of 15 an international cross-sectional study in continuous ambula• calculation of BSA. The correction requires tory peritoneal dialysis patients. Am J Kidney Dis 23:412- also three steps: 420, 1994 Step A: Same as step A in the correction of II. Kopple JD, Jones MR, Keshaviah PR, Bergstrom J, Lind• V in amputees. say RM, Moran J, Nolph KD, Teehan BP: A proposed glossary for dialysis kinetics. Am J Kidney Dis 26:963-981, 1995 Step B: BSA at the hypothetical non-ampu• 12. Tzamaloukas AH, Saddler MS, Murphy G, Morgan tated weight is calculated by one of the three K, Goldman RS, Murata GH, Malhotra D: Volume of distri• BSA formulae above. bution and fractional clearance of urea in amputees on contin• Step C: The fraction of BSA corresponding to uous ambulatory peritoneal dialysis. Perit Dial Int 14:356- 361, 1994 the amputated limb(s) (fBsA ) is obtained from 13. Tzamaloukas AH, Murata GH: Estimating urea vol• Table 11-4 derived from Herndon. Corrected BSA ume in amputees on peritoneal dialysis by modified anthropo• metric formulas. Adv Perit Dial 12:143-146, 1996 is BSA Non-amputated/(l - fBSA ).16 14. Hopkins B: Assessment of nutritional status. Nutri• tional Support Dietetics Corte Curriculum (ed 2). Silver APPENDIX E REFERENCES Spring, MD, American Society for Parenteral Nutrition, 1993, pp 15-70 I. Watson PE, Watson ID, Batt RD: Total body water vol• 15. Tzamaloukas AH, Malhotra D: Creatinine clearance umes for adult males and females estimated from simple anthro• in amputees on CPD. Perit Dial Int 16:426 1996 pometric measurements. Am J Clin Nutr 33:27-39, 1980 16. Anonymous: Pre-hospital management, transportation 2. Hume R, Weyers E: Relationship between total body and emergency care, in Herndon DN (ed): Total Burn Care. water and surface area in normal and obese subjects. J Clin Philadelphia, PA, Saunders, 1996, pp 36-38 Pathol 24:234-238, 1971 17. U.S.Renal Data Systems: The USRDS dialysis mor• 3. Mellits ED, Cheek DB: The assessment of body water bidity and mortality study (wave I), in National Institutes of and fatness from infancy to adulthood. Monographs Soc Res Health, National Institute of Diabetes and Digestive and Kid• Child Dev, Serial 140. 35:12-26,1970 ney Diseases (ed): U.S. Renal Data Systems 1997 Annual 4. Du Bois D, Du Bois EF: A formula to estimate the Data Report, chapter 4. Bethesda, MD, 1997, pp 49-67

Appendix F: Detailed Rationale for Guideline 12

GUIDELINE 12 (SGA). For pediatric PO patients, nutritional sta• tus should be assessed using the PNA and other Assessment of Nutritional Status (Opinion) standard nutritional assessments (see Guideline Nutritional status of adult PO patients should 14: Use of the Modified Borah Equation to As• be assessed on an ongoing basis in association sess Nutritional Status of Pediatric PO Patients). with Kt/V urea and CCr measurements using the Rationale Although nutritional status de• Protein equivalent of Nitrogen Appearance pends on many nondialysis-related factors, appe• (PNA) and Subjective Global Assessment tite suppression, nausea, and vomiting are major S126 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

clinical features of inadequate dialysis. There• relates directly to the contribution of protein ca• fore, nutritional status is also an important mea• tabolism to uremic toxicity. sure of PD adequacy. Of the available measures In patients on hemodialysis, nitrogen balance of nutrition, PNA is recommended because it studies have been performed to estimate total ni• provides an estimate of protein catabolic rate trogen output or appearance (skin losses were (PCR) and other protein losses. The SGA is rec• estimated and breath losses were ignored).2 The ommended because it is a clinical assessment of relationship of urea nitrogen appearance to total patient nutritional status and is strongly associ• nitrogen output was assumed to be fixed and a ated with patient survival. Both measures are dis• formula was developed, known as the Borah cussed in detail below. equation, to calculate the PCR directly from urea Protein Equivalent of Nitrogen Appearance nitrogen appearance2: (PNA). PNA is a useful tool for monitoring the absolute level of changes in dietary protein in• PCR (g/d) = 6.49*UNA + 0.294*V (1) take. Nitrogen intake is almost entirely (95%) where UNA represents the net production or ap• in the form of protein. Therefore, total nitrogen pearance of urea nitrogen in body fluids (any excretion in stable humans multiplied by 6.25 increase in body fluid nitrogen concentration (there are approximately 6.25 grams of protein times body water volume) and all measurable per gram of nitrogen) should be a good estimate outputs in g/d; V is the volume of distribution of protein intake. 1 This relationship does not hold of urea in liters. In hemodialysis patients with true for: individuals in a state of catabolism no direct protein losses in dialysate or urine, this where body cell mass may be contributing to PCR also represents an estimate of dietary pro• nitrogen excretion; conditions of anabolism tein intake and is the protein equivalent of total where the opposite occurs; and inconsistencies nitrogen appearance (PNA). in absolute or time-averaged blood concentra• In dialysis patients with substantial urinary or tions of BUN or creatinine. dialytic protein losses (>0.1 g/kg), the direct pro• In normal humans and in dialysis patients in tein losses must be added to the PCR to yield nitrogen balance who have no direct protein the true PNA as an estimate of dietary protein losses in urine, dialysate, or feces, the total daily 3 intake. ,4 Thus, in PD: excretion of nitrogen in urine, dialysate, feces, breath, and skin losses is in the form of low PNA = PCR + protein losses (2) molecular weight nitrogenous metabolites (such as urea, creatinine, urate, amino acids, ammonia, Nitrogen balance studies have also been per• and peptides ).1 formed in PD patients; the measured total nitro• The excretion of nitrogen as low molecular gen output (appearance) included estimates of weight metabolites multiplied by 6.25 approxi• skin and fecal nitrogen losses plus measurement of all nitrogen (including protein nitrogen) in di• mates the amount of nitrogen in ingested protein. 4 This calculation has been termed the protein cata• alysate and urine. ,5 The average daily dialysate bolic rate (PCR).1,2 PCR actually represents the protein loss in the CAPD patients was 7.3 g. net amount of protein catabolism exceeding pro• Urine protein losses were < 1 g/24 hrs. A formula tein synthesis required to generate an amount of for the calculation of PNA from UNA was devel• nitrogen equal to that excreted. The nitrogen in oped: ingested protein enters the body nitrogen pools; PNA (g/d) = 1O.76*(0.69*UNA + 1.46) (3) nitrogen excreted in urine, feces, breath, skin, and dialysate represents the metabolism of a vari• This calculation incorporates the average dial• ety of body substances in these pools, such as ysate protein loss of 7.3 g/d. creatine and purine, in addition to body proteins. Calculations ofPNA in PD patients with Equa• Thus, although PCR is a reasonable estimate of tions 2 and 3 have been shown to yield nearly 3 5 protein intake, not all excreted nitrogen comes identical results. - Also, subtracting protein directly from protein. The protein catabolic losses in dialysate from Equation 3 yields values equivalent of nitrogen excretion is actually a net nearly identical to the PCR calculated by Equa• catabolic equivalent, rather than an absolute. It tion 1, as developed in HD patients.4 APPENDICES S127

If daily peritoneal dialysate protein losses ex• patient-to-patient comparisons and to follow ceed 15 g, PNA calculated from Equation 2 will PNA !lleasurements serially in an individual pa• exceed PNA calculated from Equation 3 by ap• tient whose weight may change. If weight is sta• proximately 0.1 g/kg standard body weight.5 ble, normalization is less important in serial mea• High transporters lose more protein into effluent surements for an individual patient (see 6 dialysate than other PO patients. ,7 Therefore, in Guideline 14 for PNA discussion on pediatric high transport patients it is best to measure pro• patients). tein losses in dialysate directly, and if dialysate Subjective Global Assessment (SGA). The protein losses exceed 15 gld (found in < 10% SGA is a valid estimate of nutritional status for of peritonitis-free patients), calculate PNA from patients treated with PO.13 Furthermore, it is as• Equation 2.5 For patients who lose large amounts sociated with the probability of patient survival. I4 of protein from any nonperitoneal source (e.g., The SGA was developed as a clinical estimate nephrotic syndrome), Equation 2 should be used. of pre-operative nutritional status. 15 For two phy• Equations 2 and 3 have been validated with sicians, the inter-observer agreement was 72% nitrogen balance studies only in CAPO and not in greater than would have been predicted by other therapies such as NIPO?-5 However, since chance alone. Validity was based on correlations CAPO and HO patients have similar PCR values with three measures of post-operative hospital (PNA - protein losses) at any given UNA, the morbidity (incidence of infection, use of antibiot• intermittent nature of NIPO would seem unlikely ics and length of stay).15 A detailed description to alter the relationships. Furthermore, the daily of the SGA was provided by Oetsky in 1987. 16 protein losses on NIPO are similar to those of The SGA was originally developed as a clini• CAPO.8 A recent study of seven patients suggests cal assessment of pre-operative nutritional status that the PNA formula of Bergstrom et al9 may be for patients prior to gastrointestinal surgery.15,16 superior. 10 However, this must be corroborated in When applied to CAPO patients,13 validity test• a study with a larger number of patients. Berg• ing reduced the number of items to four (weight strom supports the use of Equation 3.1 loss, anorexia, loss of subcutaneous tissue, and In summary, the most accurate determination muscle mass). To increase the ability of the SGA of PNA in patients undergoing PO uses Equation to detect a change in nutritional status, the scor• 2, but this requires measurement of UNA and ing scale was increased from a 3-point to a 7- dialysate protein losses. Equation 3 is a suitable point scale. Ouring the development phase, the substitute and requires only measurement of SGA was determined by physicians, research UNA. However, if dialysate protein exceeds 15 nurses, and nurse cliniciansl6 but was determined gld (many high transporters may fall into this by dialysis nurses and dietitians when used in category) and in all pediatric patients, Equation the CANUSA study.14 2 is preferred. The SGA, as modified for use in CAPO pa• Methods of normalizing PNA are still under tients,13 uses a 7-point scalel4 which any health• debate. The Work Group recommends normal• care professional can apply following a short ization by standard weight, which has been ap• training period. plied extensively. Standard weight is equal to VI The four items used to assess nutritional status 0.58. 11 PNA normalized by either standard in CAPO patients are: weight change, anorexia, weight or actual weight tends to be high in mal• subcutaneous tissue, and muscle mass. nourished, underweight PO subjects. 12 Normal• Weight change is addressed by the question, ization ofPNA to fat-free, edema-free body mass "What was the patient's weight change over the provides appropriately low nPNA values in un• past 6 months?" Ideally, this should be documented derweight individuals. 13 The Work Group recom• by the actual weights, but historical information mends that fat-free, edema-free body mass, esti• from the patient is acceptable. A loss of > 10% is mated from creatinine kinetics (see Section II: severe, 5% to 10% is moderate while 5% is mild. Measures of PO Oose) should be used, in addi• This is rated subjectively on a scale from 1 to 7, tion to standard weight, to normalize PNA in where 1 or 2 is severe malnutrition, 3 to 5 is moder• underweight PO subjects, defined by Table x• ate to mild malnutrition and 6 or 7 is mild malnutri• l in Appendix E. Normalizing is important for tion to normal nutritional status. If the weight change S128 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

was intentional, the weight loss would be given less orexia, loss of subcutaneous tissue, and loss of subjective weight while edema might obscure muscle mass (muscle wasting).13 Evidence for greater weight loss. validity was provided by the correlations with Anorexia is addressed by the question, "Has serum albumin concentration, bioelectrical im• the patient's dietary intake changed?" Supple• pedance, anthropometric measurements and nor• mental questions determine whether a decrease malized protein catabolic rate. in dietary intake is by prescription or due to de• Using the SGA as originally described,16 59% creased appetite. Nausea and vomiting are ad• of prevalent CAPO patients were well-nour• verse factors for this item. Again, the interviewer ished. 17 Mild and severe malnutrition was re• will rate intake on the 7-point scale with higher ported in 33% and 8% of the patients, respec• scores indicative of better dietary intake, better tively. In 263 hemodialysis patients and 224 appetite, and the absence of nausea and vomiting. CAPO patients, 18 the SGA covaried with low vis• Subcutaneous tissue (fat and muscle wasting) ceral (i.e., serum) protein concentrations, midarm can be examined in many areas. A very detailed muscle circumference (somatic protein mass), and illustrated brochure is available from Baxter and body fat stores. Healthcare (publication #BRU-008-312-2000). In the CANUSA study of peritoneal dialysis, 14 Although the history-taking format is more de• weight loss, anorexia, loss of subcutaneous tis• tailed than required, the description of how to sue, and loss of muscle mass (muscle wasting), as determine muscle wasting and subcutaneous tis• identified above as being statistically associated sue is excellent. with the SGA,13 were used to generate the SGA Subcutaneous fat can be assessed by examin• for the CANUSA study. To make the scale more ing the fat pads directly below the eyes and by discriminative, the 3-point scale was expanded gently pinching the skin above the triceps and biceps. The fat pads should appear as a slight to a 7-point scale, with I and 2 corresponding to bulge in a normally nourished person but are severe malnutrition, 3-5 corresponding to mild "hollow" in a malnourished person. When the to moderate malnutrition and 6-7 corresponding skin above the triceps and biceps is gently to mild malnutrition to normal nutritional status. pinched, the thickness of the fold between the In a multivariate analysis, a higher SGA was examiner's fingers is indicative of the nutritional associated with a lower relative risk of death. status. The examiner then scores the observations A one unit increase on the 7-point scale was on a 7-point scale. associated with a 25% decline in the relative risk Muscle mass and wasting can be assessed by of death (relative risk 0.75). Ouring the first 6 examining the temporalis muscle, the promi• months of dialysis, the mean SGA increased 0.72 nence of the clavicles, the contour of the shoul• units. There was a statistically significant correla• ders (rounded indicates well-nourished; squared tion between the increment in adequacy due to indicates malnutrition), visibility of the scapula, the addition of peritoneal clearance (KtIV urea and the visibility of the ribs, and interosseous muscle CCr) to RRF. Over the next 12 months, there was mass between the thumb and forefinger, and the a small decrease in SGA and this correlated quadriceps muscle mass. These are scored on a with loss of RRF estimated by CCo but not with 7-point scale. KtIV urea' 19 The four-item scores are then aggregated into The SGA has not been validated as a means a global score. The global score is not a simple of nutritional assessment in the pediatric PO pop• arithmetic average of the four items. The exam• ulation. Thus, the SGA cannot be recommended iner can apply different weights to the items. for use in children. For example, if the physical examination items clearly indicate severe malnutrition, but the pa• APPENDIX F REFERENCES tient indicates only a moderate decrease in weight and a good appetite, the examiner might 1. Kopple JD, Jones MR, Keshaviah PR, Bergstrom J, weight the physical examination items higher Lindsay RM, Moran J, Nolph KD, Teehan BP: A proposed glossary for dialysis kinetics. Am J Kidney Dis 26:963-981, than the historical items. 1995 In 23 CAPO patients, four items were statisti• 2. Borah MF, Schoenfeld PY, Gotch FA, Sargent JA, cally associated with the SGA: weight loss, an- Wolfson M, Humphreys MH: Nitrogen balance during inter- APPENDICES S129

mittent dialysis therapy of uremia. Kidney Int 14:491-500, 12. Harty JC, Boulton H, Curwell J, Heelis N, Uttley L, 1978 Venning MC, Gokal R: The normalized protein catabolic rate 3. Keshaviah PR, Nolph KD: Protein catabolic rate calcula• is a fla~ed marker of nutrition in CAPD patients. Kidney Int tions in CAPD patients. ASAIO Trans 37:M400-M402, 1991 45:103-109, 1994 4. Randerson DH, Chapman GV, Farrell PC: Amino acid and 13. Enia G, Sicuso C, Alati G, Zoccali C: Subjective dietary status in CAPD patients, in Atkins RC, Farrell PC, global assessment of nutrition in dialysis patients. Nephrol Thompson N (eds): Peritoneal dialysis, chapter 18. Edinburgh, Dial Transplant 8:1094-1098, 1993 United Kingdom, Churchill Livingston, 1981, pp 179-191 14. Churchill DN, Taylor DW, Keshaviah PR: Adequacy of dialysis and nutrition in continuous peritoneal dialysis: 5. Usha K, Moore H, Nolph KD: Protein catabolic rate in Association with clinical outcomes. J Am Soc NephroI7:198- CAPD patients: Comparison of different techniques. Adv 207, 1996 Perit Dial 12:284-287, 1996 15. Baker JP, Detsky AS, Wesson DE, Wolman SL, Stew• 6. Malhotra D, Tzamaloukas AH, Murata GH, Fox L, art S, Whitewell J, Langer B, Jeejeebhoy K: Nutritional as• Goldman RS, Avasthi PS: Serum albumin in continuous peri• sessment. A comparison of clinical judgement and objective toneal dialysis: Its predictors and relationship to urea clear• measurements. N Engl J Med 306:969-972, 1982 ance. Kidney Int 50:1501-1507, 1996 16. Detsky AS, McLaughlin JR, Baker JP, Johnston N, 7. Nolph KD, Moore HL, Prowant B, Twardowski ZJ, Whittaker S, Mendelson RA, Jeejeebhoy KN: What is subjec• Khanna R, Gamboa S, Keshaviah P: Continuous ambulatory tive global assessment of nutritional status? J Parenter Enteral peritoneal dialysis with a high flux membrane. ASAIO J 904- Nutr II :8-13, 1987 909, 1993 17. Young GA, Kopple JD, Lindholm B, Vonesh EF, De• 8. Kathuria P, Goel S, Moore HL, Khanna R, Twardowski Vecchi A, Scalamogna A, Castelnovo C, Oreopoulos DG, ZJ, Nolph KD: Protein losses during CAPD. Perit Dial Int Anderson GH, Bergstrom J, Dichiro J, Gentile D, Nissenson 16:S9, 1996 (abstr; suppl 2) A, Sakhrani L, Brownjohn A, Nolph KD, Prowant BF, Al• 9. Bergstrom J, Furst P, Alverstrand A, Lindholm B~o­ grim CE, Martis L, Serkes KD: Nutritional assessment of tein and energy intake, nitrogen balance and nitrogen losses continuous ambulatory peritoneal dialysis patients: An inter• in patients treated with continuous ambulatory peritoneal di• national study. Am J Kidney Dis 17:462-471, 1991 alysis. Kidney Int 44:1048-1057, 1993 18. Cianciaruso B, Brunori G, Kopple JD, Traverso G, Panarello G, Enia G, Strippoli P, De Vecchi A, Querques M, 10. Mandolfo S, Zucchi A, D'Oro Le, Corradi B, Imbasciati Viglino G, Vonesh E, Maiorca R: Cross-sectional comparison E: Protein nitrogen appearance in CAPD patients: What is the of malnutrition in continuous ambulatory peritoneal dialysis best formula? Nephrol Dial Transplant 11:1592-1596, 1996 and hemodialysis patients. Am J Kidney Dis 26:475-486, II. Nolph KD, Moore HL, Prowant B, Meyer M, Twar• 1995 dowski ZJ, Khanna R, Ponferrada L, Keshaviah P: Cross 19. McCusker FM, Teehan BP, Thorpe K, Keshaviah P, sectional assessment of weekly urea and creatinine clearances Churchill D: How much peritoneal dialysis is required for and indices of nutrition in continuous ambulatory peritoneal the maintenance of a good nutritional state? Kidney Int dialysis patients. Perit Dial Int 13: 178-183, 1993 50:S56-S61, 1996

Appendix G: Detailed Rationale For Guideline 15

GUIDELINE 15 clearance of 84 L. For a patient with a total body water of 42 L, this corresponds to a weekly Kt! Weekly Dose of CAPD (Evidence) Vurea of 2.0. Others, using the concept of the Dial• For CAPD, the delivered PD dose should be ysis Index, suggested that a similar patient would a total KtIV urea of at least 2.0 per week and a require 13.5 L daily of equilibrated drained dialy• total creatinine clearance (CCr) of at least 60 LI sate to maintain nitrogen balance, and this would 2 2 week/1.73 m • produce a weekly KtlVurea of 2.25. The differ• Rationale The evidence supporting this guide• ence between these two projections is due to the line is derived from theoretical constructs and higher target protein intake used in the latter cal• cohort studies which use either univariate or mul• culation. Using the peak urea concentration hy• tivariate statistical analyses. pothesis, a weekly KtIV urea of 2.0 is equivalent The original description of CAPD! suggested to a single pool hemodialysis KtlVurea of 1.3 for that an anephric 70-kg patient with a total body patients receiving thrice weekly dialysis.3 These water of 42 L would remain in nitrogen balance theoretical constructs suggest that a weekly Kt! with a daily dialysis prescription of 10 L given V urea of 2.0 to 2.25 would be appropriate. as five 2-L exchanges. Full equilibration of urea Validation of these theoretical constructs re• between plasma and dialysate and 2 Lid of net quires clinical study. A series of cohort studies 4 8 ultrafiltration were assumed. This would produce addressed this issue. - Initially, no relationship a daily urea clearance of 12 L or a weekly urea was found between urea clearance and patient S130 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

survival,6 but a re-analysis, using an anthropo• 7% increase in the risk of death. The risk of metric9 estimate for total body water, found that technique failure increased with decreased creati• patients with a weekly KtIV urea < 1.5 had an in• nine clearance, but was not associated with Kt! creased risk of death compared to patients with a V urea' Hospitalization increased with decreased weekly KtlVurea ~1.5. In another study,1O a mean CCr. Using data derived from the multivariate weekly KtIV urea> 1.89 was associated with a de• analysis, the predicted 2-year survival associated creased risk of death compared to patients with with a constant weekly KtlVurea of 2.1 was 78%. less dialysis, while yet another4 reported that pa• The corresponding weekly CCr was 70 Ll1.73 m2. tients surviving for a 12-month follow-up had a Thus, there is both a theoretical rationale and mean weekly KtlVurea of 2.0 compared to a mean convincing evidence supporting an association of 1.7 among those who did not survive for 12 between greater clearance of urea and creatinine months. A Belgian group reported that 16 pa• and better patient survival. There is also evidence tients surviving 5 years on CAPD had a mean supporting an association between greater CCr to weekly KtlVurea of 2.0. 5 These studies all used longer technique survival and less hospitaliza• l 3 univariate analysis and therefore did not simulta• tion. In summary, theoretical constructs - sug• neously evaluate the association between other gest that the minimum weekly KtIV urea should important variables (eg, age, diabetes, cardiovas• be 2.0. Cohort studies using univariate statistical cular disease) and patient survival. analysis support this "target. ,,4-8,11,12 The CA• Several studies have used multivariate statisti• NUSA study predicts, among North American cal analysis to evaluate the association between patients, a 78% two-year survival with a weekly adequacy of PD and survival while controlling KtIV urea of 2.1.12 7 811 for other variables. • . ,12 In one such study a There are no theoretical data to support a spe• lower serum albumin concentration, increased cific CCr target. The CCr which corresponds to a age, greater time on dialysis, and lower weekly weekly KtIV urea of 2.1 in the CANUSA study KtIV urea were associated with a decreased proba• was 70 Ll1.73 m2/wk. The Italian group found bility of patient survival. 11 A French group re• that a weekly KtIV urea of 1.96 corresponded to a ported that patients with a weekly KtIV urea > 1.7 weekly CCr of 58 L. The CANUSA study in• and a weekly CCr of >50 Ll1.73 m2 at initiation volved incident patients with significant RRF, of dialysis had better survival than those with while the Italian study evaluated prevalent pa• lower values at initiation.7 However, these inves• tients with much less RRF. 8 The target of 60 LI tigators did not evaluate the effect of changes in 1.73 m 2/wk was selected by the Work Group adequacy over time due to loss of RRF, nor did because it is more relevant to patients with dimin• they attempt to evaluate any association of higher ished renal function. weekly KtIV urea or CCr with survival. An Italian There are few data to address the issue of ade• group evaluated the association between esti• quate compared to optimal dialysis. The latter is mates of adequacy and patient survival in a co• defined in part as the dialysis dose above which hort of 68 prevalent continuous PD patients fol• the incremental clinical benefit is not justified by lowed over 3 years.8 A mean weekly KtIV urea of the social cost to the patient or the financial cost 1.96 was associated with better survival than to society. Whether or not increased weekly Kt! lower values. No further benefit was observed V urea greater than 2.0 will be associated with im• with a KtlVurea higher than 1.96. Among these proved clinical outcomes requires further study. patients, a weekly KtIV urea of 1.96 corresponded The relative importance of RRF compared to to a weekly CCr of 58 liters/1.73 m2. peritoneal clearance and the relative importance The CANUSA study evaluated the association of urea compared to CCr are important and inter• between adequacy of PD and patient survival, related issues. The convention has been to con• technique survival, and hospitalization among sider RRF and peritoneal clearance to be equiva• 680 incident patients (new to starting PD) treated lent and therefore additive. Some believe that with continuous PD.12 A decrease of 0.1 in renal clearance is more important, but in the ab• weekly KtlVurea was associated with a 5% in• sence of data establishing the magnitude of that crease in the relative risk of death, and a decrease difference, the assumption of equivalence was of 5 Ll1.73 m 2/wk in CCr was associated with a adopted by the Work Group. CCr appeared more APPENDICES S131

important than urea clearance in the CANUSA predictive of well-being, morbidity, or mortality. study.12 The former was associated with patient There.are no data regarding the real protein needs survival, technique survival, and hospitalization, of children, especially young children, on dial• while the latter was associated only with patient ysis. It is the opinion ofthe Work Group that the survival. One potential explanation for this find• nutritional requirements per kilogram of body ing is that CCr is more strongly associated with weight are higher in children than in adults. better RRF than was KtIV urea' This explanation Therefore, PD doses in children, and especially is based on the assumption that RRF is better small infants who have very high protein intakes, than peritoneal clearance, an opinion not yet sup• may have to be higher than PD doses in adults. ported by evidence. Until evidence to the contrary is available, the APPENDIX G REFERENCES Work Group recommends that renal and perito• 1. Popovich RP, Moncrief JW: Kinetic modeling of perito• neal clearances be considered equivalent. Ifthere neal transport. Contrib Nephrol 17:59-72, 1979 2. Teehan BP, Schleifer CR, Sigler MH, Gilgore GS: A is discordance between achieving the target Kti quantitative approach to the CAPD prescription. Perit Dial V urea and CCo the KtIV urea should be the immedi• Bull 5:152-156, 1985 ate determinant of adequacy since it reflects pro• 3. Keshaviah PR, Nolph KD, Prowant B, Moore H, Pon• tein catabolism. However, the reason for the dis• ferrada L, Van Stone J, Twardowski ZJ, Khanna R: Defining crepancy should be sought and the patient adequacy of CAPD with urea kinetics. Adv Perit Dial 6: 173- 177, 1990 monitored closely for clinical signs of under

GUIDELINE 19 CAPD patients self-report full compliance with 82.8% of their exchanges? One exchange per Identify and Correct Patient-Related Failure week is missed by 11.5% of patients and 2 to 3 to Achieve Prescribed PD Dose (Opinion) exchanges per week are missed by 4.5% of pa• Potential patient-related causes of failure to tients, all self-reported. Other estimates vary be• achieve prescribed peritoneal dialysis dose should tween 5% and 38%.4 Thus, noncompliance is a be investigated and corrected. These inch,lde: major cause of a delivered PD dose being less than • Failure to comply with the prescription the desired dose and is potentially preventable . • Lack of understanding of the importance of Lack of understanding of the importance of ad• adherence to the full prescription herence to the full prescription. Medical litera• • Sampling and collection errors ture about conditions associated with noncompli• Rationale It is the opinion of the Work Group ance in PD is inadequate. The Work Group that to increase the likelihood of achieving a pre• reviewed published information on compliance in 5 8 scribed dose of PD, it is necessary to elucidate hemodialysis - and in drug treatment for chronic 9 ll the patient-related causes of failure to achieve a illness. - The Work Group believes that some of prescribed dose. Selection of inappropriate can• the conclusions in this literature may be applicable didates for PD may result in failure to achieve a to PD. An important conclusion of the studies on prescribed dose due to medical, technical, and/ drug compliance is that lack of education regard• ing the importance of adherence to the full pre• or psycho-social reasons. The issue of medically scription partially contributes to compliance fail• appropriate patient selection is dealt with at ure. Compliance with the drug prescription length in Section VIII, Suitable Patients for PD. improves when the patient is convinced that the In addition to the medical reasons for selecting diagnosis is accurate, the reasons for the pre• patients for PD or HD discussed in Section VIII, scribed treatment are correct, and the prescribed patient compliance is of paramount importance treatment is beneficial. 11 Some contend that pa• and should be explored. tients on dialysis are more likely to follow the Failure to Comply With the Prescription. Pa• prescribed treatment if they can be convinced that tients may decrease the delivered dose of PD in adherence to the prescription is in their own inter• several ways. Some of the ways are listed below: est. 8 By not understanding the significance, impor• • Skipping exchanges tance, value, or relevance of the collections or the • Shortened exchange times exchanges, noncompliance could occur without • Dialysate dumping: too much flushing re• patient concern. Therefore, proper education about sulting in too little filll the treatment may increase compliance in many • Delayed dumping: This is achieved by par- PD patients. Patients should be educated that dial• tially draining before the dwell is completed. ysis prescription may change over time (different • Reduction of total cycler time modality and/or increase in the number or volume • Unscheduled dry days on CCPD of exchanges) due to loss of residual renal function A validated method to measure patient compli• (see Guideline 6: Assessing Residual Renal Func• ance is not currently available. Methods pro• tion). The method of'education should emphasize posed for evaluating compliance include moni• the expected positive results (improved survival, toring for variations in creatinine output in well-being) of adherence to the PD prescription, dialysate and urine2 as detailed in Guideline 7: rather than the negative outcomes (morbidity, PD Dose Troubleshooting. Evidence for this rec• mortality) of non-adherence, to prevent the devel• ommendation is currently not available. The rec• opment of excessive anxiety, which has adverse ommendation represents, therefore, the opinion effects on compliance. of the Work Group members. Patient education should be continuous In the absence of a validated method to measure throughout the course of PD. Patients should be patient compliance, its prevalence in the PD popu• told the results of the repeated clearance measure• lation is not known. Preliminary data from the ments and should be aware of the target values USRDS DMMS Wave II project show that 487 for KprtIV and CCr and of the clinical significance APPENDICES S133

of these clearances. Prevention of non-compliance liter bags are being used and only 2.5 liters are should include monitoring the patient's psycho• exch~ged. logical status. In studies on compliance to drugs, • Incomplete urine collections for the RRF de• certain psychiatric conditions, such as hostility to• termination. ward authority, depression and memory impair• Many of these errors can be prevented by careful ment, financial problems, impaired mobility, and patient instruction about the details and signifi• language or ethnic barriers, have been associated cance of the clearance procedure. with poor compliance. I I In addition, complexity of APPENDIX H REFERENCES 9 the prescription and chronicity of the treatmene I 1. Caruana RJ, Smith KL, Hess CP, Perez JC, Cheek PL: increased noncompliance. In the case of prolonged Dialysate dumping: a novel cause of inadequate dialysis in treatment, repetition of the teaching at 6-month continuous ambulatory peritoneal dialysis (CAPD) patients. Perit Dial Int 9:319-320, 1989 intervals improved compliance. 11 In studies on l2 2. Keen M, Lipps B, Gotch F: The measured creatinine compliance in RD, male gender and young age13 generation rate in CAPD suggests only 78% of prescribed were predictors of poor compliance with different dialysis is delivered. Adv Perit Dial 9:73-75, 1993 aspects of HD prescription. Preliminary informa• 3. U.S. Renal Data Systems: The USRDS dialysis mor• tion suggests that a general negative attitude ofthe bidity and mortality study (wave I), in National Institutes of Health, National Institute of Diabetes and Digestive and patients predicts noncompliance in PD.14 Finally, Kidney Diseases (ed): U.S. Renal Data Systems 1997 An• drug compliance improves with better education nual Data Report, chapter 4. Bethesda, MD, 1997, pp 49- of the providers about compliance issues.1O The 67 Work Group thinks that all of these issues-are 4. Amici G, Viglino G, Virga G, Gandolfo C, Da Rin G, Bocci C, Cavalli PL: Compliance study in peritoneal relevant to PD. The psychological profile which dialysis using PD Adequest software. Perit Dial Int is predictive of noncompliance and the best 16:S176-S178, 1996 (suppl 1) method of characterizing this profile should be a 5. Wolcott DL, Maida CA, Diamond R, Nissenson AR: subject for research in the future. For the present, Treatment of compliance in end-stage renal disease. Am J Nephrol 6:329-338, 1986 the Work Group's opinion is that monitoring of 6. King K: Noncompliance in the chronic dialysis popu• patients' psychological status should be aimed at lation. Dial Transplant 20:67-68, 1991 detecting conditions associated with increased risk 7. Rocco MV, Burkart J: Prevalence of missed treat• of noncompliance, and particularly at detecting a ments and early sign-offs in hemodialysis patients. J Am Soc NephroI4:1178-1183, 1993 negative patient attitude towards PD. Teaching 8. Lundin AP: Causes of noncompliance in dialysis pa• patients about their PD prescription should be re• tients. Dial Transplant 24:174-176, 1995 peated at intervals of 6 months or less. 9. Porter AM: Drug defaulting in a dialysis patients. Br Sampling and collection errors. Sampling Med J 1:218-222, 1969 10. Innui TS, Yourtee GL, Williamson JW: Improved and collection errors committed by patients dur• outcomes in hypertension after physician tutorials. A con• ing the clearance study preclude accurate mea• trolled trial. Ann Intern Med 84:646-651, 1976 surement of clearance. Such errors include: 11. Anderson RJ, Kirk LM: Methods for improving pa• • Batch method: the patient may not recognize tient compliance in chronic disease states. Arch Intern Med 142:1673-1675,1982 the importance of accidentally spilled dialysate. 12. Everett KD, Sletten C, Carmach C, Brantley PJ, Jones • Aliquot method: Inaccurate weighing of GN, McKnight LT: Predicting noncompliance to fluid restrictions drain volumes which may be the result of inaccu• in hemodialysis patients. Dial Transplant 22:614-622, 1993 rate scales or misreading. Disproportionate filling 13. Cummings KM, Becker MH, Kirscht JP, Levin NW: Psy• of the syringe with dialysate. chosocial factors affecting adherence to medical regimens in a group of hemodialysis patients. Med Care 20:567-580, 1982 • Errors in weighing bags for variable fill vol• 14. Bernardini J: Predicting compliance in home perito• umes: for example, when (due to cost issues) 3- neal dialysis (PD) patients. ASAIO J 42:102, 1996 (abstr)

IMPORTANT NOTICE

Articles which were selected for structured review will be listed separately in a specially designed NKF-DOQI Program Caddy which will contain Executive Summaries in addition to the Clinical Practice Guidelines. The articles and guidelines will also be available on the W orId Wide Web on NKF's cyberNephrology web site at www.kidney.org. XI. Biographical Sketches of the NKF -DOQI Peritoneal Dialysis Adequacy Work Group Members

The following are brief sketches that describe miology and Biostatistics, with a focus on evi• the professional training and experience, as well dence-based medicine. Dr. Churchill reported an as principal business affiliations of the Work affiliation with Althin Med Inc. Group members. All Work Group members com• John Burkart, MD, is Associate Professor of pleted a disclosure statement certifying that any Internal Medicine/Nephrology and Director of potential conflict of interest would not influence Outpatient Dialysis Services (CAPD and HD) for their judgment or actions concerning the NKF• the Section of Nephrology at Wake Forest Uni• DOQI. versitylBowman Gray School of Medicine in Thomas A. Golper, MD, FACP (Work Group Winston-Salem, North Carolina. Dr. Burkart has Chair), is Medical Director, Director of Clinical been an active clinical nephrologist for 11 years Research, and Professor of Medicine for the Di• and manages a large nephrology and dialysis vision of Nephrology at the University of Arkan• practice. He currently cares for 70 end-stage re• sas for Medical Sciences. Dr. Golper has been nal disease patients and participates in a group working to improve patient outcomes for the past practice serving more than 250 dialysis patients. 23 years. He serves as a CQI advisor to industry He serves as a consultant to industry and has and has participated in guideline development for actively participated in clinical trials and educa• the management of peritonitis since 1986. Dr. tional workshops. Dr. Burkart has written or co• Golper has published 136 articles and currently authored more than 90 publications, and is cur• directs research on topics such as peritoneal rently involved in research related to adequacy fluid-drug interactions, peritonitis in PD, and ath• of peritoneal dialysis, improving outcomes on erosclerotic/thrombotic risk factors in ESRD. peritoneal dialysis, and alternative osmotic Active in many professional societies, at present, agents for peritoneal fluid. He is a co-investigator he serves on the Board of Directors for the Renal in the Hemodialysis Study and serves on the edi• Physicians Association as well as the American torial boards of Peritoneal Dialysis International Association of Kidney Patients. Dr. Golper re• and Advancement of Renal Replacement Ther• ported an affiliation with the University of Ar• apy. Dr. Burkart reported an affiliation with Bax• kansas for Medical Sciences Kidney Center and ter Healthcare. QualChoice, a managed care organization. Catherine Firanek, RN, CNN, MBA, is Peri• David Churchill, MDCM, FRCPC, FRCP toneal Dialysis Nurse Manager at Circle Medical (Edin) (Work Group Vice-Chair), is Professor of Management Dialysis Services in Chicago, Illi• Medicine and Director, Division of Nephrology, nois. Ms. Firanek has been involved with perito• lt McMaster University in Hamilton, Ontario, neal dialysis program establishment, patient care, Canada. His funded clinical research activities and education for the past 14 years. She is an have included urolithiasis research, polycystic advocate of patients and self-care dialysis kidney disease, quality of life in ESRD, economic through patient education programs. Nationally, malysis, erythropoietin in ESRD, and adequacy Ms. Firanek has been actively involved in nurs• )f peritoneal dialysis. He is co-principal investiga• ing educational program planning for the Na• :or of the CANUSA on the adequacy of peritoneal tional Kidney Foundation, and serves on a nurs• jialysis. He has over 100 peer-reviewed publica• ing advisory board for industry for creation of ions. Dr. Churchill has served on the scientific nursing tools for CQI programs and nursing edu• 'eview panels for the NIH, Kidney Foundation of cation. Ms. Firanek has published several articles Canada, and the Ontario Ministry of Health, as in the areas of peritoneal dialysis in an urban .veIl as on advisory committees for erythropoietin population, peritonitis incidence, and adequacy . md dialysis services in Ontario. He has a cross• She is the past Chairperson of the Illinois Council Ippointment in the Department of Clinical Epide- Of Nephrology Nurses and Technicians, a mem• ber of CNNT, ANNA, the National Kidney Foundation of Illinois Board of Directors, and © 1997 by the National Kidney Foundation, Inc. the International Society of Peritoneal Dialysis 0272-638619713003-0220$3.0010 Subcommittee on International Studies.

3134 AmericanJournal of Kidney Diseases, Vol 30, No 3, Suppl 2 (September), 1997: pp S134-S136 BIOGRAPHICAL SKETCHES OF THE NKF-DOQI PD ADEQUACY WORK GROUP MEMBERS S135

Denis Geary, MB, MRCP(UK), FRCPC, is sor of Medicine and Director of the Nephrology Associate Professor, Department of Pediatrics, at Divisi9n at the University of Missouri-Colum• the University of Toronto, and Chief of Nephrol• bia. Dr. Nolph has served as President of the ogy at the Hospital for Sick Children in Toronto, American Society for Artificial Internal Organs Canada. Dr. Geary's clinical training took place and as President of the International Society for in the United States, United Kingdom, and Ire• Peritoneal Dialysis. He has written more than land. He has practiced as a staff physician in 500 scientific publications and has edited four pediatric nephrology for 15 years. Dr. Geary's major textbooks on peritoneal dialysis. Dr. Nolph publications involve the subjects of dialysis, and was Director of the Clinical Coordinating Center the growth and development of children with of the National Institutes of Health USA CAPD chronic renal failure. Registry from 1981-1988. He has been the pro• Frank Gotch, MD, is Medical Director of Di• gram chairman of the Annual Peritoneal Dialysis alysis Treatment and Research at Davies Medical Conference for 17 years. Dr. Nolph serves as Center in San Francisco, California. Dr. Gotch a consultant to Baxter Healthcare, Inc., and is has worked in clinical dialysis and dialysis re• affiliated with Dialysis Clinics, Inc. search, particularly quantification of therapy, for Neil Powe, MD, MPH, MBA, FACP, is As• 30 years. He chaired the NIH Hemodialyzer sociate Professor of Medicine at the Johns Hop• kins University School of Medicine, with joint Evaluation Study Group in 1972 and the National appointments in the Departments of Epidemiol• NIH Conference on Adequacy of Hemodialysis ogy and Health Policy & Management at the in 1975. He served on the planning committee Johns Hopkins University School of Hygiene and and as kinetic consultant to the National Cooper• Public Health. Dr. Powe is Director of the Dial• ative Dialysis Study. Dr. Gotch presently serves ysis Care Patient Outcomes Research Team at on the steering committee of the current HEMO Johns Hopkins University, a national study of study and is Co-Principal Investigator of a coop• patient outcomes in peritoneal dialysis and hemo• erative study of randomized peritoneal dialysis dialysis. He has trained in internal medicine, epi• prescriptions and clinical outcome. He has more demiology and health services research. He has than 100 publications and provides consultation extensive experience in developing and measur• in dialysis kinetics and dialysis systems develop• ing outcomes in dialysis patients using data from ment to industry. Dr. Gotch serves as a consultant prospective studies, the USRDS, Medicare re• to Fresenius Medical Care. cords, and patient surveys. Dr. Powe has pub• Linda W. Moore, RD, is currently Manager, lished more than 70 articles on such topics as the Clinical Marketing for SangStat Medical Corpo• effectiveness of recombinant human erythropoie• ration, a company focused solely on transplanta• tin therapy in dialysis patients, treatment choices tion, applying a disease-management approach of peritoneal dialysis versus hemodialysis pa• for improving outcomes and decreasing the costs tients, vascular access management, and cost-ef• of therapy. She has worked in dialysis and trans• fectiveness analysis of alternative treatments for plantation for many years at the University of ESRD. Dr. Powe was a member of the Institute Tennessee-Memphis where she focused on clin• of Medicine Committee on Measuring, Manag• ical outcomes in developing research protocols ing and Improving Quality of Care in the ESRD and patient care protocols. Ms. Moore has pub• Treatment Setting. He has testified before the lished more than 40 articles on renal nutrition U.S. Congress on the role of patient outcomes and metabolism and consults both for industry research in improving the quality of care in the and private, not-for-profit organizations. She has Medicare ESRD program. served on many local and national committees Harmeet Singh, MD, is currently in private committed to defining and improving patient out• practice with Western Nephrology and Metabolic comes. Ms. Moore is currently Chair-Elect of Bone Disease P.C., in Denver, Colorado. He fin• the Council on Renal Nutrition for the National ished a clinical and research fellowship in ne• Kidney Foundation and also serves as a work• phrology at the University of Colorado Health group member of the ESRD Core Indicators for Sciences Center and was on the clinical faculty the Health Care Financing Administration. in the nephrology division at the University of Karl Nolph, MD, FACP, FRCPS, is Profes- Arkansas for Medical Sciences prior to his cur- S136 GUIDELINES FOR PERITONEAL DIALYSIS ADEQUACY

rent appointment. In his limited academic tenure, and teaching clinical nephrology for 21 years and he has written on a variety of clinical nephrology has been responsible for the peritoneal dialysis topics, including sodium and potassium disorders program of the Albuquerque V A for a number and contrast nephrotoxicity. of years. He has more than 170 scientific publica• Brendan Teehan, MD, FACP, is Clinical tions. His current research is directed toward is• Professor of Medicine at Jefferson Medical Col• sues related to adequacy of peritoneal dialysis. lege in Philadelphia and has been Chief of the He is active in many professional societies. Division of Nephrology at Lankenau HospitaU Bradley Warady, MD, is Director of Dialysis Lankenau Medical Research Center from 1992- and Transplantation and Chief of Nephrology at 1996. He has served as principal investigator at The Children's Mercy Hospital, and Professor of the NIH-sponsored National Cooperative Dial• Pediatrics at the University of Missouri-Kansas ysis Study and as principal investigator for the City School of Medicine. Dr. Warady's clinical ongoing Morbidity and Mortality in Hemodialy• and research focus is end-stage renal disease, sis Study. Dr. Teehan was also the Principal In• with particular emphasis on peritoneal dialysis. vestigator at the Lankenau site for the CANUSA He established the Pediatric Peritoneal Dialysis Adequacy of Peritoneal Dialysis Study. He has Study Consortium and currently directs research written numerous articles and chapters dealing with kinetic modeling, nutrition, and adequacy projects on a number of topics including: the of peritoneal dialysis. Dr. Teehan is the Past Pres• impact of "Flush Before Fill" on peritonitis in ident of the American Society for Artificial Inter• patients receiving automated peritoneal dialysis; nal Organs and is presently a member of the long-term function of the peritoneal membrane; finance committee of the Internal Society of Peri• and growth hormone usage in pediatric dialysis toneal Dialysis. He reported an affiliation with patients. Dr. Warady has published more than R.C.G.I. and Main Line Suburban Dialysis Cen• 100 articles and currently serves on the executive ters. committee of the Pediatric Nephrology and Urol• Antonios Tzamaloukas, MD, FACP, is Act• ogy Council of the National Kidney Foundation ing Chief of the Renal Section of the Albuquer• and the scientific advisory committee of the que VA Medical Center and Professor of Medi• North American Pediatric Renal Transplant Co• cine at the University of New Mexico School of operative StUdy. He reported receiving research Medicine. Dr. Tzamaloukas has been practicing grant funding from Baxter Healthcare Inc.