Consensus Conference on Standardized Listing Criteria for Renal Transplant Candidates

January 30, 1997, Natcher Conference Center, National Institutes of Health, Bethesda, Maryland WELCOME; Leslie Miller, MD, President, American Society of Transplant Physicians This conference was convened to review and discuss scientific data pertinent to the development of standardized listing criteria for renal transplantation and to develop a consensus on the major issues. It was one of a series of conferences on standardized listing criteria for each of the major transplanted organs. The conference was sponsored by the American Society of Transplant Physicians (ASTP) with the co-sponsorship of the United Network for Organ Sharing (UNOS) and the National Institutes of Health. A number of members of the American Society of Transplant Surgeons (ASTS) also played a key role in bringing this conference about. INTRODUCTION; Bertram Kasiske, MD The major goals of this conference were to define the issues, review the pertinent scientific data, and develop specific recommendations that the ASTP could make to UNOS. It was stressed that it was the responsibility of UNOS to create and implement any standardized listing criteria. Listing criteria should address when patients should be placed on the UNOS waiting list and when patients should begin accruing waiting time points. Listing criteria should not necessarily address who should otherwise be a candidate for transplantation. Moreover, it was not the purpose of the conference to discuss revisions of the current UNOS point system for allocating organs. The ideal listing criteria should maximize the benefit for patients. Standardized listing criteria should be enforceable, equitable, and easy to understand and carry out. The possibilities include (1) having no criteria, (2) requiring that patients be on dialysis, (3) requiring that patients should have a glomerular filtration rate (GFR) estimated by serum creatinine or creatinine clearance below a defined level, (4) handling diabetics differently than nondiabetics, and (5) requiring certain signs and symptoms of uremia be present. SESSION I: DEFINING THE ISSUES; Chairs: Bertram Kasiske, MD; Robert Gaston, MD A Brief History of the Listing Criteria Initiative; Douglas Norman, MD In 1993, UNOS established the Ad Hoc Committee on Standardized Listing Criteria, chaired by John O'Connell, MD, and charged the committee with developing standardized listing criteria for all organ transplantation. The committee met and ultimately asked the UNOS organ-specific committees to help develop specific listing criteria. In 1994, the president of UNOS disbanded the ad hoc committee and gave full responsibility to the organ-specific committees to develop the listing criteria. Over the next year, the UNOS Kidney-Pancreas Committee drafted a proposal for criteria. In March 1995, the UNOS Board of Directors approved the draft version for public comment. That same year, the ASTP Board of Directors charged the Clinical Practice Guidelines Subcommittee of the Patient Care and Education Committee to develop an ASTP response to the UNOS call for public comment. In May 1996, the committee recommended to the ASTP Board that the UNOS version should be revised and strongly recommended that a consensus conference should be convened for this purpose. The ASTP Board further revised the subcommittee's proposal and brought the resulting draft to the ASTP business meeting in May 1996. At the business meeting, it became clear during a floor debate that there was no consensus on either the UNOS proposal or on the ASTP Board's counterproposal. The ASTP Board decided to convene a consensus conference. In the meantime, the UNOS Board of Directors decided to table their proposal and to entertain completely new proposals after the ASTP consensus conference. Do Data Suggest Some Patients Are Listed Earlier than Others? Wendy London, PhD; Bertram L. Kasiske, MD An analysis was carried out to determine whether there were differences in patients who listed earlier, i.e., patients who were not yet on hemodialysis, compared with those who were on hemodialysis at the time of listing. Although it is imperfect, being on dialysis is the only variable in the UNOS database that might indicate how advanced a patient's renal disease was at the time of listing. The 41,362 registrations for kidney or kidney-pancreas transplantation made between April 1, 1994, and June 30, 1996, were analyzed. Multiple listings accounted for only 4% of the total, so that removing these or not was unlikely to alter the results. Overall, 18.4% of the registrants were not yet on dialysis at the time of being placed on the waiting list. An attempt was made to estimate the renal function of patients not yet on dialysis. To date, 3,150 of these patients had received a transplant and 1,460 had data allowing the calculation of creatinine clearance at the time of transplantation. Among this very small and potentially biased sample, the creatinine clearance estimated by the Cockcroft-Gault method was 16.7±10.2 ml/min at the time of transplantation. A multivariate, logistic regression analysis demonstrated that several patient characteristics were associated with being on dialysis at the time of listing. The reference group in this analysis was white race, male, age 18-55 years, no previous kidney transplant, employment less than full time, a source of payment other than Medicare or private insurance, transplanted in 1996 at a center that performed less than 400 transplants in 1996, not insulin-dependent, and high school education. Characteristics associated with a greater likelihood (odds ratio [OR] >1.00) of listing before dialysis included age <18 years, OR=1.9; age 56 and older, OR=1.1; previous kidney transplant, OR=1.8; full-time employment, OR=2.0; and private insurance, OR=1.1. Characteristics associated with a lower likelihood of listing before dialysis (OR<1.00) included black race, OR=0.5; Hispanic, OR=0.6; Asian and other race, OR=0.5; and payment by Medicare, OR=0.4. In addition, a center-effect analysis demonstrated that after adjustment for the above patient characteristics, 36% of centers were significantly different with regard to the proportion of patients listed before dialysis. In summary, these data suggest that there are differences between patients who are or are not listed before beginning dialysis. This could be due to differences in access to health care and referral for renal disease treatment and not necessarily due to differences in selection for placement on the transplant waiting list. Further research is needed to better understand when patients are placed on the waiting list and why. Can Signs and Symptoms Reliably Predict End-Stage Renal Disease? Eli Friedman, MD Is the deterioration in renal function over time a continuous process that can be predicted and are there signaling signs and symptoms to tell us which point a patient has arrived at? At a creatinine clearance greater than 20 ml/min, there are few overt signs and symptoms of uremia. Indeed, most transplant physicians and surgeons would consider that an allograft with a creatinine clearance of 20-25 ml/min is functioning adequately. It is with creatinine clearances between 8 and 19 ml/min that most signs and symptoms of uremia occur. However, with renal function in this range, signs and symptoms of uremia are relatively nonspecific. Below a creatinine clearance of 7 ml/min, signs and symptoms of uremia become more predictable. There are three that are not found in other diseases: (1) an unsolicited description of a "metallic" taste, (2) a characteristic reversal of the sleep-wake cycle, and (3) episodic emesis with an immediate return of hunger. These three findings signal advanced uremia, and transplant candidates should be listed by this time. The emergencies of the uremic syndrome, such as pericardial tamponade, occur too late to be helpful as listing criteria. In summary, the unpredictability of uremic signs and symptoms, along with their tendency to occur relatively late in the course of renal disease, precludes the use of signs and symptoms as listing criteria. Are there any clinical predictors of renal disease progression? There are few publications in the literature reporting disease-specific histologic or clinical findings that predicted the rate of renal disease progression. Even in these reports, the deterioration was very variable. The possible exception is diabetic nephropathy, where the course seems to be more predictable. What is the Best Measure of Renal Function? Andrew Levey, MD Data from the Modification of Diet in Renal Disease (MDRD) study were used to define the best (both accurate and practical) method for estimating GFR determined by iothalamate clearance. There were 1628 patients with complete baseline data that could be analyzed. Methods for estimating GFR that were compared were (1) inverse serum creatinine, (2) creatinine clearance predicted from the Cockcroft-Gault formula, (3) measured creatinine clearance, (4) the mean of creatinine and urea clearances, (5) a multiple regression model using urine creatinine, creatinine clearance, and race, (6) a multiple regression model using serum creatinine, age, gender, race, serum urea nitrogen, and urine urea excretion; and (7) a multiple regression model including the same variables used in method 6 but substituting serum albumin urine urea. The regression coefficients for the variables in each of these models were determined using a sample of 1070 patients randomly selected from the MDRD patient population. The predictability of each of these models was then validated using a separate sample of 558 MDRD study patients. The results showed that the two multiple regression models (methods 6 and 7) were the best predictors of GFR, leaving only 8.8% and 9.7% of the variability in GFR unexplained, respectively. There was little substantive difference between these two models, so that a urine collection appeared to be unnecessary. The final model was: GFR = 170 × SCr-.999 × Age-.176 × [0.762 if Female] × [1.180 if Black] × SUN-.170 × Salb.318 where SCr is serum creatinine, SUN is serum urea nitrogen, and Salb is serum albumin. Creatinine was determined in the MDRD study using the kinetic alkaline picrate assay that is used in standard practice. The MDRD study excluded patients with type I diabetes and patients with type II diabetes treated with insulin. It excluded patients less than 18 and greater than 70 years old. Also excluded were renal transplant recipients. The analysis was not limited to individuals with GFR less than 30 ml/min, the group of greatest relevance to possible listing criteria. Therefore, the method may need additional study and validation in these populations. The mean GFR at the time or end-stage renal disease (ESRD) was 9.0±3.0 ml/min, n=93. Whether there are clinical features that predict which patients are most likely to have renal disease progression and whether there is a level of GFR below which all patients progress to ESRD were also using the data from the MDRD study the clinical in the there were 6 predictors of renal disease urine kidney disease, serum black race, and serum However, only one of the variability in renal disease progression was predicted by the of these There was no between baseline GFR and in a level of GFR any other clinical was a of renal disease progression in this population. in for Transplant Patients MD; Robert Data from the United Renal Data were to determine whether patient would relatively better than with transplantation compared with dialysis. The has made that the for diabetics after transplantation is relatively better than that of nondiabetics, that diabetics should be listed than The only dialysis patient with and to that of transplant is the group of patients on the UNOS waiting list not yet Therefore, the of patients after transplantation was compared with that of patients dialysis on the UNOS waiting list. The analysis included patients dialysis during and of were A multivariate, regression model was used for this The model was for age, race, of of and time from ESRD to The results that was for compared with dialysis, only in the immediate In the was better with transplantation compared with dialysis. Indeed, the rate after transplantation was only of that for all patients on dialysis. for transplant all patient from transplantation. There were small differences in the time after transplantation when the became less compared with dialysis, but all patient that point the for all for patients with for and for patients with other renal There were also differences in the diabetics patients with and patients with other renal disease These data to that any patient group had greater than any other in the of of Renal Disease MD A review of the literature was to address whether there are renal disease-specific differences in the rate of progression that would suggest specific patient should be listed than The was whether patients who a defined level of renal function progress in a to for Although the rate of progression is it is also variable patients with the same renal disease Moreover, a number of patients not renal function over as as in the MDRD of patients showed no of and this was of the level of baseline There are also results that treatment with or can the rate of progression in patients. to the rate of progression should be before any patient on the transplant waiting list. There is no that the rate of progression be different in patients with of renal patients with diabetes not have a rate of of renal function and as a not progress any than patients with other renal There are also data that treatment may the rate of progression of diabetic In summary, it would be to any group of patients on a transplant list before on that the rate of renal disease progression would be the of urea can be an by the creatinine clearance this of transplantation should be this is What are by Listing MD the in is and equitable, the of transplant is to more their are to criteria and to how scientific data are to those differences in patients differences in how are What the public What of to have in It is to the of that the public the system of is It is not who their to be used are that there is in the of organ transplants have a that the system may not be In the public the system to the number of not and all with of public criteria are At it may be not to all as having might be a there is to patients it is best to the same define on how In the of this might be by age, gender, race, and It is to make a for any are define may not have but it regard to this might mean (1) the (2) those most likely to (3) those most (4) those who have it by and (5) of transplant has not a of this too that the data tell us where the for This is not need to the What should be is a of benefit is possible and is to of in need to address how are to to that not should consider that criteria, for might best be of the transplant and in the SESSION Chairs: MD; MD be Listed MD The UNOS proposal that patients with diabetes progress more to ESRD and more on dialysis when compared with There are several that diabetics should be listed The 1996 data that the rate on dialysis is for diabetics than patients with other renal These data also indicate that diabetics begin dialysis with lower serum creatinine and creatinine clearances than Data from the UNOS that patient after kidney transplantation is more in diabetics compared with In addition, these data that dialysis patients more for diabetics than There is a but not as of dialysis on data from the and UNOS transplant patients are on dialysis for less than more years, the patient rate after transplantation is significantly diabetics and However, the of diabetics who had on dialysis before transplant for more than years is lower than that of not as differences were for These data that diabetics more than after transplantation if not a kidney years of dialysis. in the data, data from the of that diabetics who were not on dialysis before transplant had significantly after transplant than diabetic patients who were on dialysis before However, the data, these data also significantly better for diabetics who were transplanted compared with diabetic patients who were on dialysis before transplant 4% difference at diabetics on dialysis at transplant have lower patient and after also data from kidney-pancreas transplants at the of The rate of due to was for dialysis before transplant compared with those regard to listing criteria, the is not to an for diabetic patients but to a clear and should not patients or physicians who have the to list patients before on dialysis. In the best of should transplant all patients before need dialysis. However, the that transplants to should not mean that it to no It is responsibility to the care of those who have the than the care of those who have the best need to listing criteria, as for and and those patients who on dialysis to on the list of whether or not are be Listed MD In an ideal all of patients would transplanted before needed dialysis, but are not in an ideal are to all of patients. The is not whether should diabetic patients to be but can other patients to be further are at a and to be further There is no to an to in the current system kidney-pancreas transplant are for to a there have over that also also at the of the list are of a kidney transplant also to the At in the years, the number of kidney transplants has the number of transplants performed has by this time, the waiting list has from 400 to at the the is not as but the same there has a in the number of However, this has in the number of kidney There has also a in the number of kidney-pancreas the number of transplants has the the waiting list is patients an in the that kidney-pancreas transplants are and the waiting time for a transplant is than for a kidney diabetics treatment in standardized listing criteria in an to their how be to What patients with access What patients who Even if that diabetics progress more for which there is little in then there are other that should also be There are other such as There is no to The and of the current UNOS system is that can patients in the and tell for organ and only waiting time and determine when a it is to in the and using the as a listing criteria define type I and type II patients the be certain that patients the system is It be and It be and easy to patient treatment is to these be MD renal disease has a number of that are to These include and and and system The number of less than the in is to the level of renal function at the time of listing for renal transplantation. than is in of patients renal function is ml/min time when there are no uremic signs or and is in more than of patients renal function is less than are in of on dialysis but in only after renal transplantation. In addition, transplantation has to and to on dialysis to be and have on the other and to of from high school and renal is associated with results in but transplantation has to In there is a of data that suggest of renal transplantation are to Data from the indicate that the of ESRD in is and has very little in the the number of transplants has over this time, the proportion of transplants performed in has is at as in as it is in of a renal transplant the are transplanted to years of ESRD of so a number be placed on the waiting list. UNOS has to in kidney years and years in to that are transplanted In summary, renal from a is the of treatment for ESRD However, not all can Therefore, standardized listing criteria should be to transplantation for Patient be Andrew MD from a review of the literature and from as a dialysis and transplant from a review of the literature was that there was no that any other patient group should be also gave as a dialysis patient of years, and as a renal transplant as a dialysis patient in an in which only limited of patients were selected for dialysis, by committees made of and Indeed, in was not a candidate for dialysis and only on dialysis could it and could that it is very for patients to that the selection process for renal transplantation is and that no one should be treated differently there are to The of kidney transplantation should be (1) a of and (2) a high There should be a between these two goals of and To one patient group over a is to all other Even when in one of patients are more that in of to make use of and need may a of a group than reference to would to and In this only patients who a kidney transplant to their in a very should be in any of the ASTP Robert Gaston, MD A was in 1996 to and of all transplant centers in the United the that were were and of the were by surgeons or one exception the results were the same for surgeons compared with of that their center placed patients on the waiting list before transplantation was to waiting time. However, that other centers in this practice. Indeed, of to list patients so not be at a to patients listed at other all the that standardized criteria should be to all the there should only be recommendations that are not and the there should be no or were asked who should develop the standardized listing criteria, only the should be UNOS should the and the ASTP should develop standardized listing criteria. more surgeons than that the should be asked level of renal function should be used as a listing criteria for a 20 ml/min and 20 ml/min should be used and In summary, this that transplant physicians and surgeons that patients are listed before transplantation is in to waiting points. that along with the and not the should develop standardized listing criteria. A of found a GFR of ml/min to be a for to diabetic and An was that of the during the an was on key issues. Among those it was that (1) standardized listing criteria should be (2) dialysis was not a criteria, (3) a serum creatinine estimate of GFR for age, gender, race, should be used as a listing criteria, (4) diabetes any other renal disease should be and (5) care should be that are not in any The to be over whether the estimated GFR for listing should be ml/min or 20 Whether to patients to be listed but not to begin accruing waiting time was also that this would the purpose of having standardized listing criteria. In a of a clear of that listing should be with beginning to waiting time. the of that there should be a process patients not the standardized criteria could to the organ for listing as an UNOS transplant candidates and have a to and of Therefore, standardized criteria should be used to determine for listing candidates for renal transplantation. criteria used to determine for listing should be on scientific To in the of scientific data, all patient should be treated Listing criteria should be to listing patients who may not yet ESRD yet to at the for transplantation for those with advanced and renal The are criteria for patients on the UNOS list to a A transplant candidate should have renal disease with GFR to be <18 ml/min, as determined GFR = 170 × × Age-.176 × [0.762 if Female] × [1.180 if Black] × SUN-.170 × Salb.318 where SCr is serum creatinine, SUN is serum urea nitrogen, and Salb is serum albumin. patient not for listing by this may for listing as an The in these be a committee or for this purpose by UNOS. A number of patients listed by any center as may be for review and for waiting time should begin accruing at the time of listing. less than 18 years of age may be listed for kidney transplantation at of GFR than may be listed for kidney transplantation as as begin to the of uremia to occur in their age MD, Center, MD L. for of MD, of of of at Wendy London, Eli Friedman, MD, Center, Robert Gaston, MD, of of MD, of of MD, Bertram L.