The early 1990s saw intense interest in the achievement

Transcription

1 Peritoneal Dialysis International, Vol. 23, pp Printed in Canada. All rights reserved /03 $ Copyright 2003 International Society for Peritoneal Dialysis CHANGES IN PERITONEAL DIALYSIS PRACTICES IN CANADA Rafael A. Perez, Peter G. Blake, Kailash A. Jindal, Kimberly Badovinac, Lilyanna Trpeski, and Stanley S.A. Fenton, on behalf of the Canadian Organ Replacement Register EPREX Study Group a Objective: Over the past decade, clinical studies and clinical practice guidelines have suggested the use of higher small solute clearance targets for patients on peritoneal dialysis (PD). This study asks whether these recommendations have translated into changes in clinical prescription of PD. Study Design: Data were collected annually from 1996 to 1999 on all prevalent dialysis patients in 24 Canadian centers, accounting for approximately 40% of the Canadian chronic dialysis population. Approximately a third of these patients were on PD. Full details of each patient s prescription were recorded, with particular attention to dwell volumes and frequency of exchanges for continuous ambulatory PD (C) and to total treatment volumes and day dwells for automated PD (). The most recent Kt/V and creatinine clearance values available were recorded a Investigators in the CORR-EPREX Study were Paul Barre (Royal Victoria Hospital, Montreal, QC), Peter Blake (London Health Sciences Centre, London, ON), Pierre Cartier (Hôpital du Sacré-Coeur de Montreal, Montreal, QC), David Churchill (St.J oseph s Health Care System, Hamilton, ON), Roland Dyck (Royal University Hospital, Saskatoon, SK), Antoine Farah (C.H. des vallées de l Outaouais Pavillon de Hull, Hull, QC), William Fay (Sudbury Regional Hospital, Sudbury, ON), Stanley Fenton (University Health Network, Toronto, ON), Adrian Fine (St.Boniface General Hospital, Winnipeg, MB), Paul Handa (Saint John Regional Hospital Atlantic Health Sciences Corporation, Saint John, NB), John Harnett (Health Care Corporation of St.J ohn s Health Sciences Centre, St.J ohn s, NF), John Jeffery (Health Sciences Centre, Winnipeg, MB), Kailash Jindal (Queen Elizabeth II Health Sciences Centre, Halifax, NS), Jacques Jobin (Centre hospitalier Angrignon, Verdun, QC), David Kates (Kelowna General Hospital, Kelowna, BC), Joanne Kappel (St.P aul s Hospital, Saskatoon, SK), Serge Langlois (Hôtel Dieu de Quebec, Quebec City, QC), Adeera Levin (St.P aul s Hospital, Vancouver, BC), Tom Liu (Grand River Hospital, Kitchener, ON), William McCready (Thunder Bay Regional Hospital, Thunder Bay, ON), Linda Nolin (Hôpital Maisonneuve Rosemont, Montreal, QC), Edwin Toffelmire (Kingston General Hospital, Kingston, ON), Richard Turcot (C.H. Regional Trois-Rivières Pavillon St.J oseph, Trois-Rivières, QC), Raymond Ulan (University of Alberta Hospital, Edmonton, AB). The Canadian Institute for Health Information (CIHI)/ Canadian Organ Replacement Register (CORR) coordinators for the study were Pat Birkland, Pauline Copleston, Jeannette Gerard, Norma Hall, Lisa Mighton, and Daria Parsons. for each patient and the overall results for each year were compared to present treatment recommendations. Setting: 24 university- and community-based hospitals. Results: From 1996 to 1999, the use of, relative to C, grew from 14% to 28% of all PD patients. Among C patients, the proportion using dwell volumes greater than 2 Lrose from 14% to 32%, and the proportion doing more than 4 dwells per day rose from 16% to 28%. The mean daily volume of prescribed fluid for C patients increased from 8.3 to 9.1 L. As a result, the proportion of patients achieving a weekly Kt/V above 2.0 rose from 54% to 72%, and those receiving a Kt/V less than 1.7 fell from 22% to 10%. For creatinine clearance, those exceeding 60 L per week rose from 63% to 73%. For, the mean treatment volume rose from 11.8 Lin 1996 to plateau at about 13.4 Lin 1998 and However, the proportion of patients receiving more than 1 day dwell grew from 31% in 1998 to 40% in 1999, and the proportion that were day dry fell from 25% to 17%. For, the proportion of patients with a Kt/V above 2.0 rose from 67% to 77%, and with a creatinine clearance above 60 L, from 62% to 70%. The proportion with no recent clearance value recorded fell during the course of the study, from 45% to 27%. Conclusion: There was a marked change in PD prescription practices in Canada during the second half of the 1990s. This occurred in response to clinical studies and publication of guidelines. There is room for further improvement, especially with respect to the proportion of patients that did not have regular clearance measurements made. Perit Dial Int 2003; 23: KEY WORDS: Automated peritoneal dialysis; adequacy; Kt/V; creatinine clearance. The early 1990s saw intense interest in the achievement of adequate small solute clearance in patients treated with peritoneal dialysis (PD). This interest culminated in the publication in 1995 of the Canada USA (CANUSA) Peritoneal Dialysis Adequacy Study, which showed a clear association between delivered clearances and patient survival (1). Correspondence to: P.G. Blake, Division of Nephrology, 800 Commissioners Rd. East, London, Ontario N6A 4G5 Canada. peter.blake@lhsc.on.ca Received 1 March 2002; accepted 20 September

2 PEREZ et al. JANUARY 2003 VOL. 23, NO. 1 PDI That study was followed by the publication of new small solute clearance targets by the National Kidney Foundation Dialysis Outcomes Quality Initiative (DOQI) Workgroup in 1997, and by the Canadian Society of Nephrology Peritoneal Dialysis Workgroup in 1999 (2,3). These developments led to changes in PD prescription practices during the middle and late 1990s (4,5). In particular, there was a notable increase in the use of automated PD () relative to continuous ambulatory PD (C) (4). In addition, there was an increased tendency for C patients to receive larger dwell volumes and extra exchanges. Similarly, patients were increasingly likely to have one or more day dwells(4). We attempted to track these changes in PD prescription patterns and the effect they had on delivered clearances. This was done using a large database called CORR-EPREX, which was set up as a collaboration between the Canadian Organ Replacement Register (CORR) (6) and Ortho Biotech, the manufacturer of Eprex, a recombinant erythropoietin product marketed in Canada. In particular, we sought to look at whether the prescription changes were leading to a greater proportion of patients achieving target clearance values. METHODS The CORR-EPREX database was established in Cross-sectional data were collected on 1J anuary of each year from 1996 to 1999, inclusive, on all prevalent chronic dialysis patients in each of the participating units. In all, data were collected from 24Canadian units, accounting for approximately 40% of all chronic dialysis patients in the country. In the case of PD patients, the information collected included the exact prescription the patient was receiving on the day concerned and the most recent Kt/V and creatinine clearance (CrCl) measurements from the chart. The methods used to measure Kt/V and CrCl were not specified but centers were requested to include both peritoneal and renal contributions in the clearance values provided. They were also asked to provide a CrCl value that had been normalized in the conventional fashion to 1.73m 2 body surface area. With respect to C patients, dwell volumes and number of exchanges daily were recorded for all years. Use of the night exchange device to do an extra exchange was only recorded for 1998 and 1999 (7). Patients using this device were counted as C and not. In the case of, the total daily fluid volume prescribed, the volume cycled at night, and the volume used for day dwells were recorded for each year. However, the number of night cycles, the cycler dwell volume, and the number and volume of individual day dwells were only recorded for 1998 and During the course of the study, a report was published each year and feedback was sent to all the participating centers, with comparisons to the data for their province and for the country as a whole. RESULTS The total number of dialysis patients participating in the study for the years 1996 to 1999 was 3923, 4520, 4678, and 4901, respectively. The number on PD in each of those years was 1423, 1540, 1431, and 1393, representing 36%, 34%, 31%, and 28% of the total patient population in the study. During this time, there was a significant decline in the proportion of chronic dialysis patients being maintained on PD in Canada (6,8). Also during this period there was a marked increase in the percentage of PD patients using, as distinct from C. This increased from 14% in 1996 to 16% in 1997, to 23% in 1998, and to 28% in Mean age of C patients was 56years in 1996, 59 in 1997, and 58 in both 1998 and Mean age of patients was 53 in 1996, and 55 in 1997, 1998, and C PRESCRIPTION PATTERNS The most marked change in C prescription patterns was the increased use of dwell volumes greater than 2L. The proportion of patients using the higher volumes rose from 14.3% to 32.2% during the course of the study (Table1). There was a marked decrease in the number of patients using dwell volumes less than 2L. The proportion of patients doing more than 4C dwells per day rose from 15.5% to 28.1% during the study. Prescriptions using fewer than 4C dwells per day were uncommon at the beginning of the study and decreased further in frequency over the period monitored (Table1). In 1998, almost half the patients doing 5exc hanges per day TABLE 1 Changes in C Dwell Volumes and Exchange Frequencies: Dwell volumes <2 L L >2 L Mean (L) Exchanges/day < > Mean (n)

3 PDI JANUARY 2003 VOL. 23, NO. 1 CHANGES IN PD PRACTICES IN CANADA were using a night exchange device. By 1999, this proportion had increased to over 85%. As a result of these changes, the mean volume of fluid used per day by C patients increased from 8.3 L in 1996 to 8.7 L in 1997, and to 9.1 L in both 1998 and Similarly, the proportion of patients receiving more than 8L per da y rose from 23.5% in 1996 to 32.8% in 1997, to 41.0% in 1998, and to 45.9% in The proportion of C patients in whom a recent Kt/V or CrCl value had been recorded increased from 41.1% in 1996 to 69.6% in 1999 (Table2). For those patients in whom a value was recorded, the proportion achieving a Kt/V over 2.0 rose from 53.9% in 1996 to 71.7% in 1999 (Table3). More notably, the proportion with a weekly Kt/V less than 1.7 declined from 21.5% in 1996 to 10.1% in With respect to CrCl, the proportion of patients with a value exceeding 60L per week rose from 63.3% in 1996 to 73.1% in The number with a value TABLE 2 C and Automated Peritoneal Dialysis () Patientswith Recorded Kt/V and or Creatinine Clearance (CrCl) Values: C Both Kt/V & CrCl Kt/V only CrCl only Neither Both Kt/V & CrCl Kt/V only CrCl only Neither TABLE 3 Distribution of Weekly Kt/V Values Achieved on C and Automated Peritoneal Dialysis (): C < Mean Kt/V < Mean Kt/V less than 50L per week dec lined from 14.7% in 1996 to 10.6% in 1999 (Table4). PRESCRIPTION PRACTICES The total fluid volume per day prescribed in was 11.8L in 1996, but rose to above 13L in 1998 and 1999 (Table5). The volume that was actually delivered by the cycler rose from an average of 9.7L per treatment in 1996 to 11.2 in 1998, then declined to 10.6L in However, the volume delivered in the form of day dwells increased from 2.1 to 2.8L during the 4years. The number of cycles delivered per night averaged 5, but the proportion of patients receiving fewer than 5 was greater in 1999 than in 1998, while the proportion of patients receiving 7 or more declined between 1998 and 1999 (Table6). The most notable change from 1998 to 1999 was in the patterns of day dwell prescription: 25.4% of patients were day dry in 1998, but only 17.4% were in The number of patients using 2 or more day dwells rose from 30.8% in 1998 to 40.1% in 1999 (Table6). The proportion of patients with a recorded Kt/V or CrCl measurement increased from 55.1% in 1996 to 73.4% in 1999 (Table 2).The mean Kt/V achieved did not change much during the 4years, but the proportion TABLE 4 Distribution of Corrected Weekly Creatinine Clearances Achieved on C and Automated Peritoneal Dialysis(): C <40 L L L >60 L Mean (L/week) <40 L L L >60 L Mean (L/week) TABLE 5 Changes in Prescribed Automated Peritoneal DialysisV olumes: Total volume cycled (L) Total day dwell volume (L) Total volume daily (L)

4 PEREZ et al. JANUARY 2003 VOL. 23, NO. 1 PDI of patients exceeding 2.0 per week, which is the Canadian clearance target, rose from 66.7% to 77.3% (Table3). The proportion with a value below 1.7 per week declined sharply, from 18.1% in 1996 to only 6.3% in With respect to CrCl, the proportion achieving over 60L per week rose from 62.0% to 70.4% between 1996 and 1999, while the proportion achieving below 50L per week did not change appreciably (Table4). DISCUSSION These results show a dramatic shift in the pattern of PD prescription in Canada during the period 1996 to It is not possible to be certain that the results are representative of all Canadian centers and it is possible that willingness to participate in the CORR-EPREX project may be a marker of generally greater interest by a center in improving quality of patient care. However, it should be kept in mind that the population surveyed comprised approximately 40% of all patients in the country. Furthermore, the participating centers came from 9 of 10Canadian provinces and included representative numbers of both academic and community centers. The most notable change was the shift from C to. While much of this shift was driven by patient lifestyle factors, there is little doubt that the intense interest in increasing delivered clearances also contributed to this trend. Other relevant factors were the relative decrease in the cost of during the time concerned and the increased availability of compact, easier to use cyclers. In C, the most notable trend was the increased use of dwell volumes greater than 2L. However, there 56 TABLE 6 Changes in Automated Peritoneal Dialysis PrescribingPractices: (%) (%) Volume cycled per night <8 L L L L >14 L Number of cycles per night < > Number of day dwells was also a trend to use more than 4exc hanges per day. Five exchanges per day increases the risk of noncompliance in C and is significantly more costly than increasing dwell volumes (9,10). However, this prescription pattern was done increasingly with the use of an automated night exchange device (7). This trend has been less apparent in data from the United States (5). With respect to, prescription patterns changed in a subtler manner. The total volumes of fluid prescribed did rise initially during the period studied but, having peaked in 1998, did not increase further in However, it could be said that the volumes of fluid were used more efficiently in terms of achieving higher clearances (11). Thus, there was a relative increase in the amount of fluid used as day dwells relative to that used to cycle at night. Day dwells are, of course, in place for longer periods of time and so achieve relatively greater degrees of clearance than does the same amount of fluid delivered by the cycler (11). This pattern of prescription is particularly apparent in the data from 1999 pertaining to the use of day dwells, which showed that 40.1% of patients had more than 1da y dwell and 12.9% had 3da y dwells. The ease of delivering such prescriptions has been facilitated by modifications in cycler and tubing technology that allow patients to attach to and detach from their cycler tubing at intervals during the daytime (12). This trend may indicate increased sophistication of PD practitioners in terms of achieving more clearance at no increased cost (10). The number of cycles delivered per night has stayed relatively stable, with the bulk of patients receiving 4, 5, or 6 per night. Recent clinical studies would suggest that this does not utilize the full potential of to achieve high clearances and that the frequencies of 7to 9cyc les per 9-hour nighttime period result in greater clearances (13). However, such prescriptions are relatively less cost-effective compared to regimens that use fewer cycles and more day dwells (10). The clinical end point in all this is the actual clearance achieved and, here, there is impressive evidence that progress has been made toward achieving target guidelines. Thus, by the end of the study period, 72% of C patients and 77% of patients were achieving the Kt/V of 2.0 per week recommended by the Canadian guidelines (3). It should be noted that the Kt/V target recommended by DOQI is also 2.0 but, for prescriptions, DOQI suggested that it should be higher, at 2.1 if the patient has fluid in place in the daytime, and 2.2 if the patient does not. The Canadian guidelines did not make this distinction between and C (14). With respect to CrCl, there was also an increase in the proportion of patients reaching the targets. By 1999, this proportion was over 70% for both modalities. Indeed, the target for low and low-average transporters,

5 PDI JANUARY 2003 VOL. 23, NO. 1 CHANGES IN PD PRACTICES IN CANADA who comprise approximately 50% of PD patients, has been set at 50 rather than 60L per week in the Canadian guidelines and, more recently, in the revised DOQI guidelines (15). Thus many of the 25%to 30% of patients not achieving 60L per week in 1999 will represent low transporters with values between 50 and 60L per week, and so the proportion of total patients achieving the targets is likely above 80%. The study did not collect information on peritoneal transport status and so the exact proportion cannot be calculated. Notwithstanding this, it is likely that, in most PD programs there will be a subset of patients that, because of physical or social limitations, will be unable to achieve the recommended targets. In such situations, a value judgment is often made to continue the patient on PD under careful monitoring. Considering such cases, it has to be kept in mind that the clearance targets have never been validated by appropriate randomized controlled trials and, indeed, recent evidence suggests that they may be unduly high (16). Thus, such clinical judgments may be very appropriate. One persisting concern from these data is the significant proportion of patients in whom regular clearance measurements were not being made during the period studied. At the start of the 4-year period, over half the PD patients had no recent Kt/V or CrCl values recorded. However, by 1999 the proportion was down to less than 30%. This is an improvement but still not ideal. It should be pointed out that such clearance measurements are not mandated in Canada by any regulatory agency and are not required for facility or physician remuneration. There is the obvious concern that centers not making clearance measurements are also less likely to be making prescription adjustments, so the clearances achieved in the patient population as a whole may be less than in those in whom measurements were actually made. It should be emphasized, however, that the impressive prescription changes reported here were based on data from all the PD patients in the study. The overall impression is that the attention given to clearances in PD over the past decade has had a major influence on how the modality is prescribed and on the clearances that are being delivered. This is an excellent example of the influence of clinical studies, of clinical practice guidelines, and of the educational efforts associated with them, on clinical prescription. It remains to be seen if these impressive changes will lead to corresponding improvements in patient outcomes. ACKNOWLEDGMENT The authors acknowledge the support given by Ortho Biotech Canada for the CORR-EPREX project. REFERENCES 1. Adequacy of dialysis and nutrition in continuous peritoneal dialysis: association with clinical outcomes. Canada USA (CANUSA) Peritoneal Dialysis Study Group. J Am Soc Nephrol1996; 7: NKF-DOQI clinical practice guidelines for peritoneal dialysis adequacy. National Kidney Foundation. AmJ Kidney Dis 1997; 30(Suppl 2):S Blake PG, Bargman JM, Bick J, Cartier P, Dasgupta MK, Fine A, et al. Clinical practice guidelines of the Canadian Society of Nephrology for peritoneal dialysis adequacy and nutrition. JAm Soc Nephrol 1999; 10(Suppl 13):S Blake PG, Bloembergen WE, Fenton SS. Changes in the demographics and prescription of peritoneal dialysis during the last decade. AmJ Kidney Dis 1998; 32 (Suppl 4):S Rocco M, Souci JM, Pastan S, McClellan WM. Peritoneal dialysis adequacy and risk of death. Kidney Int 2000; 58: Canadian Institute for Heath Information. Canadian Organ Replacement Register Annual Report Vol1: Dialysis and Renal Transplantation. Ottawa: Canadian Institute for Heath Information; Pagé DE. C with a night exchange device is the only true C? Adv Perit Dial 1998; 14: Blake PG, Finkelstein FO. Why is the proportion of patients doing peritoneal dialysis declining in North America? Perit Dial Int 2001; 21: Blake PG, Korbet SM, Blake R, Bargman JM, Burkart JM, Delano BG, etal. A multicenter study of non compliance with continuous ambulatory peritoneal dialysis exchanges in US and Canadian patients. AmJ Kidney Dis 2000; 35: Blake PG, Floyd J, Spanner E, Peters K. How much extra does adequate peritoneal dialysis cost? Perit Dial Int 1996; 16(Suppl 1):S Blake P, Burkart JM, Churchill DN, Daugirdas J, Depner T, Hamburger RJ, etal. Recommended clinical practices for maximizing peritoneal dialysis clearance. Perit Dial Int 1996; 16: Diaz-Buxo JA. Enhancement of peritoneal dialysis: the PDPlus concept. AmJ Kidney Dis 1996; 27: Perez RA, Blake PG, McMurray S, Mupas L, Oreopoulos DG. What is the optimal frequency of cycling in automated peritoneal dialysis? Perit Dial Int 2000; 20: Blake PG. Comparison between DOQI and Canadian guidelines for peritoneal dialysis. Perit Dial Int 2000; 20: National Kidney Foundation. K/DOQI clinical practice guidelines for peritoneal dialysis adequacy, AmJ Kidney Dis 2001; 37(Suppl 1):S Paniagua R, Amato D, Vonesh E, Correa Rotter R, Ramos A, Moran J, etal. Effect of increased peritoneal clearances on mortality in peritoneal dialysis: ADEMEX, a prospective randomized controlled trial. JAm Soc Nephrol 2002; 13: