Changes in the Peritoneal Equilibration Test in Selected Chronic Peritoneal Dialysis Patients1

hanges in the Peritoneal Equilibration Test in Selected hronic Peritoneal Dialysis Patients1 Wai-Kei Lo, Alessandra Brendolan, Barbara F. Prowant, Harold L. Moore, Ramesh Khanna, Zbylut J. Twardowski, and Karl D. Nolph2 W.-K. Lo, A. Brendolan, B.F. Prowant, H.L. Moore, P. Khanna, Z.J. Twardowski, K.D. Nolph, Division of Nephrology, Department of Internal Medicine and Dalton Research enter, University of Missouri-olumbia, olumbia, MO (J. Am. Soc. Nephrol. 1994; 4: 1466-1474) ABSTRAT Fifty-five patients on chronic peritoneal dialysis with Iwo or more peritoneal equilibration tests (PET) performed between 1983 and 1992 with a mean interval of 2 1.9 ± 22.7 months were studied retrospectively. Repeated PET were performed when transport changes were suspected rather than routinely. According to the initial PET, there were 16 high (HI), 17 high-average (HA), 15 low-average (LA), and 7 low (LO) transporters. There was a significant decrease in the mean creatinine dialysate to plasma ratio (D/ P creatinine) in the HI transporters and an increase in the LA and LO transporters. The mean dialysate to instilled glucose ratio (D/Do) significantly increased in the HI transporters. The change in both the D/P creatinine and the D/Do of an individual strongly and inversely correlated to their respective initial values. The change in D/P creatinine and D/Do were significantly and inversely correlated to each other, mdicating an actual transport change. No correlation was found between the change in transport with peritonitis episodes or frequencies. The centrifugal change of transport toward average described here may explain why low clearances or low ultrafiltration rates due to rapid transport are infrequent causes of peritoneal dialysis technique failure, and why patients who have been dialyed for a long period are usually HA transporters. I Receivea May 3, 1993. Accepted August 16, 1993. 2 orrespondence to Dr. K.D. Nolph, Division of Nephrology, Department of internal Medicine. MA436. Health Sciences center. One Hospital Drive, columbla, MO 65212. 146-6673/47.1466$3.OO/O Journal of the American society of Nephrology opyright 1994 by the Amr$can Society of Nephroiogy Key Words: Transport, peritonealmembrane, peritonealequilibration test, membrane durabiilty. membrane stability S ince the peritoneal equilibration test (PET) was introduced by Twardowski et al. in 1987 (1), the abridged (standard (PET) has been widely accepted as a useful means with which to identify the peritoneal solute transport rate and thereby tailor an mdividual s peritoneal dialysis prescription (2-4). The test consists of measuring the dialysate to plasma ratio (D/P) of creatinine at.5, 1, 2, and 4 h and dialysate dextrose to baseline dialysate concentration ratio (D/Do). Dialysate drainage volume (DV) at 4 h is determined by completely draining the dialysate. A simplified fast PET, with only the ratios and drainage volume at 4 h measured, was later developed and accepted as a useful alternative to the original PET. (5). According to the 4-h DIP creatinine result, patients can be categoried into high, high-average, low-ayerage. and low transporters (3, 6). Similarly, the D/Do ratio can be used in such a way. The high transporters usually have good dialysis clearances but poor ultrafiltration. Intermittent dialysis, particularly nocturnal automated, is the preferable therapy. The low transporters usually have good ultrafiltration but low small solute clearances with standard continuous ambulatory peritoneal dialysis (APD) (3, 7). The high-average and low-average transporters, particularly the former, usually have adequate dialysis clearances and ultrafiltration, and typical APD or continuous cyclic peritoneab dialysis is usually the suitable treatment mode. The result of PET repeated within a short period was demonstrated to be very consistent (1, 8). The reproducibility of the test is important for its value in predicting the best treatment modality for patients. However, changes in the result of the test on a long-term basis are not well delineated. If the transport rate changes with time, it means that the best mode of therapy for a patient may also change. Since the PET was first used in our center in 1983, we have accumulated a long observation interval to evaluate change in the peritoneal transport rate as indicated by the PET. In this article, we report our data from the analysis of repeated PET in our patients. It is important to stress that this study was not a prospective routine follow-up of our total pop- 1466 Volume 4. Number 7 1994

Lo et al ulation; the repeated PET was done in cases where change was suspected. Thus, the incidence of change could be different from a prospective study of a total population. PATIENTS AND METHODS All chronic peritoneal dialysis patients in our dialysis center with two or more standard PET performed with 2 L of 2.5% dextrose dialysis solution more than 1 week apart were included in this retrospective analysis. No PET included in the analysis was done within 1 mo of the completion of the peritonitis antibiotic treatment. No PET was performed fewer than 1 days after catheter insertion. The 4-h D/P glucose corrected creatinine ratio (5), the D/Do ratio, and the DV were used as the parameters for study. Patients were categoried into the four different groups of transporters (high, high average, low ayerage and low) according to the DIP creatinine ratio of the initial PET, as described by Twardowski (dividing ratios, >.81. >.65 to.81, >.5 to.65, and.5 or less, respectively (3). A change in DIP creatinine in one individual that was more than 1 SD of the initial DIP creatinmne of this study population was regarded as significant. Statistical differences between groups were analyed by t test, x2 test, and Fisher s exact test where appropriate. A paired t test was used to analye changes within the same group. RESULTS In an unpublished prior control study, we assessed the reproducibility of 4 1 paired PET repeated over short intervals (fewer than 2 days); the mean coeffident of variation was 1.75 ± 1.35% (SD), and the absolute mean difference in DIP creatinine between tests was.24 ±. 17. From 1 983 to November 1 992, there were 55 patients who had two or more PET performed more than 1 wk apart. They represented nearly 25% of all chronic peritoneal dialysis patients over the same retrospective period. The number of PET performed on each patient ranged from two to seven (mean, 2.85). According to the initial PET, 16 patients were high transporters, 1 7 were high-average transportcr5, 1 5 were bow-average transporters, and 7 were low transporters. The mean initial DIP creatinine was.687 ±. 1 53. The initial mode of dialysis therapy was APD, except for two patients on nocturnal intermittent peritoneal dialysis (Table 1). The initial PET were performed at a mean of 5.46 ± 1 1.8 (SD) mo (range,. 1 to 47. 1) from the start of peritoneal dialysis. and the final PET were performed at 27.3 ± 28. 1 mo (range, 1.3 to 1 32.7). The interval was 21.9 ± 22.7 mo (range,.4 to 12.2). The demographic data, the results of the initial and final PET, and the averaged daily dialysate concentration used at the time of initial and final PET of different transporters are shown in Table 1. The mean DIP creatinine decreased and the mean D/Do increased in the initial high and high-average transporters, but they moved in opposite directions, respectively, in the low-average and low transporters. Statistically significant changes were seen in the high. low-average, and low transporters for DIP creatinine (P <.1) and in the high transporters for D/Do (P <.1). The final mean DIP creatinine was still significantly higher in the initial high transporters than the final value in the initial low transporters (P <. 1 ); final means for low transporters were also significantly different from finah means for highaverage transporters (P <.2) and low-average transporters (P <.1). The final mean D/Do of the high transporters remained the lowest and that of the low transporters the highest, with P <.1 between the high and low transporters and P <.3 between the high-average and bow transporters. There was also an increase in the DV in the high transporters and a decrease in the DV in the low transporters, but the changes were not statistically significant (Table 1 ). The average dialysate daily concentration used at the time of the initial PET was significantly higher in the high than in the low transporters (P <.1). Although there was a decrease in the diabysate concentration used in the high transporters and an increase in that used in the lowaverage and low transporters. the changes were not statistically significant. The change in DIP creatinine was found to be strongly and inversely correlated with the initial D/P creatinine value (r = -.746; P <.1; Figure 1). Patients with a high initial DIP creatinine had a negative change, whereas those with a low DIP creatinine had a positive change. A similar relationship was found between the change in D/Do and the initial D/Do (r = -416; P <.1 ; Figure 2) and between the change in liv and the initial DV (r = -.844; P <.1; Figure 3). The change in D/Do correlated with the change in DIP creatinine significantly (r = -.79; P <.1 ; Figure 4). No correlation was found between the change in DV and the change in D/P creatinine or the change in D/Do. The high transporters were found to have a significantly older age than the low transporters (P <. 1 ; Table 1 ). A significant correlation between the DIP creatinine and the age of patient was found with the initial transporter (r =.37 1 ; P <. 1 ; Figure 5 but not with the final PET (Figure 5b). Similarly, the initial D/Do correlated with age in an inverse manner (r = -.353; P <.1 ; Figure 6, but not the final D/Do (Figure 6b). Journal of the American Society of Nephrology 1467

Peritoneal Transport in Repeated PET TABLE I. Results of different rates of transporters according to the D/P creatinine ratio of initial PET#{176} Parameter Hign High Avg Low Avg Low P N 16 17 15 7 Initial PD mode I NTPD I IPD Age, mean 57.6b 52. 52.6 42.5b <. lb SD 8.1 14.2 12.9 9.1 PD months, mean 3.3 7.9 7.2.7 NS SD 8.7 13.9 13.7 1.4 Interval, mean 21.9 25.8 16.4 24.3 NS SD 2.2 25.4 24.7 12.2 Initial DIP, mean.873.713.579.435 SD.4.4.36.63 Final DIP, meanc.74ld.698.698d.559d <. Id SD.81.115.86.11 Initial D/Do, mean.215.389.424.527 SD.63.97.52.72 FinalD/Do,mean.358 s.373.396.463 <.OOld SD.63.82.74.76 InitialDV(mL),mean 2171 2169 2297 2534 SD 216 683 358 168 FinalDV(mL),mean 2296 2184 225 2415 NSd SD 242 67 236 154 f95b Initial concn, mean 2.84b 2.43 2.28 <. lb SD.58.55.66.65 Final concn, mean 2.75 2.41 2.42 2.16 NSd SD.81.5.68.42 a Age and duration of peritoneal dialysis (PD) referred to the initial PET. concn. average daily dialysate concentration (% dextrose). PD mode; all are capd/continuous cyclic PD (PD) unless specified. NS. not significant; NTPD, nightly tidal PD; IPD. intermittent PD. b p compared between high and low transporters. p value between low and high transporters. <.1; low and high-average transporters, <.2; low and low-average transporters. <.1. p compared between initial and final value.. p value between low and high transporters, <.1; low and high-average transporters. <.3..3 - ALL PATIENTS.4 ALL PATIENTS.2.3 I -.1.1. -.2 -.3 r- -.746 p<.1 #{149}#{149} I.2.1. -.1 -.2 -.3 -.8 16 p<.1. -.4 -.2.3.4.5.6.7.8.9 1. INITIAL D/P reatinine -.4 -..1.2.3.4.5.6.7.8 INITIAL D/Do Figure 1. The correlation between the change in DIP creatinine in individual patients and their corresponding initial D/P creatinine value (r= -.746; P<. 1). Figure 2. The correlation between the change in D/Do in individual patients and their corresponding initial D/Do value (r= -.816; P<.1). 1468 Volume 4 - Number 7 1994

Lo et al ALL PATIENTS ALL PATIENTS 15 -.4 1 >.3 5.2.1. -5 r= -.844 p<.ool I -.1 -.2 -.3 r--.79 p<. 1-1 I I I 5 1 15 2 25 3 35 INITIAL DRAINAGE VOME -.4 I I I -.4 -.3 -.2 -.1 HANGE IN..1.2.3 D/P reatinine.4 Figure 3. The correlation between the change in DV in individual patients and their corresponding initial DV (r = -.844; P<. 1). Figure 4. The correlation between the change in D/Do in individual patients and their corresponding change in D/P creatinine (r= -.79; P<. 1). 1. ALL INITIAL PETs ALL FINAL PETs 1..9.9.8 4) :.7 6).8.7. %...6 4) U-.6.5.3 r.37 1 p<.ol.4.3 P-NS.2-2 3 4 5 6 7 8 9 AGE.2-2 3 4 5 6 7 8 9 AGE Figure 5. ( The correlation between the initial D/P creatinine and the age of the patients (r =.371; P <.1). (b) No correlation was found between the final D/P creatinine and the age of the patients. NS, not significant. Figure 7 shows the change in DIP creatinine of different individual patients over time. For those who had a decrease in DIP creatinine, the decrease took place in the first 1 8 mo; the DIP creatinine would usually rise again after 24 mo of peritoneal dialysis. Thirty-eight patients had their PET repeated within the first 1 8 mo of peritoneal dialysis. Among these 38 patients, a significant drop of DIP creatinine was found in the high transporters (P <.1) and a signicant rise was found in the low-average (P <.1) and low transporters (P <.1 ; Figure 8). Fourteesi patients, none of whom were low trans- porters, had at least one PET performed after 1 8 mo of peritoneal dialysis and another PET repeated after 24 mo of dialysis. The DIP creatinine increased in these three groups of transporters (Figure 9). Although the numbers in each group are too small for statistical analysis, the overall change in all 1 4 patients was significant. The DIP creatinine increased from.663 ±.86 to.725 ±.89 (P <.1). hanges in transport were not rebated to peritonitis episodes or their frequency. All PET studies were performed at least 1 mo after any peritonitis episodes resolved. hanges in transport in each group were in Journal of the American Society of Nephrology 1469

Perltoneal Transport in Repeated PET.8 ALL INITIAL PETs FINAL PETs.8 -.7.6 r- -.353 p<.1.7.6 PNS.5.5.4.4 : #{149}..3.3 :.. #{149}.,..2.2.1.1. 2 3 4 5 6 7 8 9. - 2 3 4 5 6 7 8 9 AGE AGE Figure 6. ( The correlatior between the initial D/Do and the age of the patients (r= -.353; P<.1). (b) No correlation was found between the final D/Do and the age of the patients. NS. not significant. I..9.8.7 :.6.5.4.3.2.1 HIGH TRANSPORTERS HIGH AVERAGE TRANSPORTERS.9.8 4,.7.6.5 c5.4..3.2.1 12 24 36 48 6 72 84 96 18121321 4 DURATION OF PERITONEAL DIALYSIS (months). 12 24 36 48 6 72 84 96 1812132144 DURATION OF PERITONEAL DIALYSIS (months) LOW AVERAGE TRANSPORTERS LOW TRANSPORTERS 1. o.5 U.4.3.2.1 o.c 12 24 36 48 6 72 84 96 1812132144 DURATION OF PERITONEAL DIALYSIS (months).7.6.5 :5.4 a..3.2 12 24 36 48 6 72 84 96 1812132144 DURATION OF PERITONEAL DIALYSIS (months) Figure 7. D/P creatinine values of individuals in the four transport groups over time. the same direction in patients without any peritonitis history. There were 28 patients in the study who never had peritonitis. The change in DIP creatinine in the first 1 8 mo was significantly correlated with the initial DIP creatinine (r = -.75; P <.1 ; Figure 1), but the change in DIP creatinine after 24 mo of peritoneal dialysis was not correlated with either the initial D/P or the first DIP creatinine after 1 8 mo of peritoneal dialysis. The final transport rate status of the different initial transporters is shown in Table 2. There was a significant increase in the proportion of high-average transporters (P <.3). Only three initial high transporters. eight high-average transporters, four low-average transporters, and 1 low transporter re- 147 Volume 4 Number 7. 1994

Lo ef al 1..9 P<.o1.8 :.7.6 a- II- - - -.4.3.2 P<.1. high trcn.port.r. n-il V high avg transport.rs n-12 U low avg tran.port.rs n-lo A low tran.port.r. n-5 3 6 9 12 15 18 MEAN DURATION OF PERITONEAL DIALYSIS (months) Figure 8. The change in D/P creatinine in 38 patients with their PET repeated within the first 18 mo of peritoneal dialysis. A statistically significant difference was reached in the high transporters (P <. 1), the low-average transporters (P <.1), and the low transporters (P<.1). NS, not significant. 1..9 Q,.8..4.7 a L 6 %..,._.5.4.3 1 I.. high traneportere n-4 V high overage transporter. n-6. low average traneporters n-4.2-24 36 48 6 72 84 MEAN DURATION OF PERITONEAL DIALYSIS Figure 9. The change in D/P creatinine in 14 patients with at least one PET performed after 18 mo and another performed after 24 mo of peritoneal dialysis. No low transporters was found by this criterion. Overall mean D/P creatinine increased from.663 ±.86 to.725 ±.89 (P<.1). mained in their initial categories. Most high transporters became high-average (1 of 1 6) or lowaverage (3 of 1 6) transporters, most low-average transporters became high-average (9 of 1 5) transporters, and most low transporters became low-average (4 of 7) or high-average (2 of 7) transporters. However, most high-average transporters remained in the same category (8 of 1 7). whereas some became high transporters and some became low-average transporters. There were 1 5 patients with a change of DIP creatinine of more than 1 SD (. 1 53) (Table 2). This significant decrease in DIP creatinine was found in four high transporters and two high-average transporters; a significant increase was found in four low transporters, four low-average transporters. and one high-average transporter. Statistical significance was reached between the high and low transporters for significant increase in DIP creatinine (P <.1) and low-average transporters for significant decrease (P <.3). Journal of the American Society of Nephrology 1471

Peritoneal Transport in Repeated PET 5) U- I -.1.3 -.2.2.1. -.3 r--o.75 p <. 1 n-4 -.4 I ll.2.3.4.5.6.7.8.9 1. INITIAL D/P reatinine Figure 1. The correlation between the change in D/P creatinine in the first 18 mo of peritoneal dialysis and the corresponding initial D/P creatinine (1= -.75; P<. 1). There were no statistical differences in the clinical outcome of the four groups of transporters in terms of death and transfer to other modes of replacement therapy. However, more low transporters were transplanted (P <.1 versus high transporters and P <.2 versus low-average transporters) and more lowaverage transporters stayed on peritoneal dialysis (P <.4 versus low transporters). There was no death directly related to inadequate dialysis. Ultrafiltration failure was the reason of transfer to hemodialysis in two patients. but high lymphatic absorption was dock umented to be the reason of ultrafiltration failure in one of them. The other patient with type I ultrafiltration failure was an initial high-average transporter with a final DIP creatinine of.822. Inadequate dialysis was the cause of transfer to hemodialysis in an initial low transporter, even though his final DIP creatinine increased to.538. DISUSSION Our study describes a selected population of patients who stayed on chronic peritoneal dialysis for at least 2 yr on average and had repeated PET. Transport changes in early dropouts are thus not included. Repeated PET studies, rather than routine studies, were done to assess possible changes in transport. Thus, this retrospective analysis of our clinical practice looks for transport changes in patients more likely to have had suspected changes. Further evaluations involving unselected chronic peritoneal dialysis patients are needed. Nevertheless, the pattern of change observed in different transport groups in our study is of interest. Our data showed that the peritoneal transport rates TABLE 2. The change in transport status in different initial transporters according to D/P creatinine#{176} Transporters According to Initial PET Final PET Tran sport Status hang e > I SD High High Avg Low Avg Low Increase Decrease HighN= 16 3 1 3 (%) 4(25%) High avg N = I 7 3 8 5 1 1 (6%) 2 (12%) Low avg N= 15 2 9 4 4 (27%) (%) Low N = 7 2 4 1 4 (57%) (%) TotalN=55 8 29 17 2 9 6 p <.3 for the change in the number of high-average transporters. by x2 test; P <.1 for the number of patients with an increase >1 SD between high and low transporters; and P<.3 for the number of patients with a decrease >1 SD between high and low-average transporters. by two-sided Fisher s exact test. TABLE 3. linical outcome of different groups of transporters High High Avg Low Avg Low I Total,N 16 17 15 7 Death, N 7 (44%) 6 (35%) 5 (33%) 1 (14%) NS hange to HD, N 4 (25%) 5 (29%) 1 (7%) 2 (28%) NS hange to IPD, N 8 (5%) 4 (24%) 4 (27%) 3 (43%) NS Transplanted, N (%) 2 (12%) (%) 3 (43%) Still on PD 5 (31%) 3 (18%) 9 (6%) I (14%) NS. not significant. b Between low and high transporters. P<.1; between low and low-average transporters. P<.2. Between low and low-average transporters, P <.4 by two-tailed Fisher s exact test. 1472 Volume 4. Number 7-1994

Lo et al of chronic peritoneal dialysis patients as indicated by PET are not static. They change in a fashion that has never been described before: the direction and the amplitude of change are dependent on their initial transport rate. The initial high transporters tend to have the transport rate decrease, and the low transporters tend to have the transport rate increase with time. Such migration toward the mean may help to explain why low clearances or low ultrafiltration with rapid transport are infrequent causes of APD technique failure (9). However, this centrifugal direction of change only occurred in the first 1 8 mo, and the transport rates tended to slowly increase thereafter, regardless of the initial transport status. On the other hand, despite changes in transport rates from initial extremes, the high transporters still have a higher transport rate than the others, and the low transporters have a lower transport rate. A simple methodobogic migration toward a mean seems unlikely because DIP creatinine and D/Do glucose changes were inversely correlated to a high degree, suggesting that actual transport changes and the mean change of DIP creatinine in the high. bow-average, and bow transporters were much higher than the coefficient of variation we analyed previously. In a previous study, we have already noted that patients with a longer duration of peritoneal dialysis were mostly high-average transporters. Our previous hypothesis was that this was because of a natural selection process: those with high-average transport stay on peritoneab dialysis. whereas those at either extreme drop out (1). In this study, we demonstrate that this phenomenon is also the result of changes toward average transport during the course of longterm peritoneal dialysis in the outlying groups. Increases in peritoneal transport rates with time have been observed by others (1, 1 1), but no change was found by some (8, 1 2). In all of the reported series, the change in DIP creatinine was not analyed according to the initial value. The high and low transporters were mixed together and studied as a group. The opposite directional changes of the transport rate in the first 1 8 mo of peritoneal dialysis in the outlying groups discovered in our analysis may explain the absence of overall changes in average values. hanges in different directions in some individuals have been noted, despite the lack of changes in a group (1, 8, 1 2), but no factors related to the change, such as age or number of peritonitis episodes, could be identified. Unfortunately, the initial DIP creatinine value was not included as one of the factors studied. Increases in transport rates were mainly reported in those series with a long follow-up (1. 1 1. 1 3). We also found that the DIP creatinine increases after the first 1 8 mo of peritoneal dialysis. regardless of the initial transport rate status or the initial direction of change. There have been some mechanisms proposed for this phenomenon, including an increase in peritoneal surface area and/or peritoneal permeability. Although it is unlikely that the peritoneal surface area will increase with dialysis. increases in peritoneal permeability may be explained by structural changes in the peritoneum and its vasculature (14, 1 5). The causes of the initial centrifugal changes in transport rate reported herein are also unknown. ould the initial high transport rates in some of the patients represent a transient reaction to catheter insertion and/or dialysate solution enhancing transport? Many initial tests were performed in the early months of peritoneal dialysis, but none fewer than 1 days after catheter insertion. Obviously, many of these patients were genuine high transporters to start with because most of their transport rates still stayed higher than the others after some time on peritoneal dialysis. The use of higher concentrations of glucose in the high transporters may play a part, but it is difficult to prove whether change relates to more hyperosmolar dialysate exposure. No correlation of the transport rate with age in adult peritoneal dialysis patients has been previously reported. However, it has been generally accepted that the transport rate is higher in children than in adults (16). Young children are often high or highaverage transporters (1 7). An age-related D/P creatinine level was described by Schroeder et a!. ( 1 8). The D/P creatinine ratio decreased with age but did not reach a significant difference between those younger and older than 3. In this report, we found a correlation of the Initial transport rate with age, whether the D/P creatinine or the D/Do is used to indicate the transport rate. However, the correlation with age disappeared with the final PET. This means that the age factor may be only applicable in the original, untouched peritoneum. As the duration of peritoneal dialysis increases, other factors may appear and the age factor no longer predominates. This may explain why our previous reports. and many other reports. did not find any age/transport rate relationship in the cross-sectional analyses with PET done at different durations of peritoneal dialysis. Similar to findings in other reports (1 ), the change in transport rate did not correlate with the number of peritonitis episodes or the peritonitis rate. This does not mean that peritonitis would not affect the transport rate. It is well established that peritonitis reduces the ultrafiltration acutely by increasing the permeability of the peritoneum (1 9. 2). In most cases, recovery occurred within a few weeks (2 1). The long-term effect is not known. but peritonitis may bead to a reduction in peritoneal surface area, resulting in a reduction in transport rate or sclerosing peritonitis with a marked decrease, or sometimes marked increase, in transport rate (22, 23). Because Journal of the American Society of Nephrology 1473

Peritoneal Transport in Repeated PET death or transfer to hemodialysis might result from a severe episode of peritonitis, the change in transport rate that might have occurred in these patients would never have a chance to be documented. In conclusion, the peritoneal transport of peritoneal dialysis patients as indicated by repeated PET is not always static. In our select group of patients with clinically suspected changes in transport, the transport rate tended to decrease in the high transporters and increase in the low transporters. This phenomenon may be beneficial to high and and low transporters by increasing ultrafiltration and increasing solute clearances, respectively. AKNOWLEDGMENTS Dr. W. K. La was sponsored by the Ho Hung-hiu Medical Education Foundation Fellowship to the Division of Nephrology. University of Missouri-olumbia. as a Research Fellow. REFERENES 1. Twardowski ZJ, Nolph KD, Khanna R, et al: Peritoneal equilibration test. Peritoneal Dial Bull 1987;7: 138-147. 2. Dia-Buxo JA: Peritoneal permeability in selecting peritoneal modalities. Perspect Peritoneal Dial 1988;5:6-1. 3. Twardowsld ZJ: linical value of standardied equilibration tests in APD patients. Blood Purif 1 989;7:95-18. 4. Wolf J, Poisky L, Ntoso KA, Koethe JD, Gerhardt RE: Adequacy of dialysis in APD and cycler PD: The PET is enough. Adv Peritoneal Dial 1992;6:28-21 1. 5. Twardowski ZJ: The fast peritoneal equilibration test. Semin Dial 199;3:141-142. 6. Twardowski ZJ, Noiph 1W, Khanna R, et at.: hoice of peritoneal dialysis regimen based on peritoneal transfer rates. Peritoneal Dial Bull 1987;7:579. 7. Dia-Buxo JA: Low peritoneal clearances-differential diagnosis and treatment. Adv Peritoneal Dial 19S9;5:31-35. 8. Davies SL, Brown B, Bryan J, Russel GI: linicab evaluation of the peritoneal equilibration test: A population-based study. Nephrol Dial Transplant 1 993;8:64-7. 9. Malorca R, Vonesh E, ancarini G, et al.: A six-year comparison of patient and technique survivals in APD and HD. Kidney mt 1988;34:518-524. 1. Procaccini PA, Querques M, Tappi A, Strippoli P: Peritoneal clearances: Long term study. Trans Am Soc Artif Intern Organs r988;34:437-44o. 1 1. Kush RD, Hallett MD, Ota K, et at.: Long term continuous ambulatory peritoneal dialysis: Mass transfer and nutritional and metabolic stability. Blood Purif 199;8:1-13. 12. Blake PG. Abraham G, Sombolos K, et at.: hanges in peritoneal membrane transport rates in patients on long term APD. Adv Peritoneal Dial 1989;5:3-7. 13. Passlick-Deetjen J, hlebowski H, Koch M, Grabensee B: hanges of peritoneal membrane function during long-term APD. Adv Peritoneal Dial 199;6:3-43. 1 4. Verger : Relationship between peritoneal membrane structure and its permeability: linical implications. Adv Peritoneal Dial 1 985; 1:87-95. 1 5. Dobbie JW: Morphology of the peritoneum in APD. Blood Purif 1989;7:74-85. 1 6. Belie JW, Vigneus A, Willumsen J, Hardy BE: The use of APD in the treatment of children with end-stage renal disease. Peritoneal Dial Bull 1981;1:25-28. 1 7. Greary DF, Harvey EA, MacMifian JH, Goodman Y, Scott M, Williamson BJ: The peritoneal equilibration test in children. Kidney Int 1992;42:12-15. 18. Schroeder H, van Dreumel M, Reddingius R, et at. : Peritoneal transport kinetics of glucose, urea and creatinine during infancy and childhood. Peritoneal Dial Int 1991 ; 11:322-325. 19. Rubin J, Ray R, Barnes T, Bower J: Peritoneal abnormalities during infection episodes of continuous ambulatory peritoneal dialysis. Nephron 1981;29:124-127. 2. Raja RM, Kramer MS. Rosenbaum JL, Boliasy, Krug M: ontrasting changes in solute transport and ultrafiltration with peritonitis in APD patients. Trans Am Soc Artif Intern Organs 1981 ;27:68-7. 2 1. Rubin J, Nolph K, Arfania D, Brown P. Prowant B: Follow-up of peritoneal clearances in patients undergoing continuous ambulatory peritoneal dialysis. Kidney Int 1 979; 16:619-623. 22. Slingeneyer A: Preliminary report on a cooperative international study on sclerosing peritonitis. ontrib Nephrol 1987;57:239-247. 23. Krediet RT, Struijik DG, Boeschoten EW, et at.: The time course of peritoneal transport kinetics in continuous ambulatory peritoneal dialysis patients who develops sclerosing peritonitis. Am J Kidney Dis 1989;13:299-37. 1474 Volume 4. Number 7. 1994