Peritoneal Dialysis International, Vol. 16, pp /96$ AUTOMATED PERITONEAL DIALYSIS

Transcription

1 Peritoneal Dialysis International, Vol. 16, pp /96$ Printed in Canada All rights reserved Copyright 1996 International Society for Peritoneal Dialysis RAPID DECLINE OF RESIDUAL RENAL FUNCTION IN PATIENTS ON AUTOMATED PERITONEAL DIALYSIS Kinya Hiroshige, Kougi Yuu, Masasuke Soejima, Masayuki Takasugi, and Akio Kuroiwa Second Department of Internal Medicine, School of Medicine, University of Occupational and Environmental Health, Kitakyushu City, Japan.Objective:To determine the effect of peritoneal dialysis modalities such as nightly intermittent peritoneal dialysis (NIPD), continuous cyclic peritoneal dialysis (CCPD), and continuous ambulatory peritoneal dialysis (CAPD) on residual renal function..design: A six-month prospective, nonrandomized comparison study..setting: Outpatient CAPD unit of a university hospital..participants: Eighteen end-stage renal disease patients treated by peritoneal dialysis (8 by NIPD, 5 by CCPD, and 5 by CAPD)..Interventions: Samples from the total dialysate, blood, and 24- hour urine collection were obtained monthly..measurements: Urea, creatinine, and beta2-microglobulin concentrations were measured. Renal and peritoneal clearances of each substance and KT/V urea were calculated. Residual renal function (RRF) was estimated by renal creatinine clearance (RCcr)..Results: No significant differences in age, sex, and primary renal disease among the three groups were noted. In all groups, anemic and hypertensive states were controlled identically, and mean weekly total (renal + peritoneal) KT/V urea (over 2.1/wk) and total creatinine clearance (over 60 L/wk/1.73 m2) were maintained during the whole experimental period. Starting mean RCcr was near 4.0 ml/min/1.73 m2 in all groups. Thereafter, a rapid and significant decline in RRF was demonstrated on NIPD and CCPD. The declining rates of RCcr values at 6 months after starting NIPD and CCPD were and ml/min/month, respectively, which were much greater than those of CAPD (+0.01 ml/min/month)..conclusion: Because of a possibly characteristic progressive loss of RRF in automated peritoneal dialysis (APD), strict regular assessment of RRF should be performed from the start of APD. KEY WORDS: Nightly intermittent peritoneal dialysis; continuous cycling peritoneal dialysis; renal creatinine clearance. Correspondence to: K. Hiroshige, Second Department of Internal Medicine, SchoolofMedicine, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahatanishi-ku, Kitakyushu City, Fukuoka 807, Japan. Received 24 November 1994; accepted 1 September R esidual renal function (RRF) in dialyzed patients is clinically important because of the addi tional factors of adequacy of peritoneal dialysis, less necessity for strict control of hydro electrolyte balance, maintenance of erythropoietin synthesis and conversion of vitamin D in its active form, and elimination of middle molecular substances, especially natural beta2- microglobulin (β2m) (1-4). Be cause of these benefits of RRF, it is important to preserve RRF as long as possible after the initiation of dialysis. The evidence that RRF is better preserved in patients on continuous ambulatory peritoneal dialysis (CAPD) than on an intermittent therapy hemodialysis (HD)was recently reviewed (5,6). Automated peritoneal dialysis (APD ), nightly intermittent peritoneal dialysis (NIPD ), or continuous cyclic peritoneal dialysis ( CCPD ), but especially NIPD, which might provide a better quality oflife, is considered an intermittent regimen. In CCPD patients, total fluid removal is performed only during the night, and the amount of small solute removal by a long daytime dwell is lower than that by intensive nighttime dialysis; thus, CCPD also seems to have intermittent characteristics. Investigation of the time course of RRF on APD will assist in determining the dialysis efficacy of APD and yield a better quality of life during the APD therapy. However, no data are available concerning the course of RRF in APD. The present study was prospectively undertaken in 13 patients treated with APD and 5 patients treated with CAPD from the start of dialysis therapy in our institute to assess the course of RRF over time and to compare the results in APD patients to those in CAPD patients. PATIEN TS AND METHODS PATIENTS At our center, between March 1992 and May 1994, 24 patients started chronic peritoneal dialysis. In the absence of specific clinical indications, the patients were allowed to choose the mode of dialysis therapy.

2 Once they chose the peritoneal dialysis regimen, they were allowed to choose the modality of peritoneal dialysis, after being fully informed of the advantages and the disadvantages of each. However, our conviction thatapd increases daytime activity, resulting in a better quality oflife, influenced some of the younger patients to choose APD. Our belief that CAPD is better tolerated by elderly patients and those with cardiovascular complications probably conditioned some of these patients to choose CAPD. NIPD is the method of choice for patients with very high and high peritoneal-membrane permeability areas, resulting in rapid glucose absorption and poor or no ultrafiltra tion on standard CAPD. However, in most of our patients at the start of dialysis, sufficient RRF was considered to easily bring minimum adequate levels of daily total (peritoneal plus renal) KTN urea and creatinine clearance (Ccr) recommended by others (7), which also allowed us to treat patients with APD from the start. One patient with abdominal discomfort and poor appetite, especially during the day, chose NIPD. Thus, 10 patients were initially treated with NIPD, 6 with CCPD, and 8 with CAPD. Two with NIPD and one with CCPD were excluded from this study because of an inadequate sample (especially 24hour urine) collection, as were 3 with CAPD because of an underlying disease such as ischemic heart disease, or cardiovalvular disease. Thus, 6 patients were excluded in total, leaving 8 with NIPD, 5 with CCPD, and 5 with CAPD included in this study. The clinical characteristics of these patients at the start of this study are shown in Table 1. There were no significant differences in age, sex, and primary renal disease among the three groups. Mean body weight in CAPD patients, which was lowest among the groups, was significantly lower than that in CCPD patients (p < 0.03). Total body water volume, calculated based on height, weight, sex, and age of the patient, using Watson's formula (8), was lower in CAPD patients, although not statistically significant when compared to APD patients. The ratio of dialy sate-to-plasma (D/P) creatinine and urea obtained from the standard peritoneal equilibration test was not significantly different in each dialysis regimen (9); however, CAPD patients showed a slight tendency to have higher values. None of the patients selected had severe cardiac, liver, or pulmonary dysfunction. All patients had renal parenchymal hypertension that was controlled with antihypertensive agents, except for an angiotensinconverting enzyme inhibitor (ACEI), during the study period. No patients were taking nonsteroi dal antiinflammatory drugs (NSAIDs) or diuretics. All of them were treated with recombinant human erythropoietin (rhuepo) from the start of dialysis to correct renal anemia. All patients were started on dialysis because of the appearance of uremic symp toms. Because none had marked fluid retention such as pulmonary edema at the start of dialysis, it was not necessary to force the rapid contraction of fluid and use the most hypertonic solution at the start (Table 2). Patients had no acute or chronic inflammatory diseases including peritonitis during the experimen tal period. DIALYSIS Dialysis prescriptions including cycler settings for nightly intensive dialysis at the start are shown in Table 2. In 8 patients, NIPD consisted of an 8 to 10hour nighttime dwell with 1.5 to 2.0 L of instilled volumes per exchange, using a P AC-X2 cycler (Baxter Healthcare Corp., Deerfield, IL). They performed four to six exchanges with each dwell time of70 to 120 minutes, according to the results of Kumano et al. (10), who showed the adequate frequency of exchange was five to six times for patients with urea D/P (2hour) ofo.6 using the kinetic model for urea removal capacity in 8-hour nightly APD. Nearly all of our patients had lower peritoneal transport properties with 0.61 ofmeand/p(2-hour) urea in NIPD and 0.63 in CCPD (see Table 1). The 5 patients on CCPD were treated with 8 to 10 hours of NIPD (four to six exchanges) plus a daytime dwell of 1.3 to 1.5 L. Five CAPD patients performed four to five exchanges per day, routinely using 1.0 to 2.0-L dialysate. Daily total dialysate volume was significantly lower in CAPD. All patients were treated using dialysate with 1.5% concentration of glucose with or without dialysate with 2.5% concentration of glucose (Table 2). EXPERIMENTAL PROCEDURES Samples from the total dialysate includingultrafiltration, blood, and 24-hour urine collection were obtained from each patient approximately monthly over 6 months to measure urea nitrogen, creatinine, and β2m. Blood samples were taken 12 hours after the midpoint of nightly dialysis in APD patients. Urea nitrogen and creatinine were measured using a Hitachi auto analyzer (Hitachi Inc., Tokyo, Japan).β2m was measured by latex aggregation reaction using an LX reagent (Eiken Co., Tokyo, Japan). The total (renal plus peritoneal) clearances of urea (C urea ), creatinine, andβ2m (Cβ2m) were calculated monthly. Clearances of each substance were calculated accord ing to the standard formula. All clearances were normalized to 1.73 m2 of the body surface area and shown as L/l.73 m2/wk. RRF measurements consisted of 24-hour Ccr measurements adjusted to the standard body surface area. TotalKT/V urea was also calculated monthly. The weeklykt/v urea was calcu

3 lated as the weekl y C urea total body water, as previously described (11). divided Blood hemoglobin b y con centration was also determined monthly. Mean blood pressure was evaluated with the weekly average of mean blood pressure obtained from patients' home records. Starting values were obtained within 2 weeks of the initiation of dialysis. These variables were then chronologically evaluated and compared among the groups. Comparisons of the degree of decline in renal creatinine clearance (RCcr) over a 6-month period on APD to those on CAPD were evaluated. STATISTICS The differences of unpaired observation between two groups were analyzed using Student's t-test. Significance was accepted for p < RESULTS Figure 1 shows the chronological data of mean blood pressure, body weight, and blood hemoglobin concentration in each group over 6 months. Blood hemoglobin concentration and blood pressure were maintained in each group, and no significant differences were noted in any group at any time of observation. Although body weight in the CAPD group was significantly lower than in the CCPD group, there was no significant change in body weight with time in any of the groups. Chronological data including clearances and urea kinetics are also illustrated in Figure 2. There were no significant differences in RCcr among the groups at the start. The RCcr significantly decreased 3 to 4 months after the start of APD, concomitant with the loss of daily urine volume. The declining rate ofrccrfrom the basal value in the CCPD group was significantly greater after 3 months when compared to the CAPD group and after 5 months when compared to the NIPD group (data not shown). The declining rates ofrccr at 6 months were ml/min/month on NIPD and 0.34 on CCPD, which were higher than that ofcapd ( ml/min/month). In APD patients, daily ultrafiltration volumes increased over time by increasing the use of dialysate with 2.5% glucose concentration compensating for the loss of urine volume, although most hypertonic solutions of dialysate with 4.25% glucose concentration were never used during the whole period. N one

4 had apparent evidence of overhydration revealed by either chest x-ray films or physical examination at any time. However, there were no significant differences in daily ultrafiltration volumes among the three groups over time. In CCPD patients, total ultrafiltration volumes per day were mainly dependent on nightly dialysis on CCPD. In CAPD patients without loss of RRF, total dialysate volumes and numbers of hypertonic solutions (2.5% glucose concentration) used did not change during the study period. As for the total elimination of solutes during the whole experimental period, total Ccr and KT/V urea tended to be slightly higher in CAPD patients. Total KTN urea over 2.1 and total Ccrover60 L/l. 73 m2/wk were obtained on APD, as on CAPD (see Figure 2), by increasing the dose of dialysate used in APD therapy, despite the loss of RRF. There were no significant differences in these values among the three groups during the study period. Actual values of serum concentrations and β2m clearances for the kidney and peritoneum are shown in Table 3. Higher levels of serum creatinine were observed in APD patients, probably because of their greater body weight. Serum β2m concentration was almost similar in all groups at the start. However, a greater increase in serum β2m concentration with time was observed in APD patients, especially those on CCPD, because of the apparent reduction in renal Cβ2m. CAPD patients had sufficient total β2m elimination throughout the study period because of the preservation of renal excretion ofβ2m. At the start of dialysis, total elimination ofβ2m mainly depended on renal excretion ofβ2m, with the ratio ofrenal to total excretion ofβ2m being 0.87,0. 74, and 0.69 on NIPD, CCPD, and CAPD, respectively. Six months after the start of dialysis, over half or the total elimination of

5 β2m still depended on renal excretion, even in APD patients with rapid decline in RRF, and the ratios on NIPD, CCPD, and CAPD were 0.71,0.47, and 0.66, respectively. DISCUSSION the avera g e of Ccr and C has been sug g When evaluating the RRF in end-stage renal disease, it can be argued that Ccr may be an unsatisfactory estimate of RRF as glomerular filtration rate (GFR) because of an inadequate increase in tubular secretion of creatinine. Comparative studies show that Ccr overestimates GFR, when compared to inulin and iodothalamate clearances, by about 0.5 ml/min when the GFR is 3 ml/min (12). Although

6 urea ested as a more accurate measurement of RRF, this combined method seems to underestimate GFR, when compared to inulin clearance, by about 0.5 ml/min at an inulin clearance of 3 ml/min; therefore it offers no advantage over the use ofccr alone (13). Milutinovic et al. indicated a reasonable correlation between Ccr and inulin clearances (13). Furthermore, many published studies of RRF after the onset of maintenance therapy have estimated RRF from RCcr (5,14-18). It can thus be acknowledged that the method used in the present study is acceptable, although, admittedly, Ccr does not exactly reflect GFR. Several benefits of RRF in dialyzed patients have been proposed by many researchers (1-4). RRF does

7 playa major role in the daily excretion of middle or large molecules, as proven by β2m elimination in our patients. Even after significant loss of RRF, β2m excretion depended mainly on RRF. Furthermore, RRF is an important additional factor for adequate dialysis (1). At the early phase of the study, sufficient dialysis, defined by total Ccr and KTN urea (7), was achieved in APD patients, despite their lower peritoneal transport properties because of sufficient RRF. However, patients with a lower peritoneal function treated by APD would easily develop underdialysis if RRF rapidly declined. We demonstrated the substantial and rapid decline in RRF within 6 months on APD. The mean rate of decline in RCcr at 6 months was apparently higher in our APD patients than in both our CAPD patients and in CAPD patients reported elsewhere (5,14-18). The declining rates ofrrf in NIPD are almost similar to or slightly greater than those of hemodialysis previously reported (Table 4). Unfavorable conditions such as severe cardiopulmonary failure, marked anemia, severe malnutrition, severe infection, and malignant hypertension were never seen in our patients, and medications such as diuretics, ACEI, and NSAIDs, which possibly affect RRF, were never used during the experimental period. Furthermore, none of our patients had marked fluid contraction caused by the use of most hypertonic solutions. The possible additional contributing factors to RRF such as age, sex, primary renal disease, and mean blood pressure should also be considered. Iestet al. reported that sex, age, and mean blood pressure at the start of hemodialysis did not contribute to the changes in RRF (14). Cancarini et al. suggested that persistent pathogenetic factors of glomerulonephritis might be a main cause of aggravated renal damage, showing the same rate of deterioration of RRF in both hemodialysis and CAPD (17). In our study, all groups were similar in age, sex, primary renal

8

9 disease, and the control of blood hemoglobin concentration and blood pressure. With regard to the primary renal disease of glomerulonephritis, some types will progress faster, promoting rapid decline in RRF even after the start of dialysis, and some types progress slowly. In 6 NIPD, 3 CCPD, and 3 CAPD patients, we could examine the declining rate in renal Ccr for at least 24 months before the start of dialysis, and found that the mean declining rate was 0.72,0.87, and 0.93 ml/mo nth, in NIPD, CCPD, and CAPD, respectively. Although the number of patients was too small to analyze the difference among the three dialysis regimens statis - tically, the declining speed ofrrf before the initia tion of dialysis would be fastest in CAPD patients. The previous review showing that CAPD could recover RRF also allowed us to confirm our previous conclusion (5,6). With the adequate increase in dialysate volume on APD, the minimum values of total KTN urea over 2.1 and total Ccr over 60 L/l.73 m2/wk were obtained, even 6 months after the start of APD. None of the patients had uremic symptoms or an overhydrated state derived from underdialysis. Frequent episodes of hypotension at the hemodialysis session, which would contribute to the decline of RRF in hemodialy sis patients, were never observed in peritoneal dialy sis (18). We could remove many of the annoying factors possibly affecting RRF and could guess that acute changes in volume state and osmotic load induced at each nightly peritoneal dialysis session, on account of the intermittent character of this therapy, would possibly result in accelerating the deterioration of renal function, although our population sizes were too small to confirm this result. The intermit tency of hemodialysis associated with rapid changes in solute concentration and body fluid volume during thrice-weekly dialysis sessions is greater than that of APD, changes being less rapid during nightly dialysis sessions. However, the fact that the reduction in RRF on APD was similar to that on hemodialysis reported in the literature allowed us to guess that the intermittency itself, not the dose of intermittent removal of solutes and body fluid, plays an important role in the reduction in RRF. Several physiological mechanisms can be invoked to explain better retention of RRF with CAPD than withapd (5,15,19). The first is avoiding the possible ischemia that occurs because of the rapid changes of osmolality and the concomitant contraction of fluid volume during nightly intensive dialysis. The second is hydrodynamic, and relates to the lower time averaged value of the preglomerular arterial pressure in CAPD patients. A third possibility is the lower average value of protein intake of CAPD patients, whose mean age was the highest of the three groups, although not statistically significant. However, other pathophysiological factors may playa role in the early decline in RRF. The operative mechanisms are also unlikely to be sorted out rapidly. The reasons for the rapid decline in RRF in hemodialysis patients is partly because hemodialysis itself may be nephrotoxic; that is, repeatedly exposing the natural kidney to the inflammatory mediators generated by extra corporeal circulation may compromise the organ, and these patients are repetitively exposed to systemic anticoagulation such as heparin (20 ). However, these factors are unlikely to be major factors for the decline in RRF of hemodialysis patients, because rapidly declining rates of RRF of our APD patients were almost equal to that of HD patients reported elsewhere. Consequently, in our NIPD and CCPD patients with lower peritoneal transport properties, a monthly, strict assessment for RRF is necessary to obtain adequate dialysis, because patients could easily develop the underdialysis state after the significant decline in RRF. Continuous therapy such as CAPD is considered the best therapy for the preservation of RRF, even if daytime activity is disrupted by the daytime dialysate exchange. REFERENCES 1. Hallett MD, Charlton B, Farrell PC. Adequacy ofperi toneal dialysis. Semin Dial 1990; 3: Jongen MJM, Van Der Vijgh WSF, Lip P, Netelenbos JC. Measurements of vitamin D metabolites in anephric subjects. Nephron 1984; 36: Amici G, Virga G, Da Rin G, et al. Serum beta-2 microglobulin level and residual renal function in perito neal dialysis. Nephron 1993; 65: Lynn RI, Feinfeld DA. Importance of residual renal function in end-stage renal disease. Semin Dial 1989; 2: RottembourgJ. Residual renal function and recovery of renal function in patients treated by CAPD.Kidney Int 1993; 43:S Nolph KD. Is residual renal function preserved better with CAPD than with hemodialysis? Am Kidney Fund Nephrol Lett 1990; 7: Keshaviah P. Urea kinetic and middle molecule ap proaches to assessing the adequacy of hemodialysis and CAPD. Kidney Int 1993; 43:S Watson PE, Watson ID, Batt RD. Total body water volumes for adult males and females estimated from anthropometric measurements. Am J Clin Nutr 1980; 33: Twardowski ZJ. Clinical value of standardized equilibration tests in CAPD patients. Blood Purif 1989; 7: KumanoK,SakaiT, YamashitaA.Adequatevolumeof infusion and frequency of dialysate exchange on automated peritoneal dialysis. Jpn J Artif Organs 1992;

10 21: Tattersall JE, Doyle S, Greenwood RN, Farrington K. Kinetic modelling and underdialysis in CAPD patients. Nephrol Dial Transplant 1993; 8: Skov PE. GFR in patients with severe and very severe renal insufficiency.acta Med Scand 1970; 187: MilutinovicJ, Cutler RE, Hoover P, MeusenB, Scribner BH. Measurement of glomerular filtration rate in patients receiving repetitive hemodialysis. Kidney Int 1975; 8: lest CG, Vanholder RC, Ringoir SM. Loss of residual renal function in patients on regular haemodialysis.int J ArtifOrgans 1989; 12: Lysaght MJ, Vonesh EF, Gotch F,et al. The influence of dialysis treatment modality on the decline of remaining renal function. Trans Am Soc Artif Intern Organs 1991; 37: Rottembourg J, Issad B, Poignet JL, et al. Residual renal function and control of blood glucose levels in insulindependent diabetic patients treated by CAPD. In: Keen H, Legrain M, eds. Prevention and treatment of diabetic nephropathy. Boston: Lancaster M. T.P.Press, 1986: Cancarini GC, Brunori G, Camerini C, et al. Renal function recovery and maintenance of residual diuresis in CAPD and hemodialysis. Perit Dial Bull 1986; 6: Hallett M, Owen J, Becker G, Stewart J, Farrell PC. Maintenance of residual renal function: CAPD versus HD (Abstract). Perit Dial Int 1992; 12(Suppl):S Conger JD. Does hemodialysis delay recovery from acute renal failure? Semin Dial 1990; 3: Cheung AK. Biocompatibility of hemodialysis mem branes. J Am Soc Nephrol1990; 1: