Adequacy of automated peritoneal dialysis with and without manual daytime exchange: A randomized controlled trial

http://www.kidney-international.org & 2006 International Society of Nephrology original article Adequacy of automated peritoneal dialysis with and without manual daytime exchange: A randomized controlled trial D Demetriou 1, A Habicht 1, M Schillinger 2,WHHörl 1 and A Vychytil 1 1 Division of Nephrology and Dialysis, Department of Medicine III, Medical University Vienna, Vienna, Austria and 2 Division of Angiology, Department of Medicine II, Medical University Vienna, Vienna, Austria Until now, it remains unclear whether the addition of manual daytime exchanges or increasing the nightly dialysate flow is the best strategy to optimize automated peritoneal dialysis (APD) treatment. In this open-label randomized controlled crossover trial, 18 patients with high-average (HA) or low-average (LA) peritoneal transport rates sequentially underwent two different APD regimens for 7 days each, with an intermittent washout period of 7 days. Manual exchange treatment was a conventional APD with low nightly dialysate flow and one manual daytime exchange. High-flow treatment was defined by cycler therapy with high dialysate flow but without manual daytime exchange. Creatinine clearances (8.5671.22 vs 7.8771.04 l/treatment, P ¼ 0.011) and urea nitrogen clearances (12.8371.98 vs 11.6871.06 l/ treatment, P ¼ 0.014) were significantly increased during high-flow treatment compared to manual exchange treatment. Sodium removal was significantly lower and glucose absorption was higher with the high-flow regimen. Phosphate clearances, b 2 -microglobulin clearances, ultrafiltration, and peritoneal protein loss were not different between the two treatment modalities. Subgroup analysis dependent on peritoneal transport types showed that the effect on clearances was most marked and significant in HA transporters, whereas sodium removal was lowest in LA transporters. We conclude that small solute clearances can be significantly improved and middle molecule clearances maintained in APD patients by increasing the nightly dialysate flow instead of adding a manual daytime exchange. However, the possible benefit of better clearances with higher nightly treatment volumes has to be weighed against increased costs and the possible negative impact of impaired sodium removal, especially in LA transporters. Kidney International (2006) 70, 1649 1655. doi:10.1038/sj.ki.5001808; published online 6 September 2006 KEYWORDS: dialysate flow; phosphate; b 2 -microglobulin; sodium; protein loss Correspondence: A Vychytil, Division of Nephrology and Dialysis, Department of Medicine III, Medical University Vienna, Währinger Gürtel 18-20, Vienna A-1090, Austria. E-mail: andreas.vychytil@meduniwien.ac.at Received 17 January 2006; revised 3 May 2006; accepted 5 July 2006; published online 6 September 2006 Automated peritoneal dialysis (APD) is a favorable option to improve peritoneal clearances. However, a continuous APD treatment (peritoneal cavity filled with dialysate between cycler sessions, with or without manual exchanges) rather than intermittent cycler therapy (peritoneal cavity empty between treatments) is preferred in these patients. Usually in patients on continuous cyclic peritoneal dialysis, a rather low dialysate flow of 0.8 1./h is used to optimize clearances, corresponding to nightly dialysate treatment volumes of 7 1 during a 8 10 h cycler session. 1,2 However, when using a low cycler treatment volume, manual daytime exchanges are often required to achieve adequate clearances, at least for patients with low residual renal function. 2 It should be considered that manual daytime exchanges impair quality of life and may also increase the risk of peritonitis. Furthermore, the question has been raised whether the recommended low dialysate volumes indeed maximally utilize the treatment options of APD. There is evidence in the literature that increases of dialysate flow up to 2 and 3 l/h improve small solute clearances as compared to lower treatment volumes. 3,4 To date, no study has opposed such newer APD treatment options to the classical strategy of optimizing dwell times with a lesser number of cycles and two daytime dwells. Furthermore, most recent studies have focused on removal of small solutes. However, when increasing the nightly dialysate flow, clearances of larger uremic toxins and sodium removal may be impaired. In the present randomized controlled crossover trial, we investigated whether a cycler therapy (tidal peritoneal dialysis) with high dialysate flow during night time and one daytime dwell (using icodextrin) provides at least similar small solute clearances, middle molecule clearances, sodium removal, peritoneal ultrafiltration, and protein loss as compared to a conventional APD treatment with low dialysate flow and one manual daytime exchange (one daytime dwell with icodextrin, followed by one daytime dwell with glucose solution) (Figure 1). RESULTS Patient characteristics and flow of participants Of the 49 APD patients screened for inclusion in the study, 22 were randomized (Figure 2). There were four dropouts Kidney International (2006) 70, 1649 1655 1649

o r i g i n a l a r t i c l e D Demetriou et al.: Adequacy of high-flow APD 'Manual exchange' treatment Fill volume BSA < 1.7 m 2 BSA 1.7 2.0 m 2 BSA > 2.0 m 2 2.5 l (tidal volume 75%) 2.5 l (tidal volume 75%) 3 l (tidal volume 75%) ICO G G 0 10 15 24 Hours 'High flow' treatment Fill volume BSA < 1.7 m 2 BSA 1.7 2.0 m 2 BSA > 2.0 m 2 (tidal volume 75%) 2.5 l (tidal volume 75%) 3 l (tidal volume 75%) ICO G 0 10 15 24 Hours Figure 1 Treatment regimens used in this randomized controlled crossover trial. BSA, body surface area; G, glucose; ICO, icodextrin; L, liters. during the study period. One patient received a kidney graft on the first day of treatment ( high-flow treatment), and one patient declined to continue the study on day 3 ( manual exchange treatment) without giving reasons for this decision. One patient was withdrawn on day 4 ( high-flow treatment) because of dehydration and hypotension owing to massive diarrhea, which occurred in combination with the continuous ultrafiltration. The reason for dropout in the fourth patient (day 3, manual exchange treatment) was intolerable abdominal distension. This latter patient had performed nightly intermittent peritoneal dialysis before the start of the study. Therefore, 18 APD patients were included in the final analysis (Figure 2). The baseline demographic and clinical characteristics of these patients are shown in Table 1. Causes for end-stage renal disease were chronic glomerulonephritis (n ¼ 5), chronic kidney failure of unknown origin (n ¼ 4), chronic interstitial nephritis (n ¼ 2), polycystic kidney disease (n ¼ 2), and miscellaneous diseases (n ¼ 5). Although diabetes was not an exclusion criterion, no diabetic patient was included in this study. Comparison of end points between the two treatments in the whole-patient group Primary end points. The average creatinine clearances were 7.8771.04 l/treatment for manual exchange treatment compared to 8.5671.2/treatment for high-flow treatment (P ¼ 0.011), demonstrating a significantly increased creatinine clearance when elevating the dialysate flow (Table 2). Average urea nitrogen clearances were 11.6871.06 l/treatment compared to 12.8371.98 l/treatment for the manual exchange and high-flow regimens, respectively (P ¼ 0.014). Secondary end points. Phosphate and b 2 -microglobulin clearances were not different between the two treatment modalities. Peritoneal sodium removal was significantly less with the high-flow treatment as compared to manual exchange treatment (Table 2). In contrast, peritoneal ultrafiltration and glucose absorption were slightly greater with the high-flow regimen (only significant for glucose absorption). There was no difference in peritoneal protein loss between both treatment modalities. Subgroup analyses Patients with body surface area 1.7 2 m 2. Analyzing the subgroup of patients with body surface area between 1.7 and 2m 2 (n ¼ 12) yielded virtually identical results to those in the entire study population. Creatinine clearances (P ¼ 0.050) and urea nitrogen clearances (P ¼ 0.040) were significantly increased after high-flow treatment compared to manual exchange treatment. Sodium removal tended to decrease (P ¼ 0.064), whereas peritoneal ultrafiltration (P ¼ 0.063) 1650 Kidney International (2006) 70, 1649 1655

D Demetriou et al.: Adequacy of high-flow APD o r i g i n a l a r t i c l e and glucose absorption (P ¼ 0.081) tended to increase with the high-flow regimen. Phosphate clearance (P ¼ 0.82), b 2 -microglobulin clearance (P ¼ 0.75), and protein loss (P ¼ 0.48) did not differ between the groups. Comparison of end points within the subgroups of highaverage and low-average transporters. To account for differences between the two transport subtypes, we separately analyzed patients categorized as high-average (HA) transporter (dialysate to plasma ratio (D/P) of creatinine at 4hX0.65, n ¼ 8) or low-average (LA) transporter (D/P of Allocated to 'manual exchange' treatment, then 'high flow' treatment (n=11) Lost to follow-up (n=0) Discontinued intervention: abdominal distension (n=1), patients decision (n=1) Analyzed (n=9) Assessed for eligibility (n=49) Randomized (n=22) Excluded (n=27) Not meeting inclusion criteria (n=15) Refused to participate (n=9) Included in other studies (n=3) Allocated to 'high flow' treatment then 'manual exchange' treatment (n=11) Figure 2 Flow of participants and study design. Lost to follow-up (n=0) Discontinued intervention: renal transplantation (n=1) hypotension (n=1) Analyzed (n=9) creatinine at 4 ho0.65, n ¼ 10) with respect to the primary and secondary study end points. In HA patients, we found consistent results as in the entire cohort with significantly higher creatinine clearances (9.3271.10 vs 8.2671.2/treatment, P ¼ 0.043) and urea nitrogen clearances (14.3871.45 vs 12.2570.78 l/treatment, P ¼ 0.013) after high-flow treatment as compared to manual exchange treatment. The difference in mean small solute clearances found in HA transporters corresponds to an increase of creatinine clearance of approximately 7 l/week when changing from manual exchange treatment to the high-flow regimen. There were no significant differences in the other end points between the two treatment modalities, except for a higher peritoneal glucose absorption during high-flow treatment (227.6786.3 vs 170.9736.9 g/treatment, P ¼ 0.022). In LA patients, sodium removal was significantly lower with the high-flow regimen as compared to manual exchange treatment (107.6767.4 vs 155.8794.3 mmol/treatment, P ¼ 0.009). We observed no significant differences in primary end points (creatinine clearance, P ¼ 0.15; urea nitrogen clearance, P ¼ 0.42) nor in any other secondary end points. However, in seven of the 10 LA patients, peritoneal creatinine and urea clearances increased when elevating the dialysate flow (data not shown). Cost analysis of treatments As shown in Table 3, the high-flow regimen was associated with a 34 59% increase of costs as compared to manual exchange treatment. Table 1 Demographic and clinical characteristics of the 18 APD patients included in the final analysis Age (years) a 53.4714.4 Gender (female/male) 5/13 Duration of dialysis (months) a 15.078.8 Body surface area (m 2 ) a 1.8570.18 Body surface area (o1.7/1.7 2/42m 2 ) 3/12/3 Body mass index (kg/m 2 ) a 23.973.1 D/P creatinine at 4 h a 0.6570.09 Peritoneal transport type (HA/LA) 8/10 APD, automated peritoneal dialysis; D/P, dialysate to plasma ratio; HA, high-average peritoneal transporters; LA, low-average peritoneal transporters. a Mean7s.d. Table 2 Comparison of peritoneal clearances, peritoneal protein loss, sodium removal, peritoneal ultrafiltration, and peritoneal glucose absorption after two different APD treatment regimens, whole-patient group (n=18) Manual exchange treatment High flow treatment P-value Primary end points Creatinine clearance (l/treatment) 7.8771.04 8.5671.22 0.011 Urea nitrogen clearance (l/treatment) 11.6871.06 12.8371.98 0.014 Secondary end points Phosphate clearance (l/treatment) 7.6071.99 7.7471.74 0.68 b 2 -microglobulin clearance (l/treatment) 1.1270.41 1.1070.37 0.78 Sodium removal (mmol/treatment) 176.4777.7 128.0777.8 0.010 Peritoneal ultrafiltration (ml/treatment) 2051.17683.5 2215.67598.2 0.10 Glucose absorption (g/treatment) 159.5729.2 189.4774.1 0.046 Protein loss (g/treatment) 4.9571.40 4.3771.71 0.17 APD, automated peritoneal dialysis. Kidney International (2006) 70, 1649 1655 1651

o r i g i n a l a r t i c l e D Demetriou et al.: Adequacy of high-flow APD Table 3 Comparison of treatment costs (only including dialysis solutions and tubing) Body surface area (m 2 ) Manual exchange treatment (h per year) High flow treatment (h per year) Increase in costs (%) o1.7 31 000.3 41 625.5 34.3 1.7 2.0 31 000.3 48 184.7 55.4 42 34 334.5 54 744.0 59.4 Adverse events Adverse events in the 18 APD patients included in the final analysis were hypokalemia (n ¼ 1) and hypotension (n ¼ 1), which required an increase of oral fluid intake and reduction of antihypertensive agents. Both events occurred during high-flow treatment. One patient complained of moderate abdominal distension during both study weeks. This patient had only a daytime fill volume of 1 l before the start of the trial. Including the patient with severe abdominal distension during manual exchange treatment and the patient with diarrhea and hypotension during high-flow treatment, who both dropped out, there was no remarkable difference in the number of adverse events between the two APD regimens. DISCUSSION Until now, it remains unclear whether addition of manual daytime exchanges or increasing the dialysate flow during night time is the best strategy to optimize APD treatment. Durand et al. 5 described that the diffusion of uremic toxins through the peritoneal membrane resembles a parabola with a maximal point. If dialysate flow is increased above this point, clearances may not increase further or may even decrease, as the time spent for inflow and outflow becomes a significant part of the entire treatment period, which is less effective for diffusion. Although these data are important, they are only based on six patients. Therefore, a final conclusion is not available based on these results. Accordingly, the maximal effective dialysate flow for small solute clearances differs between several clinical studies. In nonrandomized trials, increases in dialysate flow from 1.1 and 1.7 l/h to 2 and 3 l/h have been reported to improve small solute clearances in APD patients. 3,4 In contrast, Amici et al. 6 found significantly higher K t /V values but unchanged creatinine clearances in patients with 9-h continuous tidal peritoneal dialysis (including two daytime dwells) after elevating the cycler treatment volume from 17 to 22, 27, and 3. Based on kinetic modeling, Blake et al. 7 showed that in continuous cyclic peritoneal dialysis patients with a cycler treatment volume of 15 l, addition of a manual daytime exchange is superior to further increases in nightly dialysate volumes. Despite a non-inferiority design in the present randomized controlled clinical trial, we found significantly higher peritoneal creatinine and urea clearances when using a high dialysate flow as compared to a conventional low-flow treatment regimen with one manual daytime exchange. Including only patients with a body surface area of 1.7 2.0 m 2 in the analysis yielded consistent results. A sub-analysis of LA and HA transporters showed that this effect on clearances was most marked in the latter patient group, but smaller and not significant in LA transporters. In contrast to peritoneal creatinine clearance, phosphate clearance did not significantly differ between the two treatments in the entire patient group nor in the subgroups of LA and HA transporters. This is in contrast to findings of Sedlacek et al., 8 who found a strong relationship between creatinine and phosphate clearance in both continuous ambulatory peritoneal dialysis and continuous cyclic peritoneal dialysis patients. However, in a recent trial, improvement of clearances in APD was markedly less for phosphate than for creatinine when nightly treatment volume was increased by 71%. 9 Further studies are needed to find an explanation for these conflicting findings. Although peritoneal small solute clearances are associated with patient survival in some studies, 10,11 they have no influence on mortality in other recent trials. 12,13 This points to the impact of other factors contributing to clinical outcome in peritoneal dialysis (PD) patients. Most importantly, there is a question as to whether the price for improving small solute clearances by high dialysate flow may be a decreased removal of larger uremic toxins and sodium. Because of their high molecular weight, middle molecules and small proteins need longer dwell times to diffuse from peritoneal capillaries into the peritoneal cavity. Equilibrium between blood and dialysate is not reached for these toxins even after a dwell time of more than 10 h. Keshaviah 14 showed that middle molecular clearances in continuous ambulatory peritoneal dialysis patients are only marginally influenced by the number of dialysate exchanges. Accordingly, it can be expected that the addition of one or more manual daytime exchanges in continuous cyclic peritoneal dialysis patients instead of a long daytime dwell will not largely improve removal of these substances. However, these data are based on computer modeling, not on actual clearance measurements. In contrast to the number of manual dialysate exchanges, the overall contact time between peritoneal membrane and dialysate may have a more important influence on diffusion of larger molecules. Based on the concept of Durand et al., 5 the maximal effective dialysate flow of middle molecules is reached earlier than for small solutes, possibly leading to a decrease of clearances of these toxins with high nightly dialysate treatment volumes. Data in patients treated by intermittent APD, however, are controversial, showing a decrease of vitamin B 12 clearances after elevating the dialysate flow from 2 to 4 l/h 15 as well as no influence of dialysate flow on b 2 -microglobulin clearances. 3 In our trial, both continuous APD treatments ( manual exchange and high flow ) were equivalent with respect to b 2 -microglobulin clearances, even in LA transporters. b 2 -Microglobulin clearances found in the present study were twice as high as in our previous study on intermittent cycler therapy, 3 confirming that it is not dialysate flow or number of manual exchanges but continuous contact of the peritoneum with dialysate that is 1652 Kidney International (2006) 70, 1649 1655

D Demetriou et al.: Adequacy of high-flow APD o r i g i n a l a r t i c l e the most important determinant of b 2 -microglobulin removal in APD patients. In prospective cohort studies, both sodium and fluid removal were identified as independent predictors of survival in PD patients. 2,16 Sodium removal in APD is lower as compared to continuous ambulatory peritoneal dialysis patients because of sodium sieving at the peritoneal membrane, which occurs during short dwell times. 17 20 Our finding of significantly higher sodium removal in manual exchange as compared to high flow treatment is in agreement with a cross-sectional study that has reported that, after controlling for ultrafiltration, inclusion of a daytime exchange and longer dwell times were independent predictors of improved peritoneal sodium elimination. 19 Furthermore, sodium sieving during high flow treatment becomes most marked in patients with LA transport rates. In contrast to sodium removal, peritoneal ultrafiltration in the present study was slightly but not significantly increased with high dialysate flow. We calculated only peritoneal glucose absorption, which was higher with increasing dialysate flow. However, because of its longer daytime dwell during high flow treatment, carbohydrate absorption derived from icodextrin may be additionally greater than after manual exchange treatment, most likely increasing the difference in caloric load between the two APD regimens. Peritoneal protein loss during PD is significant and may contribute to malnutrition in these patients. 21 We observed no difference in protein loss between high-flow treatment and the manual exchange treatment in our study. This finding extends observations by Twardowski et al., 22 Perez et al., 4 and also by our own group 3 showing that not only during intermittent APD but also with continuous cycler treatment, increase of dialysate flow has no influence on peritoneal protein loss. What are the practical consequences of our study? The lack of necessity of manual daytime exchanges may have several advantages. The first and most apparent advantage is the improvement of quality of life, as manual exchanges decrease daytime flexibility and may be even more problematic for PD patients who work or who require assistance from another person. Secondly, the peritonitis risk increases with the number of manual dialysate exchanges. 23 Therefore, a high-flow regimen as performed in this study may decrease the risk of infection as compared to APD treatment modalities including manual daytime exchanges. However, data on infection rates between APD patients and continuous ambulatory peritoneal dialysis patients, who also perform more connections/disconnections at the catheter, are controversial. 24,25 Particularly in LA patients, possible advantages of increased small solute clearances during high-flow treatment may be eclipsed by the significant decrease in sodium removal, which independently of ultrafiltration influences clinical outcome. Furthermore, the higher cost of high-flow APD may hinder a more liberal application of this treatment modality. The contact of the peritoneal membrane with larger amounts of fresh dialysate when changing from the conventional to the high-flow regimen may be another point of concern. Although clinical consequences are discussed controversially, 26,27 the use of more biocompatible solutions is recommended in this situation. In conclusion, small solute clearances can be significantly improved and middle molecule clearances maintained in APD patients by increasing the nightly dialysate flow instead of adding a manual daytime exchange. However, the possible benefit of better small solute clearances with higher nightly treatment volumes has to be weighed against increased costs and the possible negative impact of impaired sodium removal. These latter aspects have to be especially considered in LA patients in whom the difference in clearances was smaller but in whom sodium sieving was more marked during high-flow treatment than in HA patients. MATERIALS AND METHODS Participants This open-label randomized controlled crossover trial was performed at the Medical University of Vienna. We included patients older than 18 years and in a clinically stable condition who were treated with APD for at least 3 months. None of the patients had experienced an episode of peritonitis within 2 months before the study began. Only patients with average peritoneal transport rates (HA or LA) were included. Exclusion criteria were participation in another study within 3 weeks before the study began, allergy to icodextrin, acute impairment of general condition, and severe chronic heart failure (class NYHA III or IV). The Ethical Review Board at the University of Vienna approved the study. All patients gave written informed consent. Study design The study period lasted 3 weeks. All patients sequentially underwent two different continuous APD regimens ( manual exchange treatment and high-flow treatment) for 7 days each (week 1 and week 3) with an intermittent washout period of 7 days. Manual exchange treatment represents a conventional APD therapy with low nightly dialysate flow and one manual daytime exchange (two daytime dwells) (Figure 1). High-flow treatment represents cycler therapy with high dialysate flow and without manual daytime exchange (one daytime dwell) (Figure 1). As the optimal treatment and fill volume depend on body size and body weight, dialysate volumes were adapted on body surface area. However, most patients (n ¼ 12) had a body surface area of 1.7 2 m 2. Body surface area was 42m 2 in three patients and o1.7 m 2 in three patients, respectively (Table 1). A tidal mode was used in all patients. This was not to improve clearances, as several studies have shown that tidal peritoneal dialysis does not increase removal of small solutes or middle molecules as compared to conventional APD when dialysate flow is kept constant. 3,28,29 The reason for choosing tidal peritoneal dialysis was to minimize outflow alarms and, more importantly, mechanical outflow problems, especially during the days when clearances were measured. However, a rather high tidal volume of 75% was used uniformly in both study weeks. During manual exchange treatment, icodextrin (Extraneal, Baxter Healthcare Corp., Norfolk, UK) was used for one of the two daytime dwells (10 h), whereas the other daytime dwell (5 h) Kidney International (2006) 70, 1649 1655 1653

o r i g i n a l a r t i c l e D Demetriou et al.: Adequacy of high-flow APD was performed with conventional glucose solution (Dianeal, Baxter Healthcare Corp., Norfolk, UK). During high-flow treatment, there was only one daytime dwell (15 h, no manual exchange). Icodextrin solution was used for this dwell. During the first 5 days of both study weeks (weeks 1 and 3), the dialysate glucose concentration was adapted according to the need of each patient (dependent on residual renal function and peritoneal ultrafiltration). However, during the last 2 days of weeks 1 and 3 (blood and dialysate sampling), all patients used exclusively 2.27% glucose solutions (except for the one daytime dwell, which was performed with icodextrin). Patients were asked to adapt their fluid intake according to the changed peritoneal ultrafiltration on these 4 days. During the washout period (second week), each patient performed her/his regular APD treatment. All treatment sessions were performed using a Home Choice PRO-cycler (Baxter Healthcare Corporation, McGaw Park, IL, USA). Patient compliance was verified by analyzing the computer card of the cycler on which all treatments were recorded. Measurements Measurement of clearances, peritoneal ultrafiltration, and protein loss were performed during the last 2 days of each study week (days 6 and 7, days 20 and 21). On these days, patients were asked to collect the entire effluent of each cycler session (including the initial drain volume) and, during manual exchange treatment, additionally the effluent obtained during the manual daytime exchange. Dialysate samples were taken from these dialysate effluents. A single blood sample was also taken on each of these 4 days (in the middle of the daytime dwell, usually at 1400 1600 hours). Peritoneal clearances, protein loss, peritoneal ultrafiltration, and glucose absorption were calculated as reported previously. 3 As each patient was her/his own control, clearances were not normalized to body surface area. Renal clearances were not measured. Plasma and dialysate sodium were determined by an indirect electrode method after pre-dilution of the sample. This method has shown a good correlation with measurements performed by the flame photometry technique. 30 Sodium removal was calculated as difference between sodium concentration in the drained dialysate total drain volume and sodium concentration in the volume of instilled dialysate total inflow volume, as reported previously. 18,20 To characterize peritoneal transport types, we used the D/P of creatinine, calculated during the peritoneal equilibration test at 4 h before the start of the study. 31,32 Patients with a D/P of creatinine between 0.50 and 0.64 at 4 h were defined as LA transporters, whereas those with a D/P of creatinine between 0.65 and 0.81 at 4 h were categorized as HA transporters. Biochemical analyses were performed in an ISO 9001 certified laboratory. The correct entry of the data in the case report forms and in the electronic database was assured by double-checking by study monitors. Study hypothesis and outcomes We hypothesized that two different APD regimens ( manual exchange treatment and high-flow treatment) would perform equally with respect to clearance capacity. The primary study end point was the difference in 24 h peritoneal small solute clearance, both creatinine and urea, between manual exchange treatment and high-flow treatment. Secondary end points were changes in 24 h phosphate and b 2 -microglobulin clearances, sodium removal, ultrafiltrate volume, peritoneal glucose absorption, and protein loss between the two treatments. Sample size As intensified APD is especially important for patients with low residual renal function, it was determined that peritoneal creatinine clearance be 7.5 8.5 l/treatment, corresponding to a weekly clearance of 50 60 l. The defined a level was 5% (P ¼ 0.05) with a minimum power of 80% (b ¼ 0.80). A difference of 10% in the means was defined as clinically relevant based on previous data, which demonstrated an increased risk for mortality with a decline of 5 l/week/1.73 m 2 peritoneal creatinine clearance. 10 The s.d. of creatinine measurements was estimated between 10 and 12%. A sample size between 16 and 23 patients was determined as necessary to demonstrate equivalence between the two APD regimens using a crossover design based on the assumptions made above. Sample size calculations were performed using the software Sample Size V2.0. Randomization Patients were randomized for the sequence of the two treatments ( manual exchange treatment in week 1 followed by high-flow treatment in week 3 or vice versa) by computer-generated random digits (software package Sample Size V2.0) using sealed opaque envelopes. These envelopes were opened for each patient immediately before the study began. Patients were enrolled by one of the investigators. Randomization was performed by an independent investigator not involved in patient care. Statistical methods Continuous variables are presented as the mean and s.d. Continuous variables derived from the measurement of clearance parameters, ultrafiltration, sodium removal, glucose absorption, and protein loss were repeated two times on two sequential days (days 6, 7 and days 20, 21), and average values from these measurements were taken for final statistical analysis. Percentages were calculated for dichotomous variables. All calculations address paired within-subject comparisons of manual exchange vs high-flow regimens. Therefore, continuous variables were compared by paired t-tests. Data were checked for a normal distribution and all comparisons were recalculated with non-parametric statistical methods (Wilcoxon paired tests). In cases of divergent results between parametric and non-parametric methods, the more conservative P-value was presented. Calculations were performed using SPSS for Windows (Version 12.0, SPSS Inc., Chicago, IL, USA) and Stata release 8.0 (Stata Corp., College Station, TX, USA). ACKNOWLEDGMENTS We gratefully thank Professor Helmut Rumpold, Clinical Institute of Medical and Chemical Laboratory Diagnostics, Medical University Vienna, for measuring b 2 -microglobulin in plasma and dialysate. REFERENCES 1. National Kidney Foundation. NKF-K/DOQI clinical practice guidelines for peritoneal dialysis adequacy: update 2000. Am J Kidney Dis 2001; 37(Suppl 1): S65 S136. 2. Brown EA, Davies SJ, Rutherford P et al. Survival of functionally anuric patients on automated peritoneal dialysis: the European APD Outcome Study. J Am Soc Nephrol 2003; 14: 2948 2957. 3. Vychytil A, Lilaj T, Schneider B et al. Tidal peritoneal dialysis for home-treated patients: should it be preferred? Am J Kidney Dis 1999; 33: 334 343. 4. Perez RA, Blake PG, McMurray S et al. What is the optimal frequency of cycling in automated peritoneal dialysis? Perit Dial Int 2000; 20: 548 556. 1654 Kidney International (2006) 70, 1649 1655

D Demetriou et al.: Adequacy of high-flow APD o r i g i n a l a r t i c l e 5. Durand PY, Freida P, Issad B et al. How to reach optimal creatinine clearances in automated peritoneal dialysis. Perit Dial Int 1996; 16(Suppl 1): S167 S170. 6. Amici G, Virga G, Da Rin G et al. Continuous tidal peritoneal dialysis (CTPD) prescription and adequacy targets. Adv Perit Dial 1998; 14: 64 67. 7. Blake P, Burkart JM, Churchill DN et al. Recommended clinical practices for maximizing peritoneal dialysis clearances. Perit Dial Int 1996; 16: 448 456. 8. Sedlacek M, Dimaano F, Uribarri J. Relationship between phosphorus and creatinine clearance in peritoneal dialysis: clinical implications. Am J Kidney Dis 2000; 36: 1020 1024. 9. Juergensen P, Eras J, McClure B et al. The impact of various cycling regimens on phosphorus removal in chronic peritoneal dialysis patients. Int J Artif Organs 2005; 28: 1219 1223. 10. Szeto CC, Wong TY, Chow KM et al. Impact of dialysis adequacy on the mortality and morbidity of anuric Chinese patients receiving continuous ambulatory peritoneal dialysis. J Am Soc Nephrol 2001; 12: 355 360. 11. Jansen MAM, Termorshuizen F, Korevaar JC et al. Predictors of survival in anuric peritoneal dialysis patients. Kidney Int 2005; 68: 1199 1205. 12. Paniagua R, Amato D, Vonesh E et al. Effects of increased peritoneal clearances on mortality rates in peritoneal dialysis: ADEMEX, a prospective, randomized, controlled trial. J Am Soc Nephrol 2002; 13: 1307 1320. 13. Lo WK, Ho YW, Li CS et al. Effect of K t /V on survival and clinical outcome in CAPD patients in a randomized prospective study. Kidney Int 2003; 64: 649 656. 14. Keshaviah P. Adequacy of CAPD: a quantitative approach. Kidney Int 1992; 42(Suppl 38): S160 S164. 15. Villarroel F. Kinetics of intermittent and continuous peritoneal dialysis. J Dial 1977; 1: 333 347. 16. Ates K, Nergizoglu G, Keven K et al. Effect of fluid and sodium removal on mortality in peritoneal dialysis patients. Kidney Int 2001; 60: 767 776. 17. Struijk DG, Krediet RT. Sodium balance in automated peritoneal dialysis. Perit Dial Int 2000; 20(Suppl 2): S101 S105. 18. Ortega O, Gallar P, Carreno A et al. Peritoneal sodium mass removal in continuous ambulatory peritoneal dialysis and automated peritoneal dialysis: influence on blood pressure control. Am J Nephrol 2001; 21: 189 193. 19. Rodriguez-Carmona A, Pérez Fontán M. Sodium removal in patients undergoing CAPD and automated peritoneal dialysis. Perit Dial Int 2002; 22: 705 713. 20. Rodríguez-Carmona A, Pérez-Fontán M, García-Naveiro R et al. Compared time profiles of ultrafiltration, sodium removal, and renal function in incident CAPD and automated peritoneal dialysis patients. Am J Kidney Dis 2004; 44: 132 145. 21. Dombros NV. Pathogenesis and management of malnutrition in chronic peritoneal dialysis patients. Nephrol Dial Transplant 2001; 16(Suppl 6): 111 113. 22. Twardowski ZJ, Nolph KD, Khanna R et al. Daily clearances with continuous ambulatory peritoneal dialysis and nightly peritoneal dialysis. Trans Am Soc Artif Intern Organs 1986; 32: 575 580. 23. De Vecchi AF, Scalamogna A, Finazzi S et al. Preliminary evaluation of incremental peritoneal dialysis in 25 patients. Perit Dial Int 2000; 20: 412 417. 24. Rodríguez-Carmona A, Pérez Fontán M, García Falcón T et al. A comparative analysis on the incidence of peritonitis and exit-site infection in CAPD and automated peritoneal dialysis. Perit Dial Int 1999; 19: 253 258. 25. Oo TN, Roberts TL, Collins AJ. A comparison of peritonitis rates from the United States Renal Data System database: CAPD versus continuous cycling peritoneal dialysis patients. Am J Kidney Dis 2005; 45: 372 380. 26. Plum J, Hermann S, Fusshöller A et al. Peritoneal sclerosis in peritoneal dialysis patients related to dialysis settings and peritoneal transport properties. Kidney Int 2001; 59(Suppl 78): S42 S47. 27. Fusshoeller A, Baehr J, Tiemann B et al. Biocompatibility parameters in in vitro simulated automated versus continuous ambulatory peritoneal dialysis. Clin Nephrol 2005; 64: 214 220. 28. Piraino B, Bender F, Bernardini J. A comparison of clearances on tidal peritoneal dialysis and intermittent peritoneal dialysis. Perit Dial Int 1994; 14: 145 148. 29. Juergensen PH, Murphy AL, Pherson KA et al. Tidal peritoneal dialysis: comparison of different tidal regimens and automated peritoneal dialysis. Kidney Int 2000; 57: 2603 2607. 30. Aanen MC, Venturoli D, Davies SJ. A detailed analysis of sodium removal by peritoneal dialysis: comparison with predictions from the three-pore model of membrane function. Nephrol Dial Transplant 2005; 20: 1192 1200. 31. Twardowski ZJ, Nolph KD, Khanna R et al. Peritoneal equilibration test. Perit Dial Bull 1987; 7: 138 147. 32. Twardowski ZJ. Clinical value of standardized equilibration tests in CAPD patients. Blood Purif 1989; 7: 95 108. Kidney International (2006) 70, 1649 1655 1655