Short Daily Versus Long Nocturnal Hemodialysis

Abstract

Interest in quotidian (daily) hemodialysis (HD) seems to be growing. A number of groups have advocated short, high efficiency daily HD, e.g., Buoncristiani and the group at Perugia, Italy, 1,2 and a few have advocated long slow HD while the patient is asleep, so-called nocturnal HD. The leader in this area is the Toronto group that was headed by the late Dr. Uldall and now by Dr. A. Pierratos. 3, 4 Several presentations have been made at major nephrologic meetings and there are a number of publications that purport improvement in the outcomes of HD patients who are treated with these methods. 5 Unfortunately, there are as yet, no randomized prospective studies comparing outcomes in patients who are dialyzing with conventional three times per week HD and either the short daily or long slow nocturnal HD methods. This is the first daily/nocturnal HD study to obtain comparative data in this field. This study commenced in November of 1998 and will be complete by the end of 2001. This article gives, for the first time, preliminary results from the study. Methods Daily/Nocturnal HD Study This is a three-year study, designed to study 20 patients, with 10 patients receiving short daily HD and 10 patients receiving nocturnal HD. All patients are monitored for 18 months and treatments are provided in the home environment. This study is funded by the Ministry of Health, the government of Ontario, Canada. Patient Recruitment The Southwestern Ontario renal program currently provides dialysis treatments for nearly 500 patients with end-stage renal failure, who may receive their dialysis treatments in a variety of manners. These include: (1) hospital in-center HD, where HD is provided to the most dependent patients in a tertiary teaching hospital; (2) satellite HD that provides dialysis to dependent, but stable patients in their local community hospital; (3) minimally assisted, self care HD that provides HD to patients who are able to be responsible for the majority of their treatments. There is a minimally assisted, self care HD facility in the city of London, but self care patients are also treated usually on different shifts in peripheral hospitals where satellite dialysis units have been created; (4) in geographic areas, remote from the major centre or the satellites, home HD is performed, usually with the patient being totally independent, although occasionally with nursing assistants provided; and (5) home peritoneal dialysis. Information regarding the daily/nocturnal dialysis study was provided to all HD patients early in 1998. Patients who were interested were asked to apply for the study, but to indicate in the application whether they would wish to be treated with short high efficiency daily HD; long, slow nocturnal dialysis; or if they had no preference. Thereafter, patients were selected by ballot. Each patient was then given additional information with the details of the study and, after informed consent was obtained, they were entered into the study. The study was approved by the Ethics Committee of The University of Western Ontario, London, Canada. Study Design This study is observational; it is not a randomized, prospective controlled study. This was done deliberately because patient preferences, regarding which arm of experimental treatment they would receive were anticipated. Randomization might, therefore, be fraught with difficulties and possibly influence patients’ quality of life. All patients were studied in detail while still receiving conventional HD, and then serially while receiving the experimental daily/nocturnal program. For each patient selected for the study, a cohort control subject was selected who matched as closely as possible for age, gender, co-morbidity, and the original dialysis modality from which the patient came. All patients selected were more than 18 years of age, able to give informed consent, had been on dialysis for a minimum of 3 months, and had a reasonable expectation of surviving 1 year, regardless of whatever co-morbidities were present. The details of this study’s treatments are shown in Table 1. The outcomes that were measured are listed in Table 2. Table 1: Characteristics of Study TreatmentsTable 2: Study OutcomesStudy Hypotheses The following were our study hypotheses: (1) both modalities would show better outcomes than those obtained with conventional HD treatments, this would also include economic assessment; (2) the study would provide information on the relative pros and cons for daily versus nocturnal home HD; and (3) the study would help in the development of appropriate equipment for all forms of quotidian HD. Patients There were 28 patients who applied for the study that met the entry criteria. Eighteen of these patients initially said that they would only participate in the short daily arm of the study, leaving 10 who had indicated their preference for slow nocturnal HD. Two of the first 18 patients subsequently elected to be considered for nocturnal dialysis. Ten patients were selected by ballot for each group and, to date, 9 of the 10 patients are receiving short daily HD in their homes, whereas all 10 of the patients opting for nocturnal HD are receiving home treatment. Details of the recruitment are shown in Figure 1. Details of the patient characteristics are shown in Table 3. Figure 1: Recruitment details of patients for study.Table 3: Patient Characteristics at Study EntryData Analysis Because this is an interim report of an ongoing study, the results will be presented for each study patient from a baseline (t0) while receiving conventional therapy. These data were uniformly collected during an intensive study week during their last week on conventional, three times per week HD, before commencing quotidian dialysis. T0 was, therefore, a variable time between November 1998 and May 2000. This is compared with the results to date, i.e., the latest results for any parameters as of June 2000 (t1). Similar data were collected for the cohort control subjects who had the same t0 and t1 dates. The elapsed study is a mean of 8.9 ± 5.4 months with a range of 1 to 18 months, and the total patient-months experience was 169 at this point. The follow-up in patient-months for the nine daily and 10 nocturnal patients is shown in Table 4; the format for data comparison is shown in Figure 2. From this figure, it can be seen that not only are the study patients acting as their own control subjects with regard to conventional HD, but that they can be studied against their control subjects both at t0 and t1. Furthermore, the control subjects can be compared with themselves over the time of follow-up. Data sets of results are therefore presented in this format under seven headings: (1) urea kinetic data, (2) anemia, (3) calcium and phosphate metabolism, (4) blood pressure and volume management, (5) hospitalization, (6) blood access, and (7) quality of life. All data are also summarized in Table 5. Table 4: Patient Follow-Up (months)Figure 2: Schematic of format for data comparisons.Table 5: ResultsStatistical Analysis Analysis of variance was used for between-group comparison at t0 and t1, whereas paired t-tests were employed for within group comparison from t0 to t1. Results Urea Kinetic Data Figure 3 shows that at baseline there was no significant difference between the weekly Kt/Vs delivered between study patients and their control subjects. Over time, the control subjects continued to receive the same therapy, whereas there were significant increases (p < 0.001) in the therapy that the study patients received. In particular, nocturnal therapy provided a considerable increase in delivered Kt/V, one that was significantly higher at t1 from the other groups (p < 0.05). The urea reduction (%) for all groups at t0 was approximately 75%. At t1, the control subject population maintained their 75% urea reduction. The mean reduction for the nocturnal group did decrease significantly (p = 0.025) to a mean of 70%, but this was not significantly different from the control subject population. The short daily study group, however, had a significant (p < 0.001) decrease in the mean percentage of urea reduction to 60%. This was statistically significant from the other three groups at t1 (p < 0.05). The predialysis ureas for all populations are shown in Figure 4, which shows that only the nocturnal group had a significant (p = 0.001) decrease in predialysis urea that was, in turn, significantly less than the other groups at t1 (p < 0.05). Figure 3: Weekly Kt/V delivered to study patients and control subjects.Figure 4: Predialysis Urea (mmol/L) levels for study patients and control subjects.Urea kinetic studies also examined normalized protein equivalent of nitrogen appearance (nPNA) (gm/kg per day) (formerly protein catabolic rate [PCR] gm/kg) as an indicator of dietary protein intake. These data are depicted in Figure 5, which shows that at t0, all groups had mean nPNAs approximately 1 to 1.1 gm/kg per day. This was not changed at t1 in the nocturnal dialysis group, nor in either control subject group, but did significantly increase (p = 0.009) in the short daily dialysis group. These data correlated well with information obtained from 3 day diaries that the patients were asked to keep. Ongoing anthropometric studies are being carried out, but to date, the results have not been analyzed. The serum albumins at t0 and t1 for all groups are shown in Table 5. The only significant change over the study period was the rise in mean serum albumin from 38.6 g/L to 40.8 g/L in the short daily study group (p = 0.001). Figure 5: nPNA (gm/kg per day) as an indicator of dietary protein intake for study patients and control subjects.Anemia Figure 6 shows the mean hemoglobin levels at t0 and t1 for all four groups. Canadian Practice Guidelines 9 (dialysis outcomes quality initiative [DOQI] equivalent) state that dialysis patients should target their hemoglobin to be in the range of 110 to 120 g/L. Figure 6 indicates this was generally adhered to in all groups, with the exception of the nocturnal study group which was just under the advised range at t0. This is mainly because one patient in the nocturnal group has aplastic anemia as a co-morbid event. Figure 6 also shows that in the short daily group alone, there was a significant rise (p = 0.019) in the hemoglobin, over the study period. In all groups, iron studies were satisfactory and all patients had serum ferritins in excess of 100 μg/L and transferrin saturations in excess of 20%. The majority of patients were receiving intravenous iron that at t1, was approximately 20 mg per week, for all patients regardless of group. The majority of patients were receiving erythropoietin. Throughout the study, the nocturnal group received the most erythropoietin (12,000 ± 10,300 U/week), which was significantly different (p < 0.05) from the short daily patients who received only 3,333 ± 2,958 U/week at t1. At t0, the short daily study patients were receiving 6,889 ± 4,807 (±SD) U/week. The decrease in erythropoietin requirements in that particular group is statistically significant (p < 0.05). As with the hemoglobin results, it is likely that the higher erythropoietin dose for the nocturnal study group was influenced by the one patient with aplastic anemia, who alone was receiving 20,000 U/week. Figure 6: Mean hemoglobin (gm/L) levels at t0 and t1 for study patients and control subjects.Calcium and Phosphate Metabolism As far as predialysis calcium levels are concerned, there were no differences between any group at t0. However, over the course of the study, there was a slight decrease in the predialysis serum calcium level in the nocturnal study group; this was of borderline significance (p = 0.09). The predialysis serum phosphate levels are shown in Figure 7. Generally speaking, phosphate control was good in all groups during the study period. Although not statistically significant, there was a downward trend of the serum phosphates over time in the nocturnal study group. This is important because this particular study group was virtually free from the use of phosphate binders. The amount (gm/day) of calcium carbonate, used for phosphate binding, is shown for all groups in Figure 8. All groups were taking the same amount of calcium carbonate at t0. The two control subject groups virtually maintained the same ingestion level at t1, but both daily and nocturnal groups (especially) showed significant decrease in their use of binders (p < 0.004). At t1, the patients receiving nocturnal HD were taking significantly fewer binders than the other three groups (p < 0.05). There were no significant changes or differences between any group in immunoreactive parathyroid hormone levels, or in alkaline phosphatase. Figure 7: Predialysis serum phosphate (mmol/L) levels for study patients and control subjects.Figure 8: The calcium carbonate dose (gm/day) for study patients and control subjects.Blood Pressure and Volume Management No differences occurred in any group over the study period, in pre or postdialysis diastolic, systolic, or mean arterial pressures (mm Hg). On the other hand, there was a statistically significant decrease in the intra-dialysis weight gain in both study groups over the period (p < 0.05 for nocturnal;p < 0.001 for daily). There were no such significant differences in the control subject populations. This was associated with a reduction in blood pressure medication taken by the two study groups. At t0, the patients in the nocturnal group were taking a mean of 1.9 blood pressure tablets per day. This was reduced to 0.6 tablets per day by t1 (p < 0.05). The short daily study group was taking a mean of 2.8 blood pressure tablets per day and this was reduced to a mean of 0.7 tablets per day by t1 (p = NS). Hospitalization During the study period, there were only four hospitalizations of patients in the two study groups. These totalled 17 days, which is equal to 1.8 days per patient-year for the whole population. At the time of this report, the data from the control subject population were not available. Blood Access Table 6 summarizes the blood access experience by t1. Again there were no comparative data from the control subject population, thus no real inferences can be made. However, in this population there have been only two total access failures over the study period, which amounts to two failures over 113 patient-months. Table 6: Blood Access by t1Quality of Life All of the patients throughout this study completed a number of quality of life questionnaires. Study patients showed significant improvement in a number of areas. These included intradialytic symptoms (p < 0.05), fatigue (p < 0.05), time for full recovery from dialysis treatment (p < 0.03), and uremic symptoms (p < 0.005). In all instances, the study patients had similar scores to their control subject groups at t0, and the control subject group scored virtually identically at t1 as at t0. There was also improvement in psychosocial stress scoring (p < 0.05). There was no evidence of any improvement in the physical component summary section of the (short form [SF] 36) and the study patients scored similarly to their control subjects over the whole study period. In the SF36 mental components summary, however, there was a statistically significant improvement in the short daily dialysis study group (p = 0.015). There was no such change in the nocturnal study group. Discussion This is an interim report on an ongoing study. For this reason, no comment will be made comparing these study data with other published observations. Nevertheless, the data to date do show that the study patients have a higher delivered Kt/V (urea), especially in the nocturnal group. There did seem to be a higher nPNA, particularly in the daily dialysis group, suggesting better protein intake. The tendency was for hemoglobin to improve according to the amount of erythropoietin taken; this association is dramatic in the daily group. Phosphate was better controlled with less binders, particularly in the nocturnal dialysis group. Blood pressure and volume control were easier with less medication. To date, we have not come across any access problems, although it is fair to indicate that the control data are not yet available. Likewise, hospitalization was rare, but again there are as yet, no comparative data for the control subject population. The study groups seem to have better quality of life, both for treatment and uremic states. As far as economics are concerned, there is a detailed economic model in place to assess the true cost of both forms of experimental therapy as compared with the conventional paradigm, but these data have not yet been analyzed in any way. In summary, the data to date, clearly suggest statistically significant benefits for patients receiving these therapies as compared with the conventional dialytic paradigm. However, at this juncture, it would be very difficult to say whether there is any clear advantage of daily versus nocturnal therapy or vice versa. The definitive answer will likely appear during the continuing course of this study. Acknowledgment The authors wish to acknowledge the Ministry of Health, Government of Ontario for their financial support.