Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
mini review http://www.kidney-international.org & 2013 International Society of Nephrology Dialysis frequency versus dialysis time, that is the question Raymond M. Hakim1 and Sharmeela Saha1 1 Division of Nephrology and Hypertension, Department of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee, USA We reviewed a number of prospective randomized and multiple retrospective cohort studies of different dialysis prescriptions: longer dialysis time, at a frequency of at least three times a week, or a frequency of daily hemodialysis with a shorter dialysis time. Interestingly, the retrospective analyses have generally found significant survival benefits in the intensive dialysis groups, whereas more modest effects were observed in the prospective randomized controlled trials. The reason for this discrepancy may be related to the retrospective nature of the studies and possible selection bias; for example, the patients who were prescribed more frequent dialysis may have had more difficulties with volume control or high blood pressure. In contrast, the randomized controlled trials of increased dialysis frequency, which have shown indirect and modest benefits in complex coprimary end points, have small sample sizes and are plagued with difficulties in recruitment and compliance with the randomly allocated more frequent dialysis. This review, which attempts to balance the potential benefits of more frequent dialysis with the burden on the patient’s lifestyle, an increased risk of access malfunction, as well as societal costs of such intensive dialysis prescriptions, concludes in favor of the conventional three times per week dialysis (at a minimum) but at longer dialysis times than is currently prescribed based on the Kt/Vurea metric alone. Kidney International (2014) 85, 1024–1029; doi:10.1038/ki.2013.474; published online 11 December 2013 KEYWORDS: hemodialysis adequacy; mortality; ultrafiltration Correspondence: Raymond M. Hakim, Division of Nephrology and Hypertension, Department of Medicine, Vanderbilt University Medical Center, 1161 21st Avenue South, S-3223 MCN, Nashville, Tennessee 37232, USA. E-mail: [email protected] Received 10 April 2013; revised 5 August 2013; accepted 15 August 2013; published online 11 December 2013 1024 Several retrospective cohort studies as well as a few prospective randomized studies have assessed the impact of changes in dialysis treatment time (TT) or dialysis frequency. The major impetus for such studies stems from the fact that the mortality and morbidity of ESRD patients undergoing hemodialysis (HD), particularly in the United States, has been unacceptably high. According to the 2012 US Renal Data System (USRDS) report, only 30% of the patients who start dialysis in the United States are still alive at 5 years, and the adjusted mortality rate in prevalent dialysis patients 65 years or older is twice as high as that of patients with cancer, and six times as high as that of the general Medicare population.1 Comparison by DOPPS (Dialysis Outcome and Practice Patterns Study) of similar outcomes of the dialysis population in other countries, such as Japan or specific European countries where mortality rates are significantly lower, have indicated that such a high disease burden is not necessarily intrinsic to the disease process or its treatment by HD; such comparative analyses highlighted a number of modifiable parameters in the dialysis prescription, such as longer dialysis time that, if implemented in the United States, may improve patient outcomes.2 One of the earliest observational studies that highlighted the improved survival from longer dialysis times but at a frequency of three times a week came from the Tassin experience where HD had been provided three times a week for 8 h using in-center dialysis. Such longer TTs (accompanied by a strong emphasis on salt restriction) resulted in a standardized mortality rate of less than half that of concurrent patients from USRDS data.3 In contrast to the focus on dialysis session length but at a frequency of three times per week, dialysis frequency (daily or 6 days per week) has been the focus of several small studies in the United States. Kjellstrand et al.4 highlighted the 425% (hazard ratio (HR) ¼ 0.73) reduction in mortality of 150 patients who were prescribed daily in-center dialysis when compared with matched USRDS HD patients and an even greater reduction in mortality (HR ¼ 0.5) when 265 patients were dialyzed daily at home. Similarly, Blagg et al.5 and Johansen et al.6 found that patients receiving short daily or nocturnal HD at home had a significantly lower mortality rate (HR ¼ 0.39 and 0.64, respectively) when compared with patients receiving conventional HD in-center, matched by propensity score; such retrospective studies provided the impetus to explore Kidney International (2014) 85, 1024–1029 RM Hakim and S Saha: Dialysis frequency vs. dialysis time longer dialysis time and more frequent dialysis as prescription tools to improve the survival of HD patients in the United States. RANDOMIZED CONTROLLED TRIALS OF DIALYSIS PRESCRIPTION—CONSTANT FREQUENCY The National Cooperative Dialysis Study (NCDS)7 and the Hemodialysis Study Group (HEMO)8 were some of the earlier prospective randomized controlled trials to address dialysis prescriptions. Both studies attempted to examine the impact of different dialysis doses on hospitalization (NDCS) and on mortality (HEMO); both studies maintained the frequency of dialysis at three times per week and randomized patients to different doses of dialysis, using urea as the surrogate molecule. NCDS, which was powered to detect the impact of dialysis dose on hospitalization, focused on ‘time-average concentration of urea’ (TACurea) and concluded that lower TACurea was associated with lower hospitalization. Although NCDS included dialysis time as one of the two main intent-to-treat interventions (in the two longer dialysis time groups, mean session length was 4.3 h, compared with 3.2 h for the shorter dialysis session), the magnitude of the time effect was large, but fell just short of achieving the ‘classical’ statistical significance (P ¼ 0.06).7 As stated by Chertow, ‘in retrospect, one might argue that the NCDS session length was the most significant (i.e., important) ‘nonsignificant’ (statistically) effect in the history of dialysis research.’9 In a reanalysis of the NCDS, the TACurea concept was later modified to analyze the data in terms of Kt/V (again with K and V determined only by urea) and concluded that the optimal dose of dialysis was achieved when Kt/Vurea is X1.0 (single pool).10 This reanalysis of the NCDS study did not consider dialysis time as an independent factor. The singular focus on Kt/V gained wide acceptance in the late 1980s and resulted in a trend in which patients were dialyzed at higher blood flows, using larger surface area dialyzers to reach the minimum Kt/V in the shortest possible time. This coincided with a consistently high mortality rate in the United States, which in the early 1990s reached an annual mortality rate close to 30%1 and led to the planning of the HEMO study in the late 1990s. The HEMO study attempted to compare two doses of dialysis, defined by Kt/Vurea, in 1846 patients, still keeping the dialysis frequency at three times a week. The HEMO study also attempted to define the impact of different dialysis membranes by comparing the use of high-flux and low-flux dialysis membranes, defined by their clearance of b2 microglobulin.8 Although in the initial ‘intent-to-treat’ analysis of the HEMO results there was no statistical difference in the mortality of patients randomized to any of the assigned therapies, in retrospect, the relatively small difference in Kt/V (20% higher in the ‘high’ Kt/V group, compared with the standard Kt/V) and a relatively small difference in dialysis time (approximately 30 min per session longer in the high Kt/V group) may explain these negative Kidney International (2014) 85, 1024–1029 mini review results. One important consideration in the HEMO study is that the exclusion criterion of ‘eKt/V of 1.3 not achieved in 4.5 h’ practically excluded all patients over 100 kg and thus 97% of randomized patients weighed o100 kg; indeed, the average weight of patients in the HEMO study was 69.2 kg and the average age was 57.6 years, both considerably less than those of the prevalent patients on dialysis at the time.8 Nevertheless, these findings led many nephrologists to conclude that there is no benefit of increasing the dialysis dose or dialysis time for patients receiving HD three times a week. Post-hoc analysis of the HEMO study, using an as-treated model, concluded that patients randomized to the high-flux group had statistically significant reductions in both the risk of death from cardiac causes and in the combined outcome of first hospitalization for cardiac causes and/or death, and suggested that dialysis time had a marked effect on survival particularly in the high-dose arm of the study.11 Although these post-hoc results were interpreted as strongly biased (dose-targeting bias),12 they also pointed out the possibility that total weekly dialysis time, independent of Kt/Vurea, may be a critical factor in patient outcomes. FHN TRIAL: RANDOMIZED CONTROLLED TRIAL OF DIFFERENT DIALYSIS FREQUENCIES Several investigators concluded that, to determine whether still higher doses of dialysis or longer dialysis time may result in improved patient outcomes, patients will need to receive HD more than three times per week and/or for much greater duration than the minimum time to achieve a Kt/Vurea of X1.0. These preliminary conclusions were the basis of the prospective randomized Frequent Hemodialysis Network trial (FHN). There were two components of the FHN trial: in the first, patients were randomly assigned to undergo HD either six times per week or three times per week for 52 weeks. The ‘intent-to-treat’ FHN trial highlighted that in the six times per week dialysis there were significant improvements in both coprimary composite outcomes (death or change in left ventricular mass and, separately, death and/or change in physical health composite score), but, importantly, because of the relatively small number of patients in each arm (approximately 120), there was no difference in death or hospitalization (unrelated to vascular access) (HR ¼ 0.93, P ¼ 0.71) between the two groups, and the major difference was in the relatively modest reduction in left ventricular mass among the survivors of the daily HD group.13 There were also no significant effects of frequent HD on cognitive performance, self-reported depression, serum albumin concentration, or the use of erythropoietin-stimulating agents.13,14 The second component of the FHN trial was to compare daily nocturnal (6–8 h) home HD to conventional three times weekly home HD. This component of the FHN trial also suffered from a slow and difficult recruiting process, which resulted in only 87 patients to be randomized and was thus significantly underpowered.15 Therefore, although the 1025 mini review delivered dose was much higher in the nocturnal FHN arm, there was no statistical difference in the same two predefined coprimary outcomes described above. Finally, although both parts of the FHN study were ‘prospective, randomized’, the participants in the study were younger and weighed less than average than patients in the United States (e.g., 50.4 years old in the FHN study, compared with 62.8-year-old incident patients in the United States).1,13 NON-RANDOMIZED RETROSPECTIVE COHORT TRIALS These results of the FHN trial are in contrast to those of several retrospective clinical studies reported by several other investigators. In a recent analysis by the International Quotidian Dialysis Registry (IQDR) in 2012, Nesrallah et al.16 compared the outcomes of 4300 patients who were receiving intensive home dialysis (4.8 sessions per week, with a mean TT of 7.4 h per session) with those of 1388 patients who were in the conventional dialysis group (three sessions per week, with a mean TT of 3.9 h per session). Comparing patient outcomes, the HR of mortality in the intensive dialysis group was close to half (HR ¼ 0.55) compared with the conventional dialysis group. Such improved patient outcomes have led Kjellstrand et al.4 and Pauly17 to conclude that intensive HD results in a probability of survival that is similar between patients on intensive dialysis and those who have undergone deceased donor transplantation. A similar retrospective study by Lacson et al.18 evaluated nearly 1000 patients undergoing in-center nocturnal dialysis three times weekly (7.85 h per treatment) compared with patients dialyzed with conventional dialysis (3.75 h per treatment). The authors concluded that these longer dialysis times were associated with a 25% reduction in the risk of death (HR 0.75, Po0.004) when compared with patients in the conventional arm, after matching and adjusting for several variables using a propensity score. More recently, using propensity matching and USRDS data, Weinhandl et al.,19 also found 13% lower mortality in patients undergoing daily home dialysis compared with three times a week in-center HD patients. Thus, these three retrospective cohort studies and the accompanying editorials concluded that there is a strong association between longer HD TT (at a minimum frequency of three times per week) and improved survival.20,21 WHY THE DIFFERENCE IN RESULTS BETWEEN THE FHN AND RETROSPECTIVE STUDIES? Although the prospective randomized FHN trial was positive in favor of more intensive dialysis, these results were certainly less marked than the results of the retrospective observational studies. It is perhaps worthwhile to speculate on the reasons for this disparity. The most important, in our consideration, was that the FHN trial required patients to be ‘blindly’ randomized to either arm of the study. Several observations point to the reluctance of patients to participate in the FHN trial, including the 4-year recruitment period where over 6000 1026 RM Hakim and S Saha: Dialysis frequency vs. dialysis time patients were screened but only 245 patients were randomized.13,22 Thus, o4% of the screened patients actually agreed to participate in the study.22 Even in those patients who were finally randomized, the acceptance of daily dialysis was relatively poor.12 This is confirmed by the large disparity in treatment attendance, which was statistically significant between the two groups (77.7% for frequent vs. 94.9% for conventional, Po0.001).13 Thus, one of the major differences between the FHN studies and the retrospective studies may well be the acceptance (or resistance) of patients to undergo more intensive dialysis. A comparative review of the benefits and barriers of intensive dialysis, which may also influence a patient’s decision to participate in trials of intensive HD, was recently provided by Chan and colleagues.23 Another important consideration in the interpretation of the FHN trial is that in both randomized groups (short daily and conventional) the rate of fluid ultrafiltration (UFR) was very similar at approximately 10–11 ml/kg/h, a critical factor that will be discussed in more detail later.13 Finally, although there was a statistical difference in the composite end point of ‘mortality and change in left ventricular mass’, patients assigned to frequent HD are also more likely to undergo vascular access interventions (HR ¼ 1.71).24 It is possible therefore that the contrast between the modest difference in the coprimary outcome seen in the FHN trial (but no statistical difference in the mortality end point) and the larger difference in mortality outcomes observed in retrospective cohort studies lies in the manner in which patients were asked to participate in such studies. One presumes that the patients in the retrospective studies were educated on the advantages of more HD, either because the patients had an underlying indication for longer or more frequent dialysis (e.g., large interdialytic weight gains) leading to results in favor of more HD, but which may be affected by a bias by indication, or because of the strong conviction and recommendation of their nephrologist. Thus, the patient’s willingness to undergo the intervention along with the associated lifestyle changes is an important but difficult-to-quantitate factor in the outcome of patients facing complex therapies. Another possible explanation for the difference in the conclusions regarding the effectiveness of frequent dialysis between the FHN trial and the retrospective cohort studies is the length of follow-up. In the FHN trial, participation was limited to 1 year, whereas the Kaplan–Meier analysis of the retrospective cohort by the IQDR indicated that the percentage survival between the conventional and intensive HD begins to diverge after 1 year.16 Potentially, the duration of participation in the FHN trial may have been too short to show a difference in mortality alone. Finally, although propensity matching or risk adjustments are partially successful in minimizing such bias, these statistical adjustments may not fully account for the differences in severity of the comorbidities or other unmeasured variations, and may also result in biased conclusions from such retrospective studies.12 Kidney International (2014) 85, 1024–1029 mini review RM Hakim and S Saha: Dialysis frequency vs. dialysis time SO, WHAT IS THE ANSWER: DIALYSIS FREQUENCY OR DIALYSIS TIME? The review of the above studies does not provide a clear answer to this question. However, it would be reasonable to conclude that longer dialysis times, preferably nocturnal for 6–7 h, provided at a minimum of three times a week (or more often if it is provided at home) appears to be the dialytic therapy with the most favorable balance between benefits, risks, and lifestyle burden, and may be the therapy that is more readily accepted by patients in the long term. In the post-hoc reanalysis of patients randomized to the high-dose arm in both the NCDS and the HEMO study (both at a frequency of three times per week), an increase in dialysis time during follow-up (in an ‘as-treated’ analysis) was associated with a marked improvement in survival, whereas shorter dialysis time was associated with increased mortality risk.27 Both the nocturnal (three times a week in-center) dialysis study by Lacson and the frequent (three to seven times per week) home IQDR dialysis study that basically analyzed ‘as-treated’ patients highlighted significant survival advantages for the patients who accepted longer dialysis time, particularly in home dialysis patients, where the advantages of a reliable, self-interested, and motivated access cannulator are important but difficult-to-quantitate factors, in contrast to the FHN trial that indirectly highlighted the patient ‘burnout’ that may occur from daily travel for in-center dialysis. There is also indirect evidence that increasing the frequency of longer dialysis (e.g., 5–7 episodes of nocturnal home dialysis) may further improve patient outcomes, compared with three times a week nocturnal dialysis. In the study by Nesrallah et al.,16 nocturnal treatment with frequencies higher than three times per week had a lower hazard risk of mortality (HR ¼ 0.55), compared with the study by Lacson et al.,18 which found an HR of 0.75 in patients undergoing in-center (three times per week) nocturnal dialysis. Nevertheless, the reduced hazard risk of mortality in the ‘longer and more frequent’ approach should be balanced against the increased burden on the patient’s quality of life, particularly if the intensive therapy is provided in-center. Longer dialysis sessions, at a frequency of at least Kidney International (2014) 85, 1024–1029 Associations between UFR and CV and all-cause mortality 2 1.8 Adjusted hazard ratio Unfortunately, to further confuse the issue of dialysis time or dialysis frequency, the same investigators who highlighted the survival advantages of intensive nocturnal home dialysis discussed above (IQDR)16 have recently published a comparison of a multinational cohort study of in-center short daily HD (weekly TT 15.7 h) compared with contemporaneous patients receiving conventional (three times weekly) dialysis. Using propensity score-based matching, these investigators concluded that patients receiving short daily HD had a significantly higher mortality (HR ¼ 1.6) compared with those receiving conventional HD.25 The mortality difference was seen in matched and unmatched adjusted analyses and was qualitatively similar across the subgroups analyzed.25 A critique of this retrospective study and its conclusions followed shortly.26 1.6 1.4 1.2 CV mortality All-cause mortality 1 5 10 15 20 UFR (ml/kg/h) Figure 1 | Cubic spline analysis of the associations between ultrafiltration rate (UFR) and cardiovascular (CV; blue line) and all-cause (purple line) mortality. HRs were adjusted for age, sex, interdialytic weight gain, race (black, non-black), smoking status (never, past, current), vintage (o1, 1–2, 2–4, X4 years), access type (graft, fistula, catheter), systolic blood pressure (o120, 120–140, 140–160, 160–180, X180 mm Hg), residual urine output (pvs. 4200 ml/day), diabetes, congestive heart failure, peripheral vascular disease, ischemic heart disease, cerebrovascular disease, serum albumin, creatinine, hematocrit (o30, 30–33, 33–36, X36%), and phosphorus, and use of a-adrenergic blocker, angiotensin-converting enzyme inhibitor/angiotensin receptor blocker, b-blocker, calciumchannel blocker, nitrates, and other antihypertensives. Estimates are presented for UFRs between 5.8 ml/kg/h (the 5th percentile of observed UFR in the study sample) and 20.4 ml/kg/h (the 95th percentile). Adapted from Flythe et al.31 three times a week, may also have physiologic advantages; short, daily dialysis primarily promotes the removal of urea and other small molecules, whereas the longer dialysis session is associated with a higher fractional removal of larger molecules compared with smaller molecules.28,29 Finally, longer dialysis sessions (accompanied by an emphasis on reduced dietary salt intake and an individualized or reduced dialysate sodium30) would allow a UFR that should not exceed 10 ml/kg/h, a critical UFR above which there is an associated increase in mortality (Figure 1).31 One potential argument against longer dialysis sessions at a frequency of only three times per week is the study by Foley et al.,32 which highlights a higher mortality rate after a long interdialytic time. Although such conclusions were derived from the USRDS data, which include primarily the traditional, brief (3.5 h) dialysis session of three times per week, it may be worthwhile to have a similar analysis in patients dialyzed for longer periods of time but at the same frequency of three times a week. If confirmed, a potential compromise is to consider alternate-day nocturnal dialysis, which may reduce the excess hazard of the longer interdialytic interval.32 WHAT SHOULD THE MINIMUM DIALYSIS SESSION LENGTH BE? Tentori et al.2 found that longer TT was strongly associated with lower mortality in Japan, Europe, Australia, and New Zealand but not in North America; however, when they 1027 mini review RM Hakim and S Saha: Dialysis frequency vs. dialysis time Association between RxDSL and mortality P-difference=0.005 P-trend=0.038 1.28 (0.98–1.67) 1.30 (1.02–1.68) < 210 min n =691 Matched pairs 210–240 min n =691 Matched pairs 1.00 (ref.) . 240 min n =691 Matched pairs 1.00 (ref.) < 240 min n =2382 Matched pairs 1 1.26 (1.07–1.48) . 240 min n =2382 Matched pairs HR (95% CI) 1.5 1.25 Figure 2 | Adjusted associations between prescribed dialysis session length (RxDSL) and mortality on the basis of the Cox regression models. Subjects were matched on gender, access type (fistula, graft, catheter), age (±2.5 years), and post-dialysis weight (±1 kg). The multivariable models contained covariate terms for age, race (black, non-black), post-dialysis weight, interdialytic weight gain (p0, 1–1.49, 1.5–2.99, 3.0–3.99, and X4.0 kg), vintage (o1, 2, 3, X4 years), diabetes, coronary artery disease, congestive heart failure, missed dialysis sessions over 30 days (0, 1, 2, 3 X4), creatinine (quartiles), albumin (p3, 3–3.5, 3.5–4, 44 g/dl), phosphorus (p4, 4–5, 5–6, X6 mg/dl), and pre-dialysis systolic blood pressure (p130, 130–150, 150–170, 4170 mm Hg). Models were stratified on matchedpair assignment. CI, confidence interval; HR, hazard ratio; ref., reference (adapted from Flythe et al.34). Increased frequency (short daily dialysis) Increased time (nocturnal dialysis) Pros Improved blood pressure control Less phosphorus binders Improved coprimary outcomes which include LVH and mortality Pros Improved large solute clearance such as β-2 microglobulin Improved blood pressure control Less phosphorus binders Lower ultrafiltration rate Lower mortality Cons Increased access complications Inconvenient for patients Higher ultrafiltration rate Higher medicare costs ?? Increased mortality Cons Hypophosphatemia Faster loss of residual renal function Figure 3 | Factors influencing modifications of current dialysis prescription with changes in frequency or time. LVH, left ventricular hypertrophy. adjusted for non-adherence, they found a similar effect of dialysis time on mortality in North America. Although longer dialysis time outside the United States is associated with financial incentives (Japan) or penalties (England) or may be the result of regulatory requirements (Germany), and the conclusion by DOPPS may reflect such ‘dose-targeting bias’, the ‘as-treated’ analysis by DOPPS provides strong support for the longer dialysis time. Longer TT was associated with 1028 lower mortality in Japan (HR ¼ 0.75 per 30 min longer TT) and in European countries, Australia, and New Zealand (HR 0.94 per 30 min longer TT).2 The study by Saran et al.33 using the DOPPS database and more recently by Flythe et al.34 using the database of a large dialysis provider have both shown that the UFR has a significant impact on HD mortality rate: UFR 410 ml/kg/h is associated with increased all-cause and cardiovascular mortality, possibly on the basis of cardiac stunning, subclinical cardiac ischemia, and hemodynamic instability during rapid ultrafiltration. Subsequently, Flythe et al.34 reported on the association between TT and survival, and found a substantial mortality increase associated with a dialysis session of o4 h, even when adjusted for body size (Figure 2). The adverse impact of short dialysis (o4 h) session on blood pressure control was also highlighted in a study by Tandon et al.35 The study by Flythe was critically analyzed by Daugirdas,12 who cautioned against interpreting these results as ‘cause and effect’ and that conclusions from such studies may reflect a ‘dose-targeting bias.’ CONCLUSIONS The weight of current evidence, particularly when considering patient acceptance, compliance, and maintenance of vascular access, is in favor of increased dialysis time to at least 4 h per session, and preferably as nocturnal dialysis with each session length of 6–8 h, with a minimum frequency of three times per week (Figure 3). Longer dialysis sessions allow for a lower UFR (preferably p10 ml/kg/h) and a higher removal of middle-size molecules. Longer dialysis but at a minimum frequency of three times weekly compared with short daily dialysis also reduces the risk of malfunction of the patient’s access, an important outcome that has been discussed recently.36 Finally, from a societal point of view, given the extraordinarily high per-patient cost of the current ESRD program, the economic viability of more frequent in-center short dialysis strategies is questionable.37 It would be ideal if our conclusions could be tested by prospective randomized studies;38–40 the likelihood of successfully financing and implementing such studies is unknown; in the meantime, we owe it to our patients to provide them with longer time therapy (44 h or nocturnal) at a frequency of at least three times a week that is based on a preponderance of evidence and which is consistent with physiologic principles of dialysis.41,42 DISCLOSURE The authors declared no competing interests. REFERENCES 1. 2. US Renal Data System. USRDS Annual Data Report: Atlas of End-Stage Renal Disease in the United States. National Institute of Diabetes and Digestive and Kidney Diseases. National Institutes of Health: Bethesda, MD, 2012. Tentori F, Zhang J, Li Y et al. Longer dialysis session length is associated with better intermediate outcomes and survival among patients on in-center three times per week hemodialysis: results from the Dialysis Outcomes and Practice Patterns Study (DOPPS). Nephrol Dial Transplant 2012; 27: 4180–4188. Kidney International (2014) 85, 1024–1029 mini review RM Hakim and S Saha: Dialysis frequency vs. dialysis time 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. Laurent G, Charra B. The results of an 8 h thrice weekly haemodialysis schedule. Nephrol Dial Transplant 1998; 13(Suppl 6): 125–131. Kjellstrand CM, Buoncristiani U, Ting G et al. Short daily haemodialysis: survival in 415 patients treated for 1006 patient-years. Nephrol Dial Transplant 2008; 23: 3283–3289. Blagg CR, Kjellstrand CM, Ting GO et al. Comparison of survival between short-daily hemodialysis and conventional hemodialysis using the standardized mortality ratio. Hemodial Int 2006; 10: 371–374. Johansen KL, Zhang R, Huang Y et al. Survival and hospitalization among patients using nocturnal and short daily compared to conventional hemodialysis: a USRDS study. Kidney Int 2009; 76: 984–990. Lowrie EG, Laird NM, Parker TF et al. Effect of the hemodialysis prescription of patient morbidity: report from the National Cooperative Dialysis Study. N Engl J Med 1981; 305: 1176–1181. Eknoyan G, Beck GJ, Cheung AK et al. Effect of dialysis dose and membrane flux in maintenance hemodialysis. N Engl J Med 2002; 347: 2010–2019. Chertow GM, Kurella M, Lowrie EG. The tortoise and hare on hemodialysis: does slow and steady win the race? Kidney Int 2006; 70: 24–25. Gotch FA, Sargent JA. A mechanistic analysis of the National Cooperative Dialysis Study (NCDS). Kidney Int 1985; 28: 526–534. Greene T, Daugirdas J, Depner T et al. Association of achieved dialysis dose with mortality in the hemodialysis study: an example of ‘dosetargeting bias’. J Am Soc Nephrol 2005; 16: 3371–3380. Daugirdas JT. Dialysis time, survival, and dose-targeting bias. Kidney Int 2013; 83: 9–13. Chertow GM, Levin NW, Beck GJ et al. In-center hemodialysis six times per week versus three times per week. N Engl J Med 2010; 363: 2287–2300. Kurella Tamura M, Unruh ML, Nissenson AR et al. Effect of more frequent hemodialysis on cognitive function in the frequent hemodialysis network trials. Am J Kidney Dis 2013; 61: 228–237. Rocco MV, Lockridge RS Jr., Beck GJ et al. The effects of frequent nocturnal home hemodialysis: the Frequent Hemodialysis Network Nocturnal Trial. Kidney Int 2011; 80: 1080–1091. Nesrallah GE, Lindsay RM, Cuerden MS et al. Intensive hemodialysis associates with improved survival compared with conventional hemodialysis. J Am Soc Nephrol 2012; 23: 696–705. Pauly RP. Survival comparison between intensive hemodialysis and transplantation in the context of the existing literature surrounding nocturnal and short-daily hemodialysis. Nephrol Dial Transplant 2013; 28: 44–47. Lacson E, Xu J, Suri RS et al. Survival with three-times weekly in-center nocturnal versus conventional hemodialysis. J Am Soc Nephrol 2012; 23: 687–695. Weinhandl ED, Liu J, Gilbertson DT et al. Survival in daily home hemodialysis and matched thrice-weekly in-center hemodialysis patients. J Am Soc Nephro 2012; 23: 895–904. Ikizler TA. Intensive hemodialysis: back to the beginning? J Am Soc Nephrol 2012; 23: 573–575. Kuhlmann MK. The eternal (nocturnal) quest for better dialysis outcomes. J Am Soc Nephrol 2012; 23: 571–573. Pauly RP, Klarenbach SW, Komenda P. Comparative survival literature in intensive hemodialysis: limitations and future directions. Semin Dial 2011; 24: 629–633. Kidney International (2014) 85, 1024–1029 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. Schachter ME, Chan CT. Current state of intensive hemodialysis: a comparative review of benefits and barriers. Nephrol Dial Transplant 2012; 27: 4307–4313. Suri RS, Larive B, Sherer S et al. Risk of vascular access complications with frequent hemodialysis. J Am Soc Nephrol 2013; 24: 498–505. Suri RS, Lindsay RM, Bieber BA et al. A multinational cohort study of incenter daily hemodialysis and patient survival. Kidney Int 2013; 83: 300–307. Kliger AS. Dialysis: does daily haemodialysis reduce survival? Nat Rev Nephrol 2013; 9: 8–9. Mehrotra R, Himmelfarb J. Dialysis in 2012: could longer and more frequent haemodialysis improve outcomes? Nat Rev Nephrol 2013; 9: 74–75. Daugirdas JT, Chertow GM, Larive B et al. Effects of frequent hemodialysis on measures of CKD mineral and bone disorder. J Am Soc Nephrol 2012; 23: 727–738. Vanholder R, Eloot S, Neirynck N et al. Home haemodialysis and uraemic toxin removal: does a happy marriage exist? Nat Rev Nephrol 2012; 8: 579–588. Mc Causland FR, Waikar SS, Brunelli SM. The relevance of dietary sodium in hemodialysis. Nephrol Dial Transplant 2013; 28: 797–802. Flythe JE, Kimmel SE, Brunelli SM. Rapid fluid removal during dialysis is associated with cardiovascular morbidity and mortality. Kidney Int 2011; 79: 250–257. Foley RN, Gilbertson DT, Murray T et al. Long interdialytic interval and mortality among patients receiving hemodialysis. N Engl J Med 2011; 365: 1099–1107. Saran R, Canaud BJ, Depner TA et al. Dose of dialysis: key lessons from major observational studies and clinical trials. Am J Kidney Dis 2004; 44: 47–53. Flythe JE, Curhan GC, Brunelli SM. Shorter length dialysis sessions are associated with increased mortality, independent of body weight. Kidney Int 2013; 83: 104–113. Tandon T, Sinha AD, Agarwal R. Shorter delivered dialysis times associate with a higher and more difficult to treat blood pressure. Nephrol Dial Transplant 2013; 28: 1562–1568. Nesrallah G. Canadian Society of Nephrology Guidelines for the Management of Patients with ESRD Treated With Intensive Hemodialysis. Am J Kidney Dis 2013; 62: 187–198. Lee CP, Zenios SA, Chertow GM. Cost-effectiveness of frequent in-center hemodialysis. J Am Soc Nephrol 2008; 19: 1792–1797. Singh AK, Kari JA, Twardowski ZJ et al. Pro: the value of randomized controlled studies in dialysis methods. Nephrol Dial Transplant 2013; 28: 834–835. Twardowski ZJ, Misra M, Singh AK. Con: randomized controlled trials (RCT) have failed in the study of dialysis methods. Nephrol Dial Transplant 2013; 28: 826–832. Van Biesen W, Lameire N. Moderator’s view: are alternative dialysis strategies superior to conventional dialysis: what Popper tells us. Nephrol Dial Transplant 2013; 28: 837–838. Chazot C, Ok E, Lacson E et al. Thrice-weekly nocturnal hemodialysis: the overlooked alternative to improve patient outcomes. Nephrol Dial Transplant 2013; 28: 2447–2455. Demirci C, Ozkahya M, Demirci MS et al. Effects of three times weekly eight-hour nocturnal hemodialysis on volume and nutritional status. Am J Nephrol 2013; 37: 559–567. 1029