Download Dialysis frequency versus dialysis time, that is the question

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