Extended Daily Dialysis Versus Continuous Renal Replacement
Therapy for Acute Kidney Injury: A Meta-analysis
Ling Zhang, MD,
Jiqiao Yang, MD,
Glenn M. Eastwood, MD,
Guijun Zhu, MD,
Aiko Tanaka, MD,
and Rinaldo Bellomo, MD, PhD
Background: Extended daily dialysis (EDD) has been suggested as an effective renal replacement therapy
for acute kidney injury. However, results from studies comparing EDD to continuous renal replacement therapy
(CRRT) are inconclusive.
Study Design: A systematic review and meta-analysis was performed by searching in MEDLINE, EMBASE,
the Cochrane Library, Google Scholar, and a Chinese database (SinsoMed).
Setting & Population: Patients with acute kidney injury.
Selection Criteria for Studies: Randomized controlled trials (RCTs) and observational studies were
included. EDD was deﬁned as extended hemodialysis or hemodiaﬁltration for more than 6 but less than 24
hours per session using a conventional hemodialysis machine.
Intervention: Renal replacement therapy comparing EDD with CRRT.
Outcomes: Mortality, kidney recovery, and ﬂuid removal.
Results: We included 17 studies from 2000 to 2014: 7 RCTs and 10 observational studies involving 533 and
675 patients, respectively. Meta-analysis of RCTs showed no difference in mortality rates between EDD and
CRRT (relative risk, 0.90; 95% CI, 0.74-1.11; P50.3). However, EDD was associated with lower mortality risk
compared with CRRT in observational studies (relative risk, 0.86; 95% CI, 0.74-1.00; P50.05). There was no
evidence of heterogeneity in RCTs (I
50%) or observational studies (I
515%). In both RCTs and
observational studies, there were no signiﬁcant differences in recovery of kidney function, ﬂuid removal, or
days in the intensive care unit, and EDD showed similar biochemical efﬁcacy to CRRT during treatment
(serum urea, serum creatinine, and serum phosphate).
Limitations: The survival beneﬁt of EDD is dependent on only observational studies and might have been
affected by allocation or selection bias.
Conclusions: EDD is associated with similar outcomes to CRRT in RCTs. The ﬁnding that EDD is asso-
ciated with a lower mortality rate relies on data from observational studies, which are potentially subject to
allocation or selection bias, making further high-quality RCTs desirable.
Am J Kidney Dis. 66(2):322-330. ª2015 by the National Kidney Foundation, Inc.
INDEX WORDS: Extended daily dialysis (EDD); continuous renal replacement therapy (CRRT); hemoﬁltration;
hemodiaﬁltration; prolonged intermittent dialysis; sustained low-efﬁciency dialysis/diaﬁltration (SLED),
hemodialysis; acute kidney injury (AKI); acute renal failure (ARF); mortality; kidney recovery; ﬂuid removal;
Acute kidney injury (AKI) is a major problem in
critical illness. When AKI is severe, renal
replacement therapy (RRT) might be required.
Continuous RRT (CRRT) is the preferred method in
intensive care unit (ICU) patients, but survival rates
Extended daily dialysis (EDD), ﬁrst per-
formed for a patient with AKI in 1945 as an alternative
modality of RRT to CRRT, is a special form of inter-
mittent dialysis with low dialysate and blood ﬂow rates
and prolonged duration.
In recent studies, EDD
appeared better tolerated hemodynamically and showed
faster normalization of deranged metabolic parameters
compared to conventional intermittent hemodialysis.
Several small randomized controlled trials (RCTs)
have compared EDD to CRRT for the treatment of
AKI. They found no signiﬁcant difference in mor-
tality between the 2 groups.
However, a recent
study reported that EDD might reduce mortality in
patients with AKI in comparison to CRRT.
sought to systematically review the current literature
and analyze all studies comparing EDD to CRRT for
the treatment of AKI in critical illness.
Division of Nephrology, West China Hospital of
Sichuan University, Sichuan, Chengdu, China;
Intensive Care Unit, Austin Health, Heidelberg, VIC, Australia;
West China School of Medicine, Sichuan University, Sichuan,
Division of Intensive Care Unit, Fourth Hospital
of Hebei Medical University, Shijiazhuang, Hebei, China.
Received November 23, 2014. Accepted in revised form
February 13, 2015. Originally published online April 2, 2015.
Address correspondence to Rinaldo Bellomo, MD, PhD, Division
of Intensive Care Unit, Austin Health, 145 Studley Road, Heidel-
berg, VIC 3084, Australia. E-mail: email@example.com
2015 by the National Kidney Foundation, Inc.
322 Am J Kidney Dis. 2015;66(2):322-330
We performed this systematic review using the guidelines pro-
posed by the Cochrane Collaboration in the Cochrane Handbook for
Systematic Reviews of Interventions (www.cochrane-handbook.
org). There was no registered protocol for this meta-analysis.
Study Selection Criteria
This review focused on patients with critical illness who
received EDD or CRRT for AKI.
For the purpose of the review, EDD was deﬁned as extended
(sessions .6 but ,24 hours) hemodialysis or hemodiaﬁltration
using a conventional hemodialysis machine, including extended
daily dialysis/diaﬁltration, sustained low-efﬁciency dialysis/dia-
ﬁltration, or prolonged intermittent dialysis/diaﬁltration. We spe-
ciﬁcally excluded intermittent hemoﬁltration in our analysis.
We used the term CRRT to describe continuous hemoﬁltration
and/or continuous hemodialysis and/or continuous hemodiaﬁltra-
tion, all intended to run on a continuous basis (24 hours a day).
Types of Outcome Measures
The primary outcomes were mortality and kidney recovery
among patients with AKI. Fluid removal, ICU days, laboratory
results, and other parameters during RRT were also analyzed.
Types of Studies
We included all RCTs and observational studies concerning
EDD versus CRRT for patients with AKI from 2000 to 2014. We
excluded reviews, commentaries, and editorials.
Search Methods for Identiﬁcation of Studies
We used the PRISMA (Preferred Reporting Items for System-
atic Reviews and Meta-analyses) and MOOSE (Meta-analysis of
Observational Studies in Epidemiology) statement methodologies
to report a systematic review and meta-analysis of RCTs and
Two independent reviewers (L.Z. and
J.Y.) conducted a search in MEDLINE, EMBASE, the Cochrane
Library, Google Scholar, a Chinese database (SinoMed), and
major nephrology journals. Trials were considered without lan-
guage or date restriction. We performed the last updated search on
August 8, 2014. The following text words and corresponding
heading terms were used as search terms: “acute kidney injury or
acute renal failure”and “dialysis or hemodialysis or renal
replacement therapy or blood puriﬁcation.”The exact search
strategy is provided in Item S1 (available as online supplementary
material). Related articles and reference lists were manually
searched to avoid omissions. After title screening, we evaluated
abstracts for relevance and identiﬁed studies as included, excluded,
or requiring further assessment. At this stage, if a paper required
further assessment, we contacted the study lead investigator by
e-mail and/or telephone with a request for further information.
Inclusion criteria were as follows: (1) studies comparing EDD
with CRRT and (2) sufﬁcient data available to calculate a relative
risk (RR) or mean difference with 95% conﬁdence interval (CI).
The following exclusion criteria were used: (1) studies comparing
standard with intensiﬁed extended dialysis; (2) studies comparing
EDD with other modalities of dialysis, such as standard intermit-
tent dialysis or peritoneal dialysis; and (3) nonhuman studies. For
studies with the same or overlapping data by the same authors, the
most suitable studies with the largest number of cases or latest
publication dates were selected.
Two investigators (L.Z. and J.Y.) assessed each trial indepen-
dently and recorded eligibility, quality, and outcomes. Disagree-
ments regarding eligibility arose with 7% of the articles (
which were resolved by consensus, with a third investigator
(G.M.E.) providing arbitration. We extracted the following study
features: ﬁrst author, publication year, country, funding source,
number of participants, RRT modalities, mortality, kidney recov-
ery, ﬂuid removal, ICU days, and parameters and laboratory results
during RRT. Outcomes reported in 2 or more articles were
extracted for meta-analysis.
Quantitative Data Synthesis
Independently and in duplicate, reviewers assessed risk of bias of
RCTs using the Cochrane collaboration tool.
For each RCT, a
description, a comment, and a judgment of “high,”“unclear,”or
“low”risk of bias was provided for each of the following domains:
adequate random sequence generation, allocation sequence con-
cealment, blinding for objective outcomes, incomplete outcome
data, freedom from selective outcome reporting, and freedom from
other bias. Trials with high risk of bias for 1 or more key domains
were considered at high risk of bias. Trials with low risk of bias for
all key domains were considered at low risk of bias. Otherwise, they
were considered at unclear risk of bias.
The Newcastle-Ottawa Scale was used in the assessment of
quality of observational studies (case-control or cohort studies).
A judgment of “high,”“unclear,”or “low”risk of bias was pro-
vided for each domains, and a “low”risk of bias was scored 1 and
a“high”or “unclear”risk of bias was scored “0.”A quality bar
was plotted for each domain to examine the limitations of the
studies. Studies of high quality were deﬁned as a score .5 points.
Before the analysis, data were standardized into equivalent
units. For dichotomous variables such as mortality, rates in the
experimental (EDD) and control (CRRT) groups were expressed
as RR and 95% CI. For continuous variables such as ﬂuid removal,
mean difference and 95% CI were calculated for each study.
Heterogeneity was evaluated using the Cochrane Q test and I
statistic to assess the degree of interstudy variation. I
0% to 24.9%, 25% to 49.9%, 50% to 74.9%, and 75% to 100%
were considered as having no, mild, moderate, and signiﬁcant
thresholds for statistical heterogeneity.
model using restricted maximum likelihood,
which is thought
to be better than the conventional DerSimonian-Laird method,
was performed to provide more conservative estimates of effect
in the presence of known or unknown heterogeneity. Sensitivity
analysis was conducted by sequentially omitting each study one at
a time in an attempt to identify the potential inﬂuence of an in-
dividual study. Data analysis was performed using Stata SE,
version 12, software (StataCorp LP).
The study selection process is presented in Fig 1.
The literature search yielded 607 potentially relevant
records. By screening the titles, we removed 249
duplicate studies. After evaluating the abstract of
each, 323 studies were excluded because they did not
meet the inclusion criteria. Subsequently, we carefully
read the full text of each of the remaining 35 trials and
excluded 18 trials. Reasons for exclusion were com-
parison with other modalities of RRT (n 58), no
comparison between RRTs (n 57), not all patients
with AKI (n 52), and overlapping data (n 51).
Extended Dialysis for Acute Kidney Injury
Am J Kidney Dis. 2015;66(2):322-330 323
Finally, 7 RCTs
and 10 observational
comparing EDD with CRRT for patients
with AKI were included in this systematic review and
The eligible studies were conducted from 2000 to
2014 with a total of 634 patients undergoing EDD and
574 patients undergoing CRRT. Six studies were from
Asia; 4, North America; 4, Europe; 2, Oceania; and 1,
Africa. A variety of outcomes were recorded in these
studies, including mortality (16 studies),
recovery (7 studies),
ﬂuid removal (9
ICU days (5 studies),
serum creatinine level (8 studies),
urea level (7 studies),
serum phosphate level
total heparin dose (3 studies),
events of increasing vasopressors dose (3 studies).
Among these studies, the duration of EDD was 6.0 to
14.9 hours per day, which was lower than CRRT (18.0-
23.5 hours per day). Characteristics of the RCTs and
observational studies fulﬁlling the inclusion criteria are
listed in Table 1 and Item S2.
Assessment of Methodological Quality
The details of risk of bias for RCTs are summarized in
Fig 2. Only 2 studies
were judged to be at low risk of
bias, one study
was judged to be at high risk of bias
due to selection bias, and the other 4 were judged to
be at unclear risk of bias. Three studies
adequate randomized sequence and reported appropriate
allocation concealment. Among all RCTs, none was
double blinded. However, blinding of patients and cli-
nicians was impossible in studies comparing EDD with
CRRT, and the authors judged that the primary outcome
(mortality) is not likely to be inﬂuenced by lack of
The Newcastle-Ottawa Scale quality assessment of
included observational studies is also listed in Fig 2. All
studies showed comparatively high quality (score .5).
The main limitation observed in 5 studies
unclear reporting of dropout rates or uncompleted
In 6 RCTs (617 patients) and 10 observational
studies (675 patients), data for mortality were re-
ported. Within RCTs (Fig 3), there was no statisti-
cally signiﬁcant difference in the risk of mortality
rate between EDD and CRRT (RR, 0.90; 95% CI,
0.74-1.11; P50.3). There was no evidence for
50). When observational studies
comparing EDD with CRRT were pooled, EDD was
associated with lower risk for mortality compared
with CRRT (RR, 0.86; 95% CI, 0.74-1.0;
P50.05). No evidence was detected for heteroge-
A total of 3 RCTs (174 patients) and 5 observa-
tional studies (238 patients) provided information on
kidney recovery. Within RCTs, as shown in Fig 3,
there was no signiﬁcant difference in kidney recovery
rate with EDD in comparison to CRRT (RR, 1.12;
95% CI, 0.83-1.76; P50.4), with moderate hetero-
565%). Similar results were obtained in
observational studies (RR, 1.14; 95% CI, 0.90-1.46;
Overall, 4 RCTs (367 patients) and 3 observational
studies (140 patients) reported data for ﬂuid removal.
Within RCTs, the overall summary mean difference
using the random-effects model was 20.10 (95%
CI, 20.39 to 0.19) L/d (I
573%), indicating a
similar effect on ﬂuid removal between EDD and
CRRT (Fig 3). On analysis of observational studies,
there were also no signiﬁcant differences in ﬂuid
removal between EDD and CRRT (mean
difference, 20.06 [95% CI, 21.03 to 0.91] L/d), with
signiﬁcant heterogeneity (I
Two RCTs (120 patients; 864 sessions) reported
data for ﬂuid removal rate during treatment (Table 2);
607 Potentially relevant studies identified by
MEDLINE (n = 206)
EMBASE (n = 267)
Cochrane (n = 89)
Others (n = 45)
358 Potentially relevant articles screened based on
249 excluded (duplicate studies)
35 full-text articles selected for full review
17 studies included in the meta-analysis
231 Reviews, comments, and editorials
45 Other modalities of RRT
21 Patients with other diagnosis
15 Studied other treatment
11 No relevant events data
8 Comparison with other modalities of
7 No comparison between RRT
2 Not all patients with AKI
Figure 1. Flow chart of selection of studies. Abbreviations:
AKI, acute kidney injury; RCT, randomized controlled trial;
RRT, renal replacement therapy.
324 Am J Kidney Dis. 2015;66(2):322-330
Zhang et al
Table 1. Characteristics of Studies Fulﬁlling the Inclusion Criteria
Study Country Design of Study Modality N Mean Age (y) Male Sex (%)
Main Outcomes FundingEDD CRRT
(2004) DE RCT EDD vs CVVH 39 50.5 62.9 11.7 23.3 Mortality, ﬂuid removal Industry
(2007) AU RCT EDD vs CVVH 16 69.5 56.3 7.3 18.4 Fluid removal NR
(2010) JP RCT EDDF vs CVVHDF 60 68.7 65.0 6.5 20.3 Mortality, kidney recovery, ICU days NR
(2011) JP RCT EDDF vs CVVHDF 50 65.9 66.0 6.0 15.2 Mortality, kidney recovery, ICU days NR
(2011) KR RCT SLED vs CVVH 46 63 63.0 10 NR Mortality NR
(2012) DE RCT SLED vs CVVH 232 66.2 67.7 14.9 19.9 Mortality, ﬂuid removal, ICU days NR
(2012) EG RCT EDD vs CVVHDF 80 47.5 65.0 6-8 NR Mortality, ﬂuid removal, ICU days NR
(2000) US Retrospective EDD vs CVVH 42 50 64.3 7.5 19.5 Mortality NR
(2004) US Prospective EDD vs CVVHD 54 52 63.0 6.7 16.8 Mortality, kidney recovery NR
(2006) CA Prospective SLED vs CVVHDF 34 58.4 61.8 7.5 21.3 Mortality, ﬂuid removal NR
(2006) PT Retrospective SLED vs CVVHDF 53 59.1 NR 6.8 22.1 Mortality NR
(2008) CN Prospective SLED vs CVVH 12 49.7 66.7 10 18 Mortality, kidney recovery Public
(2009) PT Retrospective SLED vs CVVHDF 63 63.3 49.2 6-12 NR Mortality, ﬂuid removal NR
(2010) CA Retrospective SLED vs CVVHDF 43 62.1 76.7 6.8 19.7 Mortality, ﬂuid removal NR
(2010) TW Retrospective SLED vs CVVH 101 67.4 65.3 8.0 NR Mortality, kidney recovery NR
(2012) NZ Retrospective SLEDF vs CVVHDF 166 58.5 62.0 7.2 NR Mortality, ICU days NR
(2014) CN Retrospective SLEDF vs CVVH 107 59.5 NR 8.8 23.5 Mortality, kidney recovery NR
Abbreviations: AU, Australia; CA, Canada; CN, China; CRRT, continuous renal replacement therapy; CVVH, continuous venovenous hemoﬁltration; CVVHD, continuous venovenous he-
modialysis; CVVHDF, continuous venovenous hemodiaﬁltration; DE, Germany; EDD, extended daily dialysis; EDDF, extended daily diaﬁltration; EG, Egypt; ICU, intensive care unit; JP, Japan;
KR, Korea; NR, not reported; NZ, New Zealand; PT, Portugal; RCT, randomized controlled trial; SLED, sustained low-efﬁciency dialysis; SLEDF, sustained low-efﬁciency diaﬁltration; TW,
Taiwan; US, United States.
Am J Kidney Dis. 2015;66(2):322-330 325
Extended Dialysis for Acute Kidney Injury
EDD was associated with a higher ﬂuid removal rate
compared to CRRT (202.23 [95% CI, 151.27-253.19]
Length of ICU Stay
A total of 4 RCTs
(422 patients) and one
(166 patients) described length
of ICU stay. Within RCTs, as shown in Table 2, there
was similar duration of ICU stay between EDD and
CRRT (mean difference, 21.51 [95% CI, 27.84 to
4.3] days), with signiﬁcant heterogeneity (I
Similar results were obtained in observational studies
(mean difference, 22.30 [95% CI, 25.61 to 1.01]
Laboratory Results During RRT
Overall, 3 RCTs
(351 patients) and 5 observa-
(724 patients) reported on
serum creatinine. Within RCTs, as shown in Table 2,
the overall summary mean difference using the
random-effects model was 0.37 (95% CI, 20.43 to
1.17) mg/dL, indicating there was no difference in
serum creatinine levels between CRRT and EDD, with
signiﬁcant heterogeneity (I
598%). Similar results
were obtained in observational studies (mean differ-
ence, 0.40 [95% CI, 20.36 to 1.15] mg/dL; I
A total of 3 RCTs
(351 patients) and 4
(690 patients) reported
plasma urea levels (Table 2). There was no signiﬁcant
difference in urea levels between EDD and CRRT in
RCTs (mean difference, 28.56 [95% CI, 237.17 to
20.04] mg/dL; I
599%) or observational studies
(mean difference, 14.90 [95% CI, 28.20 to 38.0] mg/
(271 patients) and one observational
(54 patients) reported on serum phosphate
levels. As shown in Table 2, there was no signiﬁcant
difference in serum phosphate levels with EDD in
comparison to CRRT in RCTs (mean difference,
20.31 [95% CI, 21.50 to 0.89] mg/dL; I
observational studies (mean difference, 0.31 [95%
CI, 20.38 to 1.0] mg/dL).
(351 patients) and one observational
(724 patients) reported on heparin dose
(Table 2). Within RCTs, there was nominally lower use
of heparin per session with EDD than with CRRT, but
the difference was not statistically signiﬁcant (mean
difference, 24.49 [95% CI, 212.09 to 3.10] kU/d;
599%). However, the difference was signiﬁcant
in observational studies (mean difference, 28.91 [95%
CI, 29.32 to 28.50] kU/d).
Episodes of Vasopressor Escalation
(89 patients; 412 sessions) and 1
(43 patients; 125 sessions)
described episodes of vasopressor escalation during
RRT. There was no signiﬁcant difference between
EDD and CRRT in RCTs (P50.8; I
observational studies (P50.2). The outcome might
be different if it were expressed as hourly rates
Figure 2. Risk of bias summary.
Abbreviation: RCT, randomized con-
326 Am J Kidney Dis. 2015;66(2):322-330
Zhang et al
because the duration of CRRT in these studies was
about twice as long as EDD; however, no relevant
data could be obtained for any of the included studies.
Three studies reported costs in different ways,
including total RRT costs in 2 studies,
per day in 2 studies,
and total costs of hospitali-
zation in 1 study.
All results indicated that costs
were lower with EDD compared to CRRT. Meta-
analysis could not be estimated due to absence of a
In order to assess the stability of results of the
current meta-analysis, we performed sensitivity anal-
ysis for mortality by omitting a single study. Statis-
tically similar results were obtained after omitting
Mortality ( Observat ional Stud ies)
Mortality (R CTs)
Kidney Recovery (RCTs)
Fluid Removal (RCTs) Fluid Removal (Observational Studies)
Effect (rem l)
Effect (rem l)
Kidney Recovery (Observational Studies)
Effect (rem l)
Figure 3. Forest plot for mortality, kidney recovery, and ﬂuid removal. The analysis was stratiﬁed by study design. Original weights
(squares) displayed; largest to smallest ratios are 13.64, 11.30, 1.51, 3.83, 10.35, and 1.11, respectively, for randomized controlled trial
(RCT) mortality, observational mortality, RTC kidney recovery, observational kidney recovery, RCT ﬂuid removal, RCT ﬂuid removal.
Abbreviations: CI, conﬁdence interval; MD, mean difference; RR, risk ratio.
Am J Kidney Dis. 2015;66(2):322-330 327
Extended Dialysis for Acute Kidney Injury
each of the studies (RR range in RCTs, 0.82-0.93; RR
range in observational studies, 0.81-0.90). We also
evaluated the effect on overall mortality of removing
studies with unclear methodological quality
and found similar RRs (0.89 and 0.86,
respectively; P.0.05) in RCTs.
We performed a systematic review of the literature
and identiﬁed 17 original studies reporting data on
EDD versus CRRT among more than 1,200 patients
with AKI. When we analyzed RCTs, we found that
patients who received EDD as an initial RRT mo-
dality for AKI had a similar risk of death compared
with those who initially received CRRT. However, a
difference in favor of EDD was seen in observational
studies. In both RCTs and observational studies, there
were no signiﬁcant differences in kidney recovery,
ﬂuid removal, and episodes of vasopressor escalation
with EDD in comparison to CRRT. Also, EDD
showed similar efﬁcacy to CRRT in laboratory results
(serum urea, serum creatinine, and serum phosphate)
during RRT and was associated with lower costs.
have compared con-
ventional intermittent RRT and CRRT and found no
signiﬁcant difference in mortality, and a recent RCT
conﬁrmed this result.
A meta-analysis found that conventional intermittent
RRT may be associated with higher rates of dialysis
dependence compared with CRRT.
because EDD is a hybrid technology combining prop-
erties from both intermittent RRT and CRRT, EDD
should not be considered a conventional form of
intermittent RRT. To date, to our knowledge, only 1
in 2008 compared EDD (in the form of
sustained low-efﬁciency dialysis [SLED]) with CRRT
for AKI, but it included just 1 trial with 54 patients. In
contrast, the present review includes data from 7 RCTs
and 10 observational studies with more than 1,200
patients with AKI. Such studies, although subject to
bias, involve a large number of patients and might be
more likely to accurately represent the efﬁcacy of EDD
for patients with AKI. In this meta-analysis, there was a
mild trend toward improved survival in favor of EDD-
treated patients with AKI, although the evidence was
weak because of a lack of signiﬁcant difference in
RCTs. There might be some reasons for this trend. First,
heparin was the main anticoagulant in included studies
(table bof Item S2), and CRRT used a relatively high
dose of heparin, which might be associated with higher
risk of bleeding. Second, a higher rate of clotting was
found in CRRT from included studies,
might contribute to inadequate treatment and increased
Third, long duration of treatment might be
associated with a higher rate of biofilm formation and
Many studies and meta-analyses have focused on
conventional intermittent RRT in comparison to CRRT
for critically ill patients with AKI and found no sig-
niﬁcant difference in mortality. However, there might
be some advantages and disadvantages with both
In this regard, EDD to some extent
combines several advantages of both modalities.
However, there are only a few high-quality studies to
compare EDD with other modalities for patients with
AKI. In addition, much more attention should be
focused on the intensity of EDD, such as dose or fre-
quency, although some evidence shows that intensiﬁed
EDD is not associated with better outcomes.
Thus, future studies comparing RRT modalities
should focus on studying EDD as a therapeutic option
in severe AKI.
Table 2. Meta-analysis of Secondary Outcomes
Outcome Study Type No. of Studies EDD vs CRRT
Fluid removal rate (in mL/h) RCT 2 202.23 (151.27 to 253.19) 98%
ICU days RCT 3 21.51 (27.84 to 4.83) 80%
Observational 1 22.30 (25.61 to 1.01) —
Serum creatinine (in mg/dL) RCT 3 0.37 (20.43 to 1.17) 98%
Observational 5 0.40 (20.36 to 1.15) 91%
Serum urea (in mg/dL) RCT 3 28.56 (237.17 to 20.04) 99%
Observational 4 14.90 (28.20 to 38.00) 95%
Serum phosphate (in mg/dL) RCT 2 20.31 (21.50 to 0.89) 98%
Observational 1 0.31 (20.38 to 1.0) —
Heparin dose (I kU/d) RCT 2 24.49 (212.09 to 3.10) 99%
Observational 1 28.91 (29.32 to 28.50) —
Episodes of vasopressor escalation RCT 2 0.85 (0.52 to 1.40) 0%
Observational 1 0.65 (0.36 to 1.18) —
Abbreviations: CRRT, continuous renal replacement therapy; EDD, extended daily dialysis; ICU, intensive care unit; RCT, ran-
domized controlled trial.
Values shown are mean difference (EDD –CRRT), except episodes of vasopressor escalation, which give relative risk for EDD
versus CRRT. Values in parentheses are 95% conﬁdence intervals.
328 Am J Kidney Dis. 2015;66(2):322-330
Zhang et al
To our knowledge, this study is the ﬁrst to sys-
tematically evaluate the effect of EDD versus CRRT
on patients with AKI. Our search strategy was broad
and included studies in English, Chinese,
languages. It included data from
more than 1,200 patients, 17 studies, and 10 coun-
tries; from different regions of Asia, North America,
Europe, Oceania, and Africa; and from both large
observational studies and RCTs. Furthermore, 2 in-
dependent investigators thoroughly evaluated meth-
However, our study also has several limitations.
First, this association with mortality is largely
dependent on observational studies and might have
been affected by allocation or selection bias.
Second, despite a total of 17 studies, eligible studies
comparing speciﬁc outcomes were limited. No sufﬁ-
cient data were obtained to evaluate adverse outcomes
such as bleeding, infection, and hypotension. In
addition, although a similar effect on ﬂuid removal
was found for EDD and CRRT, data regarding ﬂuid
intake and volume balance were lacking. Furthermore,
the evidence on secondary outcomes might be at
high risk of reporting bias because only a few
included studies (#50%) reported these outcomes
and did so with high heterogeneity. Thus, in our
opinion, these results do not warrant changing clinical
practice, but rather support the need for additional
Third, although we extracted data for mortality at
the end of follow-up, the duration of each study
varied from 10 days in 1 study
to 30 days in 6
to 90 days in 2 studies,
or to ICU
mortality in 1 study
and in-hospital mortality in 6
Even so, although end points of
different follow-up periods could modify the absolute
risk, they should not bias the overall RR.
Last, but not least, only published studies with selective
databases were included for data analysis. The unavail-
ability of unreported outcomes possibly could result in
reporting bias. For instance, 2 studies reported insufﬁcient
data for mortality.
Regardless of these limitations, we
have minimized bias throughout the process by our
methods of study identiﬁcation, data selection, and sta-
tistical analysis and in our controlling publication bias and
sensitivity. These steps should strengthen the stability and
accuracy of the meta-analysis.
In conclusion, available RCTs do not show a dif-
ference in mortality between EDD and CRRT.
However, observational studies suggest that EDD
may be associated with a greater survival rate.
Because these studies might be associated with allo-
cation or selection bias, further high-quality RCTs
focused on mortality according to different RRT
modalities are necessary to fully understand the ef-
fects of EDD for patients with AKI.
Financial Disclosure: The authors declare that they have no
relevant ﬁnancial interests.
Contributions: Research idea and study design: LZ, RB; data
acquisition: LZ, JY; data analysis/interpretation: LZ, JY, AT;
statistical analysis: LZ, GZ; supervision or mentorship: RB, GME.
Each author contributed important intellectual content during
manuscript drafting or revision and accepts accountability for the
overall work by ensuring that questions pertaining to the accuracy
or integrity of any portion of the work are appropriately investi-
gated and resolved. LZ and RB take responsibility that this study
has been reported honestly, accurately, and transparently; that no
important aspects of the study have been omitted, and that any
discrepancies from the study as planned have been explained.
Item S1: Electronic search strategies.
Item S2: Additional characteristics of studies fulﬁlling inclusion
Note: The supplementary material accompanying this article
(http://dx.doi.org/10.1053/j.ajkd.2015.02.328) is available at
1. Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure
in critically ill patients: a multinational, multicenter study. JAMA.
2. Vinsonneau C, Camus C, Combes A, et al. Continuous
venovenous haemodiaﬁltration versus intermittent haemodialysis
for acute renal failure in patients with multiple-organ dysfunc-
tion syndrome: a multicentre randomised trial. Lancet. 2006;368:
3. Kielstein JT, Schiffer M, Hafer C. Back to the future:
extended dialysis for treatment of acute kidney injury in the
intensive care unit. J Nephrol. 2010;23:494-501.
4. Kumar N, Ahlawat RS. Extended daily dialysis in acute renal
failure: a new therapeutic approach. Iran J Kidney Dis. 2007;1:63-72.
5. Cornelis T, van der Sande FM, Eloot S, et al. Acute he-
modynamic response and uremic toxin removal in conventional
and extended hemodialysis and hemodiaﬁltration: a randomized
crossover study. Am J Kidney Dis. 2014;64:247-256.
6. Kielstein JT, Kretschmer U, Ernst T, et al. Efﬁcacy and
cardiovascular tolerability of extended dialysis in critically ill
patients: a randomized controlled study. Am J Kidney Dis.
7. Abe M, Okada K, Suzuki M, et al. Comparison of sustained
hemodiaﬁltration with continuous venovenous hemodiaﬁltration
for the treatment of critically ill patients with acute kidney injury.
Artif Organs. 2010;34:331-338.
8. Schwenger V, Weigand MA, Hoffmann O, et al. Correction:
Sustained low efﬁciency dialysis using a single-pass batch system
in acute kidney injury—a randomized interventional trial: the
REnal Replacement Therapy Study in Intensive Care Unit Pa-
tiEnts. Crit Care. 2012;16:451.
9. Wu VC, Wang CH, Wang WJ, et al. Sustained low-
efﬁciency dialysis versus continuous veno-venous hemoﬁltration
for postsurgical acute renal failure. Am J Surg. 2010;199:466-476.
10. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA
Group. Preferred reporting items for systematic reviews and meta-
analyses: the PRISMA statement. BMJ. 2009;339:b2535.
11. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of
observational studies in epidemiology: a proposal for reporting.
Am J Kidney Dis. 2015;66(2):322-330 329
Extended Dialysis for Acute Kidney Injury
12. Higgins J, Altman DG, Gøtzsche PC, et al; Cochrane Bias
Methods Group; Cochrane Statistical Methods Group. The
Cochrane Collaboration’s tool for assessing risk of bias in rand-
omised trials. BMJ. 2011;343:d5928.
13. Wells G, Shea B, O’Connell D, et al. The Newcastle-
Ottawa Scale (NOS) for assessing the quality of nonrandomised
studies in meta-analyses. http://www.ohri.ca/programs/clinical_
epidemiology/oxford.asp. Accessed February 26, 2015.
14. Higgins JP, Thompson SG. Quantifying heterogeneity in a
meta-analysis. Stat Med. 2002;21:1539-1558.
15. Higgins JP, Thompson SG, Deeks JJ, Altman DG.
Measuring inconsistency in meta-analyses. BMJ. 2003;327:
16. Kontopantelis E, Reeves D. metaan: random-effects meta-
analysis. Stata J. 2010;10:395.
17. Cornell JE, Mulrow CD, Localio R, et al. Random-effects
meta-analysis of inconsistent effects: a time for change. Ann Intern
18. Baldwin I, Bellomo R, Naka T, Koch B, Fealy N. A pilot
randomized controlled comparison of extended daily dialysis with
ﬁltration and continuous veno-venous hemoﬁltration: ﬂuid
removal and hemodynamics. Int J Artif Organs. 2007;30:1083-
19. Abe M, Maruyama N, Matsumoto S, et al. Comparison of
sustained hemodiaﬁltration with acetate-free dialysate and
continuous venovenous hemodiaﬁltration for the treatment of
critically ill patients with acute kidney injury. Int J Nephrol.
20. Shin YB, Cho JH, Park JY, et al. Sustained low-efﬁciency
dialysis as an alternative therapy to continuous renal replacement
therapy in critically ill patients with acute kidney injury. Korean J
21. Badawy SS, Hassan AR, Samir EM. A prospective ran-
domized comparative pilot trial on extended daily dialysis versus
continuous venovenous hemodiaﬁltration in acute kidney injury
after cardiac surgery. Egypt J Cardiothorac Anesth. 2013;7:69-73.
22. Kumar VA, Craig M, Depner TA, Yeun JY. Extended daily
dialysis: a new approach to renal replacement for acute renal
failure in the intensive care unit. Am J Kidney Dis. 2000;36:294-
23. Kumar VA, Yeun JY, Depner TA, Don BR. Extended daily
dialysis vs. continuous hemodialysis for ICU patients with acute
renal failure: a two-year single center report. Int J Artif Organs.
24. Berbece AN, Richardson RM. Sustained low-efﬁciency
dialysis in the ICU: cost, anticoagulation, and solute removal.
Kidney Int. 2006;70:963-968.
25. Marcelino P, Marum S, Fernandes AP, Ribeiro JP. Hybrid
or continuous renal replacement techniques for unstable haemo-
dynamic patients in the intensive care unit. Acta Med Port.
26. Fieghen HE, Friedrich JO, Burns KE, et al. The hemody-
namic tolerability and feasibility of sustained low efﬁciency dial-
ysis in the management of critically ill patients with acute kidney
injury. BMC Nephrol. 2010;11:32.
27. Khanal N, Marshall MR, Ma TM, Pridmore PJ,
Williams AB, Rankin AP. Comparison of outcomes by modality for
critically ill patients requiring renal replacement therapy: a single-
centre cohort study adjusting for time-varying illness severity and
modality exposure. Anaesth Intensive Care. 2012;40:260-268.
28. Chen X, Ma T. Sustained low-efﬁciency daily diaﬁltration
for diabetic nephropathy patients with acute kidney injury. Med
Princ Pract. 2014;23:119-124.
29. Lu R-h, Yan Y-c, Gu Y, et al. Effects of CVVH and SLED
in patients with acute renal failure after liver transplantation.
J Shanghai Jiaotong University (Medical Science)(China). 2008;8:
30. Birne R, Branco P, Marcelino P, et al. A comparative study
of cardiovascular tolerability with slow extended dialysis versus
continuous haemodiaﬁltration in the critical patient. Port J
Nephrol Hypertens. 2009;23:323-330.
31. Schneider AG, Bellomo R, Bagshaw SM, et al. Choice of
renal replacement therapy modality and dialysis dependence after
acute kidney injury: a systematic review and meta-analysis.
Intensive Care Med. 2013;39:987-997.
32. Pannu N, Klarenbach S, Wiebe N, Manns B, Tonelli M;
Alberta Kidney Disease Network. Renal replacement therapy in
patients with acute renal failure: a systematic review. JAMA.
33. Rabindranath K, Adams J, Macleod AM, Muirhead N.
Intermittent versus continuous renal replacement therapy for acute
renal failure in adults. Cochrane Database Syst Rev. 2007;3:
34. Bagshaw SM, Berthiaume LR, Delaney A, Bellomo R.
Continuous versus intermittent renal replacement therapy for
critically ill patients with acute kidney injury: a meta-analysis. Crit
Care Med. 2008;36:610-617.
35. Schefold JC, Haehling S, Pschowski R, et al. The effect
of continuous versus intermittent renal replacement therapy on
the outcome of critically ill patients with acute renal failure
(CONVINT): a prospective randomized controlled trial. Crit Care.
36. Moore I, Bhat R, Hoenich NA, et al. A microbiological
survey of bicarbonate-based replacement circuits in continuous
veno-venous hemoﬁltration. Crit Care Med. 2009;37:496-500.
37. Kanagasundaram NS, Moore I, Hoenich NA. Bioﬁlm in
bicarbonate-based replacement ﬂuid circuits in CVVH. Kidney Int.
38. Fliser D, Kielstein JT. Technology insight: treatment of
renal failure in the intensive care unit with extended dialysis. Nat
Clin Pract Nephrol. 2006;2:32-39.
39. VA/NIH Acute Renal Failure Trial Network; Palevsky PM,
Zhang JH, O’Connor TZ, et al. Intensity of renal support in crit-
ically ill patients with acute kidney injury. N Engl J Med.
40. Faulhaber-Walter R, Hafer C, Jahr N, et al. The Hannover
Dialysis Outcome study: comparison of standard versus intensiﬁed
extended dialysis for treatment of patients with acute kidney injury
in the intensive care unit. Nephrol Dial Transplant. 2009;24:2179-
330 Am J Kidney Dis. 2015;66(2):322-330
Zhang et al