Human albumin for intradialytic hypotension in haemodialysis

patients (Review)

Fortin PM, Bassett K, Musini VM

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2010, Issue 11
http://www.thecochranelibrary.com

Human albumin for intradialytic hypotension in haemodialysis patients (Review)

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . . 18
INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

Human albumin for intradialytic hypotension in haemodialysis patients (Review) i

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
[Intervention Review]

Human albumin for intradialytic hypotension in haemodialysis

patients

Patricia M Fortin1 , Ken Bassett1 , Vijaya M Musini1

1 Department of Anesthesiology, Pharmacology and Therapeutics, University of British Columbia, Vancouver, Canada

Contact address: Patricia M Fortin, Department of Anesthesiology, Pharmacology and Therapeutics, University of British Columbia,
2176 Health Sciences Mall, Vancouver, BC, V6T 1Z3, Canada. patricia@ti.ubc.ca.

Editorial group: Cochrane Renal Group.

Publication status and date: New, published in Issue 11, 2010.
Review content assessed as up-to-date: 29 September 2010.

Citation: Fortin PM, Bassett K, Musini VM. Human albumin for intradialytic hypotension in haemodialysis patients. Cochrane
Database of Systematic Reviews 2010, Issue 11. Art. No.: CD006758. DOI: 10.1002/14651858.CD006758.pub2.

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

ABSTRACT
Background
Intradialytic hypotension (IDH) occurs in 20% to 55% of haemodialysis sessions and is more frequent among patients on long-term
haemodialysis. Symptomatic IDH is generally defined as a decrease in systolic blood pressure (BP) of at least 10 mm Hg or a systolic BP
less than 100 mm Hg, with symptoms such as cramps, nausea, vomiting, and dizziness. IDH is managed acutely by volume expansion
through the intravenous administration of fluids.
Objectives
To compare the benefits and harms of volume expansion with human albumin, alone or in combination with crystalloid or non-protein
colloids, for treating IDH in haemodialysis patients.
Search methods
The Cochrane Renal Group’s Specialised Register and the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane
Library 2010, Issue 9) MEDLINE (1966 to Oct 2009), and EMBASE (1980 to Oct 2009) were searched.
Selection criteria
Randomised controlled trials (RCTs), quasi-RCTs as well as randomised crossover studies were to be included.
Data collection and analysis
Two authors independently extracted data and assessed trial quality. Relative risk (RR) was to be used to analyse dichotomous variables
and mean difference (MD) used to analyse continuous variables.
Main results
One double blind randomised crossover trial met the inclusion criteria and compared 5% albumin to normal saline in patients with a
previous history of IDH. Results from 45 assessable participants did not lead to rejection of the null hypothesis of no difference between
5% albumin and normal saline in the primary outcome measure of percentage target ultrafiltration achieved, nor in 11/12 secondary
outcomes. Additional (unblinded) saline was given less often when 5% albumin was used compared with saline (16% versus 36%, P =
0.04). However, the volume of additional fluid administered was similar in both groups. There were no significant differences in the
nursing time required to treat IDH and the time to restore BP.
Human albumin for intradialytic hypotension in haemodialysis patients (Review) 1
Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Authors’ conclusions

No randomised or controlled trial was identified comparing albumin to crystalloids (other than normal saline) or non-protein colloids, or
a combination of both, in the treatment of symptomatic hypotension during dialysis. One double blind crossover RCT in 45 assessable
patients showed that 5% albumin is not superior to normal saline for the treatment of symptomatic hypotension in maintenance
haemodialysis patients with a previous history of IDH. Given the cost and relative rarity of albumin use compared to saline, saline
should be first line of therapy for treatment of IDH in stable dialysis patients.

PLAIN LANGUAGE SUMMARY

Human albumin for hypotension in haemodialysis patients

Intradialytic hypotension (IDH), which is caused by a decrease in blood volume, is a common complication of haemodialysis treatment.
It is characterized by a sudden drop in blood pressure with symptoms of nausea, sweating, cramping, or dizziness and causes much
discomfort to the patient. IDH is managed by giving fluids to the patient. These can include saline, albumin or other fluids such as
gelatins and starches. Because albumin is a relatively rare and expensive blood product, we asked if albumin had an advantage over
other fluids to treat IDH. We found one trial comparing albumin to normal saline; no trials were found comparing albumin to other
fluids used to treat hypotension. This trial showed no difference between albumin and normal saline in all outcomes except for the
amount of additional saline given, which was less in the group treated with albumin. We concluded that normal saline should be the
first choice for treating IDH.

BACKGROUND bonate in the dialysate solution. The temperature of the dialysate

Intradialytic hypotension (IDH) occurs in 20% to 55% of
haemodialysis sessions and with increasing frequency over time.
IDH can result from an excessive rate or volume of ultrafiltration.
Other contributing factors include left ventricular dysfunction,
reduced cardiac output due to drugs or co-morbid conditions,
autoimmune disorders, or reduced vascular tone. Components of
the dialysis procedure, including the effects of low calcium and
low sodium in dialysates or bio-incompatible dialyser membranes,
may also be contributors to IDH (Emeli 1999; Knoll 2004; van
der Sande 1999; Van der Sande 2000).
Symptomatic IDH is defined as a decrease in systolic blood pres-
sure (BP) of at least 10 mm Hg or a systolic BP less than 100
mm Hg, with symptoms such as cramps, nausea, vomiting, and
dizziness (Irwin 2003). IDH may affect health-related quality of
life and usually requires the administration of intravenous fluids
in addition to increased medical and nursing care. It may cause
early termination of dialysis resulting in inadequate fluid removal
and reduced efficacy of the dialysis therapy. More serious com-
plications such as cardiac or cerebral ischaemia may also develop
(Emeli 1999; Knoll 2004; van der Sande 1999; Van der Sande
2000).
IDH is managed pre-emptively or acutely. Pre-emptive strategies
include using higher concentrations of sodium, calcium and bicar-
Human albumin for intradialytic hypotension in haemodialysis patients (Review)
Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
can also be lowered to increase myocardial contractility and pe-
ripheral vasoconstriction, the ultrafiltration rate can be closely
regulated, or the time on dialysis or cardiovascular tolerance can
be increased with sequential ultrafiltration. Pharmaceutical inter-
ventions include withholding antihypertensive medications before
each treatment, or administration of alpha-adrenergic agonists,
mannitol or sertraline (Irwin 2003).
IDH is managed acutely by volume expansion through the intra-
venous administration of fluids. Fluids most commonly used for
managing IDH include normal saline, hypertonic saline, dextran,
pentastarch, and albumin (Irwin 2003).
Albumin is a human blood protein extracted from plasma. It is
reconstituted with water to reflect normal blood osmolality and
used clinically to replace lost plasma volume. As a protein, it carries
the risk of allergic reactions including anaphylaxis. As a human
blood product, it carries the risk of viral disease transmission (for
example hepatitis A, parvovirus) and prion disease transmission
(for example Creutzfeldt-Jacob disease). In 2005 to 2006, albu-
min cost approximately CAD 66 per 25 grams. In a pilot study
of albumin use in British Columbia (Canada), use of albumin for
IDH accounted for approximately 19% of total in-hospital albu-
min use (Fan 2005).
Recent controversies have surrounded the use of albumin in spe-
2
cific indications. A 1998 Cochrane review, ’Human albumin so-
lution for resuscitation and volume expansion in critically ill pa-
tients’ (Alderson 1998), examined the safety of fluid resuscitation
in patients with hypovolaemia, burns, and hypoproteinaemia by
comparing albumin to no albumin or to crystalloid. The origi-
nal review included 30 randomised controlled trials (RCTs) with
variable duration from one day to an unspecified duration and
found an increase in mortality with the use of albumin as com-
pared to crystalloid solutions in 1419 patients. The pooled relative
risk (RR) of death with albumin administration was 1.68 (95%
confidence interval (CI) 1.26 to 2.23). The pooled difference in
the risk of death with albumin was 6% (95% CI 3% to 9%)
with a fixed-effect model. However, the updated review (Albumin
Review 2004) included the SAFE trial (N = 6997) (SAFE Study
2004), which received 91% of the weight, and found the RR of
death with albumin administration was not significantly different
from that with crystalloids (RR 1.01, 95% CI 0.92 to 1.10). In the
SAFE Study 2004, patients admitted to the ICU were randomised
to receive either albumin (N = 3497) or normal saline (N = 3500)
for intravascular resuscitation. The primary outcome measure was
death from any cause during a 28-day period after randomisation.
There were 726 deaths in the albumin group as compared with
729 deaths in the saline group (RR of death 0.99, 95% CI 0.91 to
1.09; P = 0.87).There were no significant differences in the pro-
portion of patients with new, single-organ or multiple-organ fail-
ure; number of days spent in the ICU; days spent in the hospital;
and days of mechanical ventilation or renal replacement therapy.
Clinical alternatives to albumin infusion include crystalloids and
non-protein colloids. Crystalloid solutions are fluids that contain
a combination of water and electrolytes that approximates plasma
osmolality. These include isotonic solutions such as normal (serum
osmotic) saline and Ringer’s lactate; and hypertonic solutions in-
cluding 1.8%, 3%, 5%, 7.5%, and 10% sodium chloride solu-
tions. Colloid solutions include larger, complex molecules formu-
lated to reflect normal serum osmotic pressure. Non-protein, syn-
thetic colloids include dextrans, gelatin-based products, and com-
plex starches such as pentastarch. The cost of the synthetic non-
protein colloid pentastarch was CA$84 per litre in 2005 to 2006,
for example; and crystalloids were much cheaper than albumin
with normal saline costing CA$1 per litre.
OBJECTIVES
The objective of this review was to compare the benefits and harms
of volume expansion with human albumin alone or in combina-
tion with crystalloid or non-protein colloids for treatment of IDH
in haemodialysis patients.
METHODS
Human albumin for intradialytic hypotension in haemodialysis patients (Review)
Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Criteria for considering studies for this review
Types of studies
All RCTs and quasi-RCTs (RCTs in which allocation to treatment
was obtained by alternation, use of alternate medical records, date
of birth, or other predictable methods). Randomised crossover
studies were also included.
Types of participants
Inclusion criteria
Adult patients undergoing long-term haemodialysis and experi-
encing episodes of symptomatic IDH, defined as a decrease in
systolic BP of at least 10 mm Hg or a systolic BP less than 100
mm Hg, with symptoms such as cramps, nausea, vomiting, and
dizziness.
Exclusion criteria
• Patients on peritoneal dialysis
• Children
• Patients having acute kidney failure or undergoing
continuous renal replacement therapy
• Patients treated with haemofiltration or haemodiafiltration
Types of interventions
• Albumin versus crystalloids (which include isotonic
solutions such as normal (serum osmotic) saline and Ringer’s
lactate; and hypertonic solutions including 1.8%, 3%, 5%,
7.5%, and 10% sodium chloride solutions)
• Albumin versus non-protein colloids (which include
dextrans, gelatin-based products, and complex starches such as
pentastarch)
• Albumin versus both saline and non-protein colloids
Studies that included non-fluid interventions were not included.
Types of outcome measures
• Mortality (all-cause mortality reported at end of follow-up
or a longer duration)
• Non-fatal serious adverse events (such as anaphylaxis)
• Treatment failure, defined as not reaching the target
ultrafiltration goal (target ultrafiltration is determined by the
nephrologist and defined as the difference between the pre-
dialysis weight and the dry weight)
3
Surrogate outcome measures
• Time to restore BP
• Nursing time to treat episode
• Recurrent episodes of hypotension
• Per cent target ultrafiltration achieved (% target
ultrafiltration = actual ultrafiltration volume/target ultrafiltration
volume)
• Percentage of patients with IDH in whom systolic BP
increased by at least 10 mm Hg or systolic BP was greater than
100 mm Hg
• post-dialysis systolic BP (mm Hg)
• post-dialysis diastolic BP (mm Hg)
• Volume of fluid used
• Other interventions to treat hypotension
• Ultrafiltration goal decreased
• Volume of additional fluids given
• Volume of additional saline used
Search methods for identification of studies
1. The Cochrane Renal Group’s Specialised Register and the
Cochrane Central Register of Controlled Trials (CENTRAL)
(The Cochrane Library 2010, Issue 9) were searched. CENTRAL
and the Renal Group Specialised Register contain the results of
handsearching conference proceedings from general and
speciality meetings. This is an ongoing activity across The
Cochrane Collaboration and is both retrospective and
prospective (Master List 2010). Therefore, we did not specifically
search conference proceedings. Please refer to the Cochrane
Renal Review Group’s Module in The Cochrane Library for the
most up-to-date list of conference proceedings.
2. MEDLINE (1966 to Oct 2009) using the optimally
sensitive strategy developed for The Cochrane Collaboration for
the identification of RCTs (Dickersin 1994) with a specific
search strategy developed with input from the Cochrane Renal
Group Trial Search Co-ordinator.
3. EMBASE (1980 to Oct 2009) using a search strategy
adapted from that developed for The Cochrane Collaboration
for the identification of RCTs (Lefebvre 1996) together with a
specific search strategy developed with input from the Cochrane
Renal Group Trial Search Co-ordinator.
4. Reference lists of nephrology textbooks, review articles, and
relevant trials.
5. Letters seeking information about unpublished or
incomplete trials that were sent to investigators known to be
involved in previous trials.
See Appendix 1 for the search terms used.
Data collection and analysis
Human albumin for intradialytic hypotension in haemodialysis patients (Review)
Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Selection of studies
The search strategy described was used to obtain titles and ab-
stracts of studies that were relevant to the review. The titles and ab-
stracts were screened independently by two authors, who discarded
studies that were not applicable, however studies and reviews that
might include relevant data or information on trials were retained
initially. Two authors independently assessed retrieved abstracts
and, if necessary the full text, of these studies to determine which
studies satisfied the inclusion criteria. Disagreements were resolved
in consultation with the third review author.
Data extraction and management
Data extraction was carried out independently by the same authors
using standard data extraction forms. Studies reported in non-En-
glish language journals were to be translated before assessment.
Where more than one publication of one study existed, reports
were to be grouped together and the most recent or most com-
plete data set was to be used. Any discrepancy between published
versions was to be highlighted. Disagreements were resolved in
consultation with the third review author.
Assessment of risk of bias in included studies
The following items were assessed using the risk of bias assessment
tool (Higgins 2008) (see Appendix 2).
• Was there adequate sequence generation?
• Was allocation adequately concealed?
• Was knowledge of the allocated interventions adequately
prevented during the study?
• Were incomplete outcome data adequately addressed?
• Were reports of the study free of suggestion of selective
outcome reporting?
• Was the study apparently free of other problems that could
put it at a risk of bias?
Measures of treatment effect
For dichotomous outcomes (death, treatment failure, non-fatal se-
rious adverse events, recurrent hypotension, systolic BP increased
by at least 10 mm Hg or systolic BP greater than 100 mm Hg,
other interventions to treat hypotension, additional fluids given)
results were to be expressed as RR with 95% CI, where possible.
Where continuous scales of measurement were used to assess the
effects of treatment (percentage of target ultrafiltration achieved,
post-dialysis systolic BP (mm Hg), post-dialysis diastolic BP (mm
Hg), volume of fluid used, time to restore BP, nursing time to treat
episode, ultrafiltration goal decreased, volume of additional saline
used), the mean difference (MD) was used. Standardised mean
difference (SMD) was to be used if included trials used different
scales.
4
Dealing with missing data
Any further information required from the original author was
requested by written correspondence and any relevant information
obtained in this manner was included in the review.
Assessment of heterogeneity
Heterogeneity was analysed using a Chi² test on N-1 degrees of
freedom, alpha = 0.05 for statistical significance, and the I² statistic
was also reported (Higgins 2003).
Assessment of reporting biases
Funnel plots were to be used to assess publication bias (Higgins
2008).
Data synthesis
Data were to be pooled using the random-effects model but the
fixed-effect model was to be analysed to ensure robustness of the
model chosen and susceptibility to outlier results.
Subgroup analysis and investigation of heterogeneity
Subgroup analysis was to be used to explore possible sources of
heterogeneity (for example participants, interventions, and study
quality). Heterogeneity among participants could be related to age
and renal pathology. Heterogeneity in treatments could be related
to prior agent(s) used and the agent, dose, and duration of therapy.
Sensitivity analysis
Conference proceedings, abstracts and poster presentations often
do not provide sufficient details to determine the validity or qual-
ity of the study or its findings. The latter requires analysis of how
the trial was conducted and how the results were analysed. That is,
without publication, safety and efficacy claims may not be repli-
cable or non-refutable. We were to carry out sensitivity analyses
based on publication type (full text versus abstract only). Adverse
effects were to be tabulated and assessed with descriptive tech-
niques as they are likely to be different for the various agents used.
Where possible, the risk difference (RD) with 95% CI was calcu-
lated for each adverse effect, either compared to no treatment or
to another agent.
RESULTS
Description of studies
See: Characteristics of included studies.
Human albumin for intradialytic hypotension in haemodialysis patients (Review)
Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
One study (Knoll 2004) comparing albumin to normal saline met
the inclusion criteria for this review. For details of the included
study, see Characteristics of included studies.
Albumin versus hypertonic saline
No controlled trials were identified that compared albumin with
hypertonic saline for the treatment of symptomatic IDH during
dialysis.
Albumin versus non protein colloids
No controlled trials were identified that compared albumin with
non-protein colloids for the treatment of symptomatic IDH dur-
ing dialysis.
Albumin versus non protein colloid and saline
No controlled trials compared albumin with non-protein colloid
and saline treatments for symptomatic IDH during dialysis.
Albumin versus normal saline
One study comparing albumin to normal saline met the inclu-
sion criteria for this review. The study was a double blind, ran-
domised three-period crossover trial conducted at one centre (the
Ottawa Hospital). Eligible patients were adults who underwent
haemodialysis treatments at the Ottawa Hospital for a minimum
period of three months and who had at least three episodes of IDH
in the 60 days preceding enrolment into the study. Patients with
known sensitivity to albumin were excluded. Seventy-two patients
were randomised to one of two treatment sequences, 37 to treat-
ment sequence 1 and 35 to treatment sequence 2. Patients in se-
quence one received 5% albumin to treat the first episode of IDH,
the second and third dialysis sessions with IDH were treated with
normal saline. Patients in sequence two received normal saline to
treat the first dialysis session with IDH, the second and third dial-
ysis sessions with IDH were treated with 5% albumin.
The primary outcome measure was the percentage of the target
ultrafiltration achieved, which was defined as the actual ultrafil-
tration volume divided by the target ultrafiltration volume. Target
ultrafiltration volume was the difference between the pre-dialysis
weight and the dry weight. Actual ultrafiltration was the difference
between the pre-dialysis and post-dialysis weights.
Secondary outcome measures were post-dialysis systolic and di-
astolic BP, volume of study fluid used, time to restore BP, total
nursing time to treat hypotensive episodes (defined as the time
from onset of symptomatic hypotension until the BP was restored
and the patient no longer needed one-to-one attention), treatment
failures (defined as the inability to restore BP with 750 mL of the
study fluid); recurrent symptomatic hypotension, and the use of
other interventions to treat hypotension (such as unblinded saline
administration).
5
Of the 72 patients randomised, 27 patients were not assessable
because they had no episode (N = 17: sequence one N = 11,
sequence two N = 6) or one or two episodes (N = 10: sequence
one N = 3, sequence two N = 7). Baseline characteristics were
presented for the assessable patients only (N = 45); however, the
authors stated that the randomised groups were well-balanced with
respect to demographic and clinical characteristics.
To determine sample size, a chart review was conducted to de-
termine the mean (SD) of the percentage of target ultrafiltration
achieved with and without hypotension. From an informal survey
of nephrologists it was determined that the minimum increase in
percentage of target ultrafiltration that would change practice was
0.07. Assuming this minimal clinically important difference, a to-
tal of 42 patients needed to be evaluable. There were 45 patients in
the final analysis, which was determined to be adequate to achieve
statistical significance.
Risk of bias in included studies
See ’Risk of bias table’ and Figure 1. This was a well-conducted
trial and appeared to be free of or at least had minimal bias that
could affect the results of the trial.

Figure 1. Methodological quality summary: review authors’ judgements about each methodological quality
item for each included study.

Effects of interventions multiple comparisons. There were no reported withdrawals and

See Table 1 for a summary of results by outcome hierarchy. no deaths. There were no significant carry-over (P = 0.38) or pe-
This RCT of 5% albumin versus normal saline for the treatment riod effects (P = 0.32). Thus, the results of this uniformly negative
of IDH in maintenance haemodialysis patients showed no statis- trial does not reject the null hypothesis of no difference between
tically significant difference in the primary outcome measurement 5% albumin and saline.
of percentage of target ultrafiltration achieved and in 11 of 12 sec- A secondary analysis was also performed, on all patients who re-
ondary outcome measurements. The P value was not adjusted for ceived at least one blinded study fluid (N = 55), with a similar
result to the primary analysis.
Human albumin for intradialytic hypotension in haemodialysis patients (Review) 6
Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Additional (unblinded) saline was given less often when 5% al-
bumin was used compared with saline (16% versus 36%, P =
0.04). However, the volume of additional volume used was similar
whether 5% albumin or saline was administered: 231 mL (136
mL) versus 260 mL (214 mL) (P = 0.41). Otherwise, there were
no significant differences in secondary outcomes.
One argument for the use of albumin has revolved around time
saving in restoring BP in IDH. In this trial, there were no signifi-
cant differences in the nursing time required to treat IDH and the
time required to restore BP.
DISCUSSION
Key appraisal issues
This is a well-conducted trial using a complex three-period
crossover design. Crossover trials are suited to investigating treat-
ments for chronic diseases where there is no hope for cure but a
hope of moderating effects on the disease or, in this case, mod-
erating the effects of a procedure. Certain therapies, those that
have rapid and dramatic effects, also appear to be more suitable for
crossover trials than therapies which have more persistent effects
(Senn 1993). In this trial, it appears that the choice of a crossover
design was appropriate for studying the rapid treatment effects of
albumin and saline for IDH in stable dialysis patients.
Carry-over, or the period of treatment interaction, is considered
to be an unresolved problem of crossover trials. Although there
are tests for carry-over, the reader is warned to be cautious about
these tests as they are impossible to interpret independently of the
treatment effect (Senn 1993). The authors tested for carry-over
and period effects at the request of the editors, and results were
non-significant. This is further supported by the nature of the
three-period design, modelled on the Jones and Kenward crossover
designs, in that it is intended to mitigate carry-over and period
effects so that they are not significant confounders (Jones 2003).
The baseline characteristics in this trial are defined according to
patients randomised to sequence one or two. However, the results
are presented as the albumin 5% group and normal saline group.
Since patients in both sequence one and two receive albumin as
well as saline, treatment results include combining patients from
both groups. In consultation with the author it was determined
that patients are only compared to themselves, thus the unit of
analysis is the patient and the denominator is 45 for both the
Human albumin for intradialytic hypotension in haemodialysis patients (Review)
Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
albumin 5% group and the normal saline group (Altman 1997).
Patients acting as their own controls eliminates between patient
variability and increases the power and efficiency of the trial.
AUTHORS’ CONCLUSIONS
Implications for practice
No randomised or controlled trials were identified comparing al-
bumin to crystalloids (other than normal saline), non-protein col-
loids, or a combination of both in the treatment of symptomatic
hypotension during dialysis. Therefore there is insufficient evi-
dence from randomised or controlled trials to support that albu-
min provides a therapeutic advantage over crystalloids or non-pro-
tein colloids in the treatment of symptomatic hypotension during
dialysis. One double blind randomised crossover RCT in 45 as-
sessable patients showed that 5% albumin is not superior to nor-
mal saline for the treatment of symptomatic hypotension in main-
tenance haemodialysis patients with a previous history of IDH.
Given the cost and relative rarity of albumin compared to saline,
saline should be first-line therapy for treatment of IDH in stable
dialysis patients.
Implications for research
Albumin may be used as prophylaxis for the preservation of blood
volume and BP during haemodialysis in patients thought to be at
risk of hypotension (Van der Sande 2000; van der Sande 1999).
Since this may be a relatively high use area for albumin, RCTs com-
paring albumin to crystalloids or non-protein colloids are needed
to determine if there is a role for albumin as prophylactic therapy
to prevent the reduction in osmolality during dialysis. Further, If a
clinician requires a colloid for treatment of IDH, randomised trials
comparing the efficacy of non-protein colloids such as pentastarch
versus albumin may be useful from a cost and albumin sparing
perspective. Finally, it is not known if albumin confers a therapeu-
tic advantage over crystalloids or other colloids in haemodialysis
patients with rare episodes of hypotension.
ACKNOWLEDGEMENTS
We would like to thank the referees for their editorial advice during
the preparation of this review.
7
 REFERENCES
References to studies included in this review
Knoll 2004 {published data only}
Knoll GA, Grabowski JA, Dervin GF, O’Rourke K. A
randomized, controlled trial of albumin versus saline for
the treatment of intradialytic hypotension. Journal of
the American Society of Nephrology 2004;15(2):487–92.
[MEDLINE: 14747397]
Additional references
Albumin Review 2004
The Albumin Reviewers (Alderson P, Bunn F, Li Wan
Po A, Li L, Roberts I, Schierhout G). Human albumin
solution for resuscitation and volume expansion in critically
ill patients. Cochrane Database of Systematic Reviews
2004, Issue 4. [Art. No.: CD001208. DOI: 10.1002/
14651858.CD001208.pub2]
Alderson 1998
Alderson P, Bunn F, Li Wan Po A, Li L, Roberts I,
Schierhout G. Human albumin solution for resuscitation
and volume expansion in critically ill patients. Cochrane
Database of Systematic Reviews 1998, Issue 3. [DOI:
10.1002/14651858.CD001208]
Altman 1997
Altman DG, Bland JM. Statistics notes. Units of analysis.
BMJ 1997;314(7098):1874. [MEDLINE: 9224131]
Dickersin 1994
Dickersin K, Scherer R, Lefebvre C. Identifying relevant
studies for systematic reviews. BMJ 1994;309(6964):
1286–91. [MEDLINE: 7718048]
Emeli 1999
Emeli S, Black N, Paul R, Rexing C, Ullian M. A protocol-
based treatment for intradialytic hypotension in hospitalized
hemodialysis patients. American Journal of Kidney Diseases
1999;33(6):1107–14. [MEDLINE: 10352199]
Fan 2005
Fan C, Phillips K, Selin S. Serum albumin: new thoughts
on an old treatment. BC Medical Journal 2005;47(8):
438–44. [: BCMJ/2005/oct˙2005/serum˙albumin.asp]
Higgins 2003
Higgins JP, Thompson SG, Deeks JJ, Altman DG.
Measuring inconsistency in meta-analyses. BMJ 2003;327
(7414):557–60. [MEDLINE: 12958120]
Human albumin for intradialytic hypotension in haemodialysis patients (Review)
Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Higgins 2008
Higgins JPT, Green S, editors. Cochrane Handbook for
Systematic Reviews of Interventions Version 5.0.0 [updated
February 2008]. The Cochrane Collaboration. 2008.
Available from www.cochrane-handbook.org.
Irwin 2003
Irwin RS, Rippe JM, editors. Irwin and Rippe’s intensive care
medicine. 5th Edition. Philadelphia: Lippincott, Williams
and Wilkins, 2003.
Jones 2003
Jones B, Kenward M. Design and analysis of crossover trials.
2nd Edition. Boca Raton: Chapman & Hall, 2003.
Lefebvre 1996
Lefebvre C, McDonald S. Development of a sensitive
search strategy for reports of randomized controlled trials
in EMBASE. Fourth International Cochrane Colloquium;
1996 Oct 20-24; Adelaide (Australia). 1996.
Master List 2010
United States Cochrane Center. Master list of journals being
searched. http://apps1.jhsph.edu/cochrane/masterlist.asp
(accessed September 2010).
SAFE Study 2004
Finfer S, Bellomo R, Boyce N, French J, Myburgh J,
Norton R. A comparison of albumin and saline for fluid
resuscitation in the intensive care unit. New England
Journal of Medicine 2004;350(22):2247–56. [MEDLINE:
15163774]
Senn 1993
Senn S. Cross-over trials in clinical research. 1st Edition.
West Sussex: John Wiley, 1993.
van der Sande 1999
van der Sande FM, Kooman JP, Barendregt JN, Nieman
FH, Leunissen KM. Effect of intravenous saline, albumin,
or hydroxyethyl starch on blood volume during combined
ultrafiltration and hemodialysis. Journal of the American
Society of Nephrology 1999;10(6):1303–8. [MEDLINE:
10361869]
Van der Sande 2000
Van der Sande FM Lak JV. Effect of intravenous fluids on
blood pressure course during hemodialysis in hypotensive-
prone patients. Journal of the American Society of Nephrology
2000;11(3):550–5. [MEDLINE: 10703679]
∗
Indicates the major publication for the study
8
CHARACTERISTICS OF STUDIES

Characteristics of included studies [ordered by study ID]

Knoll 2004

Methods • Study design: Double blind randomised three-period crossover study

Participants • Setting: Hospital, single centre

• Country: Canada
• Adult patients who underwent haemodialysis treatments for a minimum of 3
months. All participants needed to have at least 3 symptomatic episodes of IDH within
the 60 days preceding enrolment. Patients with known sensitivity to albumin were
excluded.
• Number: 72
◦ Sequence 1: 37
◦ Sequence 2: 35
• Mean age (± SD)
◦ Sequence 1: 65.4 ± 14.9 years
◦ Sequence 2: 64.5 ± 14.0 years
• Sex (M/F)
◦ Sequence 1: 48% female
◦ Sequence 2: 32% female

Interventions Sequence 1
• 5% albumin to treat the first dialysis episode of hypotension, the 2nd and 3rd
dialysis sessions with hypotension were treated with normal saline.
Sequence 2
• Normal saline to treat the first dialysis episode of hypotension, the 2nd and 3rd
dialysis sessions with hypotension were treated with 5% albumin.

Outcomes • Primary outcome: percentage of target ultrafiltration achieved

◦ definition: % target ultrafiltration achieved = actual ultrafiltration volume/
target ultrafiltration volume.
• Secondary outcomes: post-dialysis systolic and diastolic BP, volume of study fluid
used, time to restore BP, total nursing time to treat hypotensive episode, treatment
failures (inability to restore BP with 750 mL of study fluid), recurrent symptomatic
hypotension, use of other interventions to treat hypotension (additional unblinded
saline).

Notes • 72 patients randomised, 27 patients were not assessable because they had:
◦ No episode of IDH (17): sequence one (11), sequence two (6), or
◦ One or two episodes of IDH (10): sequence one (3), sequence two (7)
• Baseline characteristics are presented for the assessable patients only (45), however
authors state that randomised groups were well-balanced with respect to demographic
and clinical characteristics.
• Target ultrafiltration volume was the difference between the pre-dialysis weight
and the dry weight. Actual ultrafiltration volume was the difference between the pre-
dialysis and post-dialysis weights.

Risk of bias

Human albumin for intradialytic hypotension in haemodialysis patients (Review) 9

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Knoll 2004 (Continued)

Item Authors’ judgement Description

Adequate sequence generation? Yes Statistician generated the randomisation

schedule using a random number table.
The randomised treatment sequence was
assigned to each patient using a unique
identification number

Allocation concealment? Yes For ensuring proper allocation conceal-

ment, the statistician was given only the pa-
tient’s identification number which could
not be linked to any patient

Blinding? Yes 5% albumin and normal saline were placed

All outcomes in identical 250 mL bottles. All bottles were
tightly wrapped with foil so that the con-
tents could not be seen. Plastic intravenous
tubing covered with an opaque sleeve was
supplied with each bottle, a small amount
of multivitamin solution was added to the
saline so that it was identical in colour
to albumin. Participants, investigators, re-
search nurses, dialysis nurses, and attend-
ing physicians were blinded to the coding
of the study solutions. The investigators re-
mained blinded until the analysis of the pri-
mary outcome was completed. No attempt
was made to judge the success of blinding
by asking patients or study personnel which
solution was given

Incomplete outcome data addressed?
All outcomes
Yes 27 patients were not assessable because they
had no hypotensive episodes (17) or only
one or two hypotensive episodes (10) dur-
ing the study period. There 45 patients in
the final analysis, 23 in sequence 1 and 22
in sequence 2

Free of selective reporting? Yes All primary and secondary outcomes in the
methods section are reported on in the re-
sults section

Free of other bias? Yes Authors tested for carry-over and period ef-
fect at the request of editor
Study was sponsored by a grant from the In-
ternational Therapeutic Fluids Group and
Bayer Canada, however the study was de-
signed, analysed, and written by the investi-
gators. The investigator collected all of the

Human albumin for intradialytic hypotension in haemodialysis patients (Review) 10

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Knoll 2004 (Continued)

data and maintained the database indepen-

dent of sponsors

IDH - intradialytic hypotension

Human albumin for intradialytic hypotension in haemodialysis patients (Review) 11

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
DATA AND ANALYSES
This review has no analyses.

ADDITIONAL TABLES
Table 1. Results by outcome hierarchy

Albumin (45) Normal saline (45) P value

Mortality None None

Non-fatal serious adverse events NR NR

Treatment failure (%) 22 24 1.0

Surrogate outcome measures

Time to restore BP (min) 7.9 ± 6.6 9.9 ± 7.5 0.09

Nursing time to treat episode 15.1 ± 7.2 15.9 ± 7.3 0.47

(min)

Recurrent hypotension (%) 36 36 1.0

Actual ultrafiltration 2.6 ± 1.4 2.6 ± 1.2 0.58

% target ultrafiltration achieved 0.84 ± 0.17 0.80 ± 0.16 0.14

% patients with systolic BP in- NR NR

creased by at least 10 mm Hg or
was > 100 mm Hg

Post-dialysis systolic BP (mm 121 ± 9 117 ± 9 0.32

Hg)

Post-dialysis diastolic BP (mm 63 ± 9 61 ± 9 0.33

Hg)

Volume of study fluid used 403 ± 170 428 ± 191 0.34

(mL)

Ultrafiltration goal decreased 11 22 0.27

(%)

Additional saline given (%) 16 36 0.04

Additional albumin given (%) 0 0 1.0

Human albumin for intradialytic hypotension in haemodialysis patients (Review) 12

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Results by outcome hierarchy (Continued)

Volume of additional saline 231 ± 136 260 ± 214 0.74

used (%)

Early termination of dialysis 40 29 0.41

(%)

NR - not reported

APPENDICES
Appendix 1. Electronic search strategies

Database Search terms

CENTRAL 1. MeSH descriptor Albumins, this term only

2. MeSH descriptor Serum Albumin, this term only
3. albumin*:ti,ab,kw
4. MeSH descriptor Fluid Therapy, this term only
5. (#1 OR #2 OR #3 OR #4)
6. MeSH descriptor Hypotension, this term only
7. hypotens*:ti,ab,kw
8. low next blood next pressure:ti,ab,kw
9. (#6 OR #7 OR #8)
10. MeSH descriptor Renal Dialysis, this term only
11. MeSH descriptor Hemofiltration explode all trees
12. (haemodialysis or hemodialysis):ti,ab,kw
13. (haemofiltration or hemofiltration):ti,ab,kw
14. (haemodiafiltration or hemodiafiltration):ti,ab,kw
15. dialysis:ti,ab,kw
16. intradialytic:ti,ab,kw
17. (#10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16)
18. (#5 AND #9 AND #17)

MEDLINE 1. Albumins/
2. Serum Albumin/
3. albumin$.tw.
4. Fluid Therapy/
5. or/1-4
6. Hypotension/
7. hypotens$.tw.
8. low blood pressure.tw.

Human albumin for intradialytic hypotension in haemodialysis patients (Review) 13

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
(Continued)

9. or/6-8
10. Renal Dialysis/
11. exp Hemofiltration/
12. (haemodialysis or hemodialysis).tw.
13. (haemofiltration or hemofiltration).tw.
14. (haemodiafiltration or hemodiafiltration).tw.
15. dialysis.tw.
16. intradialytic.tw.
17. or/10-16
18. and/5,9,17

EMBASE 1. Albumin/
2. Human Albumin/
3. Human Serum Albumin/
4. Serum Albumin/
5. Recombinant Human Serum Albumin/
6. albumin$.tw.
7. or/1-6
8. Hypotension/
9. hypotens$.tw.
10. low blood pressure.tw.
11. or/8-10
12. Hemodialysis/
13. Hemofiltration/
14. Hemodiafiltration/
15. dialysis.tw.
16. (haemodialysis or hemodialysis).tw.
17. (haemofiltration or hemofiltration).tw.
18. (haemodiafiltration or hemodiafiltration).tw.
19. intradialytic.tw.
20. or/12-19
21. and/7,11,20

Appendix 2. Risk of bias assessment tool

Potential source of bias Assessment criteria

Was there adequate sequence generation?
Yes (low risk of bias): random number table; computer random
number generator; coin tossing; shuffling cards or envelopes;
throwing dice; drawing of lots; minimization (minimization may
be implemented without a random element, and this is considered
to be equivalent to being random)

No (high risk of bias): sequence generated by odd or even date of

birth; date (or day) of admission; sequence generated by hospital
or clinic record number; allocation by judgement of the clinician;

Human albumin for intradialytic hypotension in haemodialysis patients (Review) 14

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
(Continued)

by preference of the participant; based on the results of a laboratory

test or a series of tests; by availability of the intervention

Unclear: insufficient information about the sequence generation

process to permit judgement

Was allocation adequately concealed?
Yes (low risk of bias): randomisation method described that would
not allow investigator/participant to know or influence interven-
tion group before eligible participant entered in the study (e.g. cen-
tral allocation, including telephone, web-based, and pharmacy-
controlled, randomisation; sequentially numbered drug contain-
ers of identical appearance; sequentially numbered, opaque, sealed
envelopes)

No (high risk of bias): using an open random allocation schedule

(e.g. a list of random numbers); assignment envelopes were used
without appropriate safeguards (e.g. if envelopes were unsealed or
non-opaque or not sequentially numbered); alternation or rota-
tion; date of birth; case record number; any other explicitly un-
concealed procedure

Unclear: randomisation stated but no information on method used

is available

Was knowledge of the allocated interventions adequately pre- Yes (low risk of bias): no blinding, but the review authors judge
vented during the study? that the outcome and the outcome measurement are not likely
to be influenced by lack of blinding; blinding of participants and
key study personnel ensured, and unlikely that the blinding could
have been broken; either participants or some key study personnel
were not blinded, but outcome assessment was blinded and the
non-blinding of others unlikely to introduce bias

No (high risk of bias): no blinding or incomplete blinding, and the

outcome or outcome measurement is likely to be influenced by
lack of blinding; blinding of key study participants and personnel
attempted, but likely that the blinding could have been broken;
either participants or some key study personnel were not blinded,
and the non-blinding of others likely to introduce bias

Unclear: insufficient information to permit judgement of ‘Yes’ or

‘No’

Were incomplete outcome data adequately addressed?
Yes (low risk of bias): no missing outcome data; reasons for missing
outcome data unlikely to be related to true outcome (for survival
data, censoring unlikely to be introducing bias); missing outcome
data balanced in numbers across intervention groups, with similar
reasons for missing data across groups; for dichotomous outcome
data, the proportion of missing outcomes compared with observed
event risk not enough to have a clinically relevant impact on the

Human albumin for intradialytic hypotension in haemodialysis patients (Review) 15

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
(Continued)

intervention effect estimate; for continuous outcome data, plau-

sible effect size (difference in means or standardized difference in
means) among missing outcomes not enough to have a clinically
relevant impact on observed effect size; missing data have been
imputed using appropriate methods

No (high risk of bias): reason for missing outcome data likely to

be related to true outcome, with either imbalance in numbers or
reasons for missing data across intervention groups; for dichoto-
mous outcome data, the proportion of missing outcomes com-
pared with observed event risk enough to induce clinically relevant
bias in intervention effect estimate; for continuous outcome data,
plausible effect size (difference in means or standardized differ-
ence in means) among missing outcomes enough to induce clini-
cally relevant bias in observed effect size; ‘as-treated’ analysis done
with substantial departure of the intervention received from that
assigned at randomisation; potentially inappropriate application
of simple imputation

Unclear: insufficient information to permit judgement of ‘Yes’ or

‘No’

Are reports of the study free of suggestion of selective outcome Yes (low risk of bias): the study protocol is available and all of the
reporting? study’s pre-specified (primary and secondary) outcomes that are of
interest in the review have been reported in the pre-specified way;
the study protocol is not available but it is clear that the published
reports include all expected outcomes, including those that were
pre-specified (convincing text of this nature may be uncommon)

No (high risk of bias): not all of the study’s pre-specified primary

outcomes have been reported; one or more primary outcomes
is reported using measurements, analysis methods or subsets of
the data (e.g. subscales) that were not pre-specified; one or more
reported primary outcomes were not pre-specified (unless clear
justification for their reporting is provided, such as an unexpected
adverse effect); one or more outcomes of interest in the review are
reported incompletely so that they cannot be entered in a meta-
analysis; the study report fails to include results for a key outcome
that would be expected to have been reported for such a study

Unclear: insufficient information to permit judgement of ‘Yes’ or

‘No’

Was the study apparently free of other problems that could Yes (low risk of bias): the study appears to be free of other sources
put it at a risk of bias? of bias.

No (high risk of bias): had a potential source of bias related to the

specific study design used; stopped early due to some data-de-
pendent process (including a formal-stopping rule); had extreme

Human albumin for intradialytic hypotension in haemodialysis patients (Review) 16

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
(Continued)

baseline imbalance; has been claimed to have been fraudulent; had

some other problem

Unclear: insufficient information to permit judgement of ‘Yes’ or

‘No’

HISTORY
Protocol first published: Issue 4, 2007
Review first published: Issue 11, 2010

CONTRIBUTIONS OF AUTHORS

• Writing of protocol and review: PF, VM, KB

• Screening of titles and abstracts: PF, VM
• Assessment for inclusion: PF, VM
• Quality assessment: PF, VM
• Data extraction: PF, VM
• Data entry into RevMan: PF
• Data analysis: PF, VM, KB
• Disagreement resolution: KB

DECLARATIONS OF INTEREST
None known

SOURCES OF SUPPORT

Internal sources
• UBC Therapeutics Initiative, Canada.
Also Drug Assessment Working Group for their assistance with critical appraisal

Human albumin for intradialytic hypotension in haemodialysis patients (Review) 17

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
External sources
• B.C. Provincial Blood Coordinating Office, Canada.

DIFFERENCES BETWEEN PROTOCOL AND REVIEW

The risk of bias assessment tool was used in place of quality assessment.

INDEX TERMS

Medical Subject Headings (MeSH)

Albumins [∗ therapeutic use]; Double-Blind Method; Hypotension [∗ drug therapy; etiology]; Randomized Controlled Trials as Topic;
Renal Dialysis [∗ adverse effects]; Sodium Chloride [therapeutic use]

MeSH check words

Humans

Human albumin for intradialytic hypotension in haemodialysis patients (Review) 18

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.