ArticlePDF AvailableLiterature Review

Loop diuretics in the management of acute renal failure: A systematic review and meta-analysis

Authors:

Abstract

Loop diuretics are commonly used in critically ill patients with acute renal failure (ARF), but their effect on clinical outcome remains uncertain. We systematically reviewed the literature comparing loop diuretics with control in the management of ARF. Studies were identified by search of MEDLINE, EMBASE, and the Cochrane Controlled Clinical Trials Register, and review of proceedings from selected scientific meetings and clinical trial registries, and bibliographies of retrieved citations. We selected randomised controlled trials (RCTs) comparing loop diuretics with control in patients with ARF. Data were extracted in duplicate by two independent reviewers on study characteristics, quality and outcomes. Primary outcomes were mortality, need for renal replacement therapy (RRT) and renal recovery. Secondary outcomes were change to urine output, serum potassium level and acid-base status, duration of ARF or RRT, length of hospital stay and toxicity. Of 62 studies reviewed, five RCTs, enrolling 555 patients, were eligible and analysed. These trials enrolled a mix of patients, but only two included critically ill patients. Overall trial quality was low. There was no statistical difference in mortality (odds ratio [OR], 1.28; 95% CI, 0.89-1.84; P=0.18) or renal recovery (OR, 0.88; 95% CI, 0.59-1.31; P=0.5) with use of loop diuretics compared with control. However, loop diuretics were associated with a shorter duration of RRT (weighted mean difference, ?1.4 days; 95% CI, ?0.2 to ?2.3 days; P=0.02), shorter time to spontaneous decline in serum creatinine level (weighted mean difference, ?2.1 days; 95% CI, ?0.4 to ?3.7 days; P=0.01) and a greater increase in urine output from baseline (OR, 2.6; 95% CI, 1.4-4.9; P=0.004). Insufficient data were available on acid-base status, hospital length of stay or health costs. Four studies reported toxicity, most commonly transient tinnitus and deafness. Loop diuretics were not associated with improved mortality or rate of independence from RRT, but were associated with shorter duration of RRT and increased urine output. However, these findings have limited relevance to critically ill patients. The relative paucity of high-quality data assessing the value of loop diuretics in ARF for the critically ill suggests a need for a suitably powered randomised trial.
REVIEWS
Critical Care and Resuscitation Volume 9 Number 1 March 2007
60
Crit Care Resusc ISSN: 1441-2772 5 March
2007 9 1 60-68
©Crit Care Resusc 2007
www.jficm.anzca.edu.au/aaccm/journal/publi-
cations.htm
Reviews
Acute renal failure (ARF) affects an estimated 6% of
patients admitted to intensive care
1
and is associated with
increased morbidity, mortality and utilisation of health
resources.
2-5
The use of diuretics is common practice — in
about 60% of critically ill patients with ARF.
6
Frusemide and other loop diuretics can reduce oxygen
demand in the medullary thick ascending loop of Henle
by inhibiting the Na
+
/K
+
/Cl
pump on the luminal cell
membrane surface. Thus, timely administration of loop
diuretics might attenuate renal injury and reduce the
severity of ARF.
7,8
Similarly, loop diuretics may have
additional benefit in patients with ARF by increasing urine
output and thereby facilitating fluid, acid–base and potas-
sium control.
Numerous small, clinical studies of loop diuretics for the
treatment of ARF have been reported without consistent
clinical benefits.
9-20
Moreover, two large observational
studies in critically ill patients have reported discrepant
findings on the effect of loop diuretics on mortality and
renal recovery after ARF.
6,21
A recent meta-analysis con-
cluded that frusemide was not associated with any signifi-
cant clinical benefit and perhaps increased risk of harm.
22
Unfortunately, this meta-analysis included studies where
frusemide was administered to both prevent and treat
ARF, and one study where frusemide was given to both
the treatment and control groups. Further, it included
duplicated control data from a study with three treatment
groups, and estimated the rates of toxicity in another.
Consequently, evidence on the role of diuretics in the
management of ARF, in particular in critically ill patients,
remains uncertain.
In an attempt to overcome the above limitations and as
part of a larger initiative to understand the therapeutic
role of loop diuretics in the management of ARF, we
conducted an up-to-date systematic review and meta-
analysis. We assessed the impact of loop diuretics on
mortality, need for renal replacement therapy and renal
recovery after ARF. We also examined the effect of loop
diuretics on physiological variables, such as urine output,
duration of ARF or renal replacement therapy (RRT),
hospital length of stay, and the occurrence of toxicity.
Finally, we applied a critical care perspective to the current
literature on the use of loop diuretics in the management
of ARF.
ABSTRACT
Background: Loop diuretics are commonly used in critically
ill patients with acute renal failure (ARF), but their effect on
clinical outcome remains uncertain. We systematically
reviewed the literature comparing loop diuretics with control
in the management of ARF.
Methods:
Studies were identified by search of MEDLINE,
EMBASE, and the Cochrane Controlled Clinical Trials Register,
and review of proceedings from selected scientific meetings
and clinical trial registries, and bibliographies of retrieved
citations. We selected randomised controlled trials (RCTs)
comparing loop diuretics with control in patients with ARF. Data
were extracted in duplicate by two independent reviewers on
study characteristics, quality and outcomes. Primary outcomes
were mortality, need for renal replacement therapy (RRT) and
renal recovery. Secondary outcomes were change to urine
output, serum potassium level and acid–base status, duration
of ARF or RRT, length of hospital stay and toxicity.
Results: Of 62 studies reviewed, five RCTs, enrolling 555
patients, were eligible and analysed. These trials enrolled a
mix of patients, but only two included critically ill patients.
Overall trial quality was low. There was no statistical difference
in mortality (odds ratio [OR], 1.28; 95% CI, 0.89–1.84; P =
0.18) or renal recovery (OR, 0.88; 95% CI, 0.59–1.31; P=0.5)
with use of loop diuretics compared with control. However,
loop diuretics were associated with a shorter duration of RRT
(weighted mean difference,
1.4 days; 95% CI, 0.2 to 2.3
days; P = 0.02), shorter time to spontaneous decline in serum
creatinine level (weighted mean difference,
2.1 days; 95%
CI,
0.4 to 3.7 days; P = 0.01) and a greater increase in urine
output from baseline (OR, 2.6; 95% CI, 1.4–4.9; P = 0.004).
Insufficient data were available on acid–base status, hospital
length of stay or health costs. Four studies reported toxicity,
most commonly transient tinnitus and deafness.
Conclusions: Loop diuretics were not associated with
improved mortality or rate of independence from RRT, but
were associated with shorter duration of RRT and increased
urine output. However, these findings have limited relevance
to critically ill patients. The relative paucity of high-quality data
assessing the value of loop diuretics in ARF for the critically ill
Crit Care Resusc 2007; 9: 6068
suggests a need for a suitably powered randomised trial.
Loop diuretics in the management of acute renal failure:
a systematic review and meta-analysis
Sean M Bagshaw, Anthony Delaney, Michael Haase,
William A Ghali and Rinaldo Bellomo
REVIEWS
Critical Care and Resuscitation Volume 9 Number 1 March 2007
61
Methods
Search strategy
Randomised controlled trials (RCTs) of loop diuretics in the
management of ARF were identified by electronic and
manual search strategies. This search was supplemented by
scanning the bibliographies of retrieved articles and review
articles, reviewing conference proceedings of selected sci-
entific meetings and clinical trial registries, and contacting
experts in the field. All languages and types of publications
were considered eligible. The comprehensive search was
performed in April 2006, and an updated verification search
in November 2006.
The databases MEDLINE, EMBASE and the Cochrane
Controlled Clinical Trials Register (from inception to April
2006 for all three) were searched via OVID using an
approach recommended for systematic reviews of ran-
domised trials.
23
MEDLINE was also searched through
PubMed.
24
Three comprehensive search themes were
derived, which were then combined using the Boolean
operator “AND”. The first theme used a recommended
highly sensitive RCT filter.
25
The second theme, diuretic, was
created by a search using an exploded medical subject
headings (MeSH) and textword search for “diuretic” or
“frusemide” or “furosemide” or “lasix” or “loop”. The
third theme, acute renal failure, was created using the
Boolean search term “OR” to search for the following terms
appearing as both exploded MeSH and text words: “acute
renal failure” or “acute renal insufficiency” or “oliguria”.
Study selection
Identified abstracts were initially screened independently
by two of the authors (S M B and A D) to confirm that they
reported original data on the use of loop diuretics in the
management of ARF. The full text articles were retrieved
and assessed to determine if they fulfilled pre-determined
eligibility criteria. The same two authors independently
applied the inclusion criteria to all retrieved articles, with
any disagreements resolved by discussion. A third reviewer
(M H) assessed articles published in German or Russian to
determine eligibility. To be eligible for inclusion, the article
had to fulfil all the following criteria:
study design — randomised clinical trial;
target population — adult patients with established
ARF;
intervention — loop diuretics compared with control;
and
outcome — reported at least one of need for RRT, death
or renal recovery.
Assessment of methodological quality
The methodological quality and validity of the included
studies were assessed using criteria defined a priori. Each
study was assessed for allocation concealment, blinding,
reporting of losses to follow-up or missing outcome
assessments, evidence of important baseline differences
between the groups, analysis on an intention-to-treat
basis and use of a sample size calculation.
26,27
When the
details of the methods for allocation concealment were
not specified or could not be clarified, it was assessed as
absent. Two authors (S M B and A D) independently
assessed methodological quality of the included studies,
with any disagreements resolved through discussion. We
did not calculate quality scores, as many of their compo-
nents were not available, and the validity of adjustment in
meta-analyses by such quality scores, rather than simple
description, has been questioned.
28
Data abstraction
Data were abstracted onto standardised data collection
forms independently by S M B and A D. Any discrepancies
in extracted data were resolved by discussion. Data
extracted included details of study protocol and regimen
for administration of loop diuretics, baseline demographic,
clinical and laboratory characteristics of the study popula-
tion, aetiology of ARF, and the primary outcomes of need
for RRT, death and renal recovery.
Secondary outcome data, when available, were
abstracted on change in urine output, change in renal
function, duration of ARF or duration of need for RRT,
hospital length of stay, and occurrence of drug toxicity.
Additional information was sought if available on acid–
Figure 1. Overview of study selection process
Citations identified and screened
for retrieval: 1336
Potentially relevant citations
identified for further review: 62
Citations deemed not relevant
and excluded: 1274
Citations excluded: 57
31 No randomisation
8 No control group
8 Review articles
5 No acute renal failure
2 Editorials
2 Duplicate publication
1 Animal study
RCTs included in final analysis: 5
REVIEWS
Critical Care and Resuscitation Volume 9 Number 1 March 2007
62
base status, serum potassium levels, oxygenation or need
for mechanical ventilation, and health care costs. In one
study where a loop diuretic was administered by two
different regimens, outcome data were pooled and com-
pared with the control.
29
In another study, both frusemide
and torasemide were used, and, as they are reportedly
equipotent, data were pooled for these arms and com-
pared with the control.
30
Quantitative data synthesis
Agreement on the inclusion of full text studies was
assessed by the
κ statistic. The presence of heterogeneity
across trials for all variables was evaluated using the
χ
2
test
for homogeneity and the I
2
statistic, with an I
2
value
> 50% indicating at least moderate statistical heterogene-
ity.
31
Dichotomous data were combined to estimate the
pooled odds ratio (OR) with 95% CIs using weighted
fixed-effects or random-effects models when appropri-
ate.
32
Continuous outcomes were pooled using standard-
ised mean differences. The potential for publication bias
was assessed using both Egger’s test and a visual assess-
ment of funnel plots for asymmetry.
33,34
All statistical
analyses were performed using Stata version 8.2 (Stata-
Corp, College Station, Tex, USA).
Results
Study selection
Database searches generated 1336 citations. Sixty-two full
text articles were retrieved and reviewed in detail, with five
RCTs fulfilling all eligibility criteria for inclusion.
29,30,35-37
All
five RCTs were identified by the electronic search strategy.
There was excellent agreement between reviewers for study
inclusion (raw agreement = 0.94, chance-corrected agree-
ment
κ = 0.89 [SE, 0.16]). The studies included and reasons
for exclusion are shown in Figure 1.
Study description
A total of 555 patients were studied, with 305 allocated to
receive loop diuretics, and 250 to control groups. Two RCTs
enrolled critically ill patients, but the proportion admitted to
an intensive care unit was not specified.
30,35
The mean (± SD)
age of patients enrolled was 58.7 (
± 0.16) years,
30,35
and
57% were male.
30,35,37
The characteristics of the five trials
are shown in Table 1.
The overall methodological quality and reporting of the
trials was poor, as shown in Table 2. Adequate concealment
of allocation was reported in only one trial, but this trial had
significant differences between groups at baseline in terms
Table 2. Summary of key indicators of trial methodological quality
Study
Allocation
concealment Blinding
Placebo-
controlled
Analysis by
intention-to-
treat
Baseline
imbalance
Pre-defined
outcomes
Loss to
follow-up
Sample size
calculation
Cantarovich
29
No No No No na No na No
Karayannopoulos
36
No No No No na No na No
Kleinknecht
37
No No No No na No No No
Shilliday
30
No Yes Yes No No Yes No No
Cantarovich
35
Yes Yes Yes No Yes Yes No Yes
na = not available or not reported.
Table 1. Study features and patient characteristics at time of trial enrolment
Study Year
No. of
patients
Mean age
(years) Male (%) Baseline kidney function Oliguria (%)
Renal
replacement
therapy (%)
Cantarovich
29
1971 47 na na na 47 (100%)* 47 (100%)
Karayannopoulos
36
1974 20 na na na na na
Kleinknecht
37
1976 66 na 31 (47%) na 66 (100%)
na
Shilliday
30
1997 92 58.8 51 (55%) Creatine clearance < 10 mL/min na 0
Cantarovich
35
2004 330 58.5 223 (68%) Serum creatinine, 406 μmol/L
Urea, 12.8 mmol/L
145 (44%) 330 (100%)
* Oliguria defined as urine output < 400 mL/day. Oliguria defined as urine output < 500 mL/day or < 20 mL/h after volume expansion. ‡ Mean values.
Year = year of publication. na = not available or not reported.
REVIEWS
Critical Care and Resuscitation Volume 9 Number 1 March 2007
63
of severity of kidney dysfunction and the proportion of
participants with sepsis.
35
The aetiology of ARF was categorised as medical in 367
patients (66%), surgical in 130 (23%), obstetric in 37 (7%),
and unspecified in 21 (4%). A total of 189 (34%) patients
had sepsis as a potential contributing factor for ARF. One
trial reported that 38% of patients had sepsis, most of
whom had septic shock.
35
Baseline kidney function was reported in two trials.
30,35
However, in three trials
29,35,37
most patients were already
receiving RRT at the time of enrolment. This suggests that
nearly all patients had established and severe ARF. Oliguria
was variably defined and present in 58% of patients across
three trials.
29,35,37
The average duration of oligo–anuria
before enrolment was 2.9 days in one trial,
29
whereas
another reported that 53% of patients had oliguria for 2
days before trial entry.
37
Administration of loop diuretics
The details of how loop diuretics were administered are
shown in Table 3. Frusemide was administered in all
trials. One trial compared torasemide with frusemide and
placebo.
30
Loop diuretics were generally administered as
a bolus dose by the intravenous route. In two studies, the
bolus infusion was given over several hours when the
dose was large.
29,35
The average daily dose of loop
diuretic delivered varied and was estimated between 600
mg and 3200 mg or 12 mg/kg.
29,30,35,36
Four trials had
confounding co-interventions in addition to loop diuret-
ics, including mannitol,
29,30,37
dopamine
30
or RRT.
29,35
Only
three trials set a target diuresis range. This varied from 20
to 100 mL/h.
29,35,37
Evidence synthesis
Mortality
Overall mortality was 40% (213/535)
29,30,35,37
(one study did
not report mortality
36
) (Table 4). The odds for death was
slightly higher with loop diuretics compared with the
control, but the pooled OR was not statistically different
(OR, 1.28; 95% CI, 0.89–1.84; P = 0.18) (Figure 2). There
was no evidence of statistical heterogeneity across studies
Table 3. Details of protocol for administration of loop diuretics in acute renal failure
Study Loop diuretic Protocol for diuretic administration Target diuresis Additional interventions
Cantarovich
29
Frusemide Fixed frusemide (600 mg/24 h IV*) v
progressive frusemide (100–3200 mg/24 h IV*)
v placebo
2000 mL/24 h All received mannitol and
conventional intermittent
haemodialysis
Karayannopoulos
36
Frusemide 1000 mg IV, titrated to a maximum 3000 mg 7days na na
Kleinknecht
37
Frusemide 3 mg/kg IV bolus, then 1.5–6.0 mg/kg every 4 h 20–100 mL/h Mannitol and renal
replacement therapy
Shilliday
30
Frusemide or
torasemide
3mg/kg IV over 1h every 6h21 days
and tapered with evidence of recovery
na Dopamine 2 μg/kg/min and
mannitol 20% 100 mL
every 6h
3 days
Cantarovich
35
Frusemide 25 mg/kg/day IV or 35 mg/kg/day orally
and tapered with evidence of recovery
> 2000 mL/24 h Renal replacement therapy
* Frusemide was administered IV over a duration of 30 min to 10 h, depending on the prescribed dose. IV = intravenous. na = not reported or not available.
Table 4. Summary of primary and secondary outcomes
Need for RRT (%) Mortality (%) Renal recovery (%)
Increased urine
output (%)
Duration of
ARF (days)
Study Loop Control Loop Control Loop Control Loop Control Loop Control
Cantarovich
29
34/34 (100) 13/13 (100) 15/34 (44) 7/13 (54) na na 22/34 (65)* 7/13 (54)* na na
Karayannopoulos
36
1/10 (10)
7/10 (70)
na na na na na na na na
Kleinknecht
37
56/66 (85) 13/33 (39) 12/33 (36) na na 10/33 (30)
6/33 (18)
19.5
§
22.1
§
Shilliday
30
21/62 (34) 12/30 (40) 42/62 (68) 15/30 (50) 14/62 (23)
7/30 (23)
32/62 (52) 7/30 (23) 5.66
§
7.6
§
Cantarovich
35
166/166 (100) 164/164 (100) 59/166 (36) 50/164 (30) 82/166 (49) 87/164 (53) na na 11.4
§
12.4
§
* Urine output increased to 400 mL/24 h. † RRT was peritoneal dialysis. ‡ Urine output increased to 500 mL/24 h.
§ Duration of ARF defined by time to reach a spontaneous decline in serum creatinine or serum urea level without further need for RRT.
Recovery defined as spontaneous fall in serum creatinine level without further need for RRT.
RRT=renal replacement therapy. ARF = acute renal failure. Loop = loop diuretic. na = not available or not reported.
REVIEWS
Critical Care and Resuscitation Volume 9 Number 1 March 2007
64
(χ
2
P = 0.5; I
2
= 0), or publication bias by either Egger’s test
(coefficient = 0.4; P = 0.5) or funnel plot asymmetry.
Need for renal replacement therapy
Renal replacement therapy was received by 85% (474/555)
of enrolled patients (Table 4). The predominant modality of
RRT was conventional intermittent haemodialysis, in 78%
(n = 371), whereas 20% (n = 95) received continuous or
daily intermittent haemodialysis, and 2% (n =8) received
peritoneal dialysis.
It should be emphasised that, in two of the five trials, all
patients were already receiving RRT at the time of randomi-
sation,
29,35
while in another trial, 85% received RRT,
although the timing of initiation and what proportion were
allocated to loop diuretic or control were not specified.
37
Thus, the need for RRT after enrolment could be formally
assessed in only two of the five trials, one of which used
peritoneal dialysis alone.
36
In those two studies, RRT was received by 37% (41/112).
There was a reduced odds for receiving RRT with loop
diuretics compared with control, but the pooled OR was not
statistically significant (OR, 0.50; 95% CI, 0.23–1.10; P =
0.09).
30,36
Further, these trials showed evidence of statistical
heterogeneity (
χ
2
P = 0.04; I
2
= 77%). One trial reported a
lack of significant statistical difference in the time to
initiation of RRT (5.2 days for loop versus 2.8 days for
placebo, P = 0.35), despite an apparent > 2 day difference in
delay in those receiving loop diuretics.
30
Renal recovery
The definitions of renal recovery differed between trials.
Only two trials reported the proportion of patients recov-
ering renal function.
30,35
In these trials, renal recovery was
defined as a spontaneous decrease in serum creatinine
level independent from RRT.
30,35
The odds of renal recovery
was slightly lower in patients receiving loop diuretics
compared with control, but the resulting pooled OR was
not statistically different (OR, 0.88; 95% CI, 0.59–1.31;
P = 0.5). There was no evidence of statistical heterogeneity
(
χ
2
P =0.9; I
2
=0).
However, loop diuretics were associated with a shorter time
to spontaneous decline in either serum creatinine
30
or urea
37
level, or a reduction in serum creatinine level to < 200 μmol/L
35
compared with the control (weighted mean difference, 2.1
days; 95% CI,
0.4 to 3.7 days; P= 0.01). There was no
evidence of statistical heterogeneity (
χ
2
P= 0.9).
Additionally, loop diuretics were associated with a shorter
duration of RRT (weighted mean difference,
1.4 days;
95% CI,
0.2 to 2.3 days; P = 0.02) compared with
control.
30,35
There was no evidence of statistical heterogene-
ity (
χ
2
P = 0.12). Two trials also reported fewer sessions of
RRT with loop diuretics compared with control (5.0–5.5
sessions for loop diuretics versus 6.1–8.0 sessions for
control).
29,37
Secondary outcomes
Both the definition and method of measuring urinary
output varied across trials. Loop diuretics were associated
with an increase in urine output, defined as diuresis 400–
500 mL/day, when compared with control (OR, 2.6; 95% CI,
Figure 3. Forest plot for urine output associated
with loop diuretics compared with control from
three randomised controlled trials
Odds ratio
Favours control Favours diuretics
0.05 0.1
0.25 0.5 1 2 4 8
Study Odds ratio (95% CI) % Weight
Cantarovich (1971)
1.57 (0.43 − 5.75) 29.7%
Kleinknecht (1976)
1.96 (0.62 − 6.21) 34.8%
Shilliday (1997) 3.99 (1.49 − 10.66) 35.5%
Overall
2.56 (1.35 − 4.85) 100.0%
Figure 2. Forest plot for mortality associated with
loop diuretics compared with control from four
randomised controlled trials
Odds ratio
Favours diuretics Favours control
0.05 0.1 0.25
0.5
1
2 4 8
Study
Odds ratio (95% CI) % Weight
Cantarovich (1971)
0.68 (0.19 − 2.44) 10.9%
Kleinknecht (1976)
1.14 (0.42 − 3.08) 14.0%
Shilliday (1997)
2.10 (0.86 − 5.12) 12.6%
Cantarovich (2004)
1.26 (0.79 − 1.99) 62.5%
Overall
1.28 (0.89 − 1.84) 100.0%
REVIEWS
Critical Care and Resuscitation Volume 9 Number 1 March 2007
65
1.4–4.9; P = 0.004).
29,30,37
There was no evidence of statisti-
cal heterogeneity (
χ
2
P = 0.5; I
2
= 0) (Figure 3). Similarly, three
trials showed that time to achieve a urine output goal
1.5–2.0 L/day was shorter with loop diuretics compared
with control (weighted mean difference of
3.1 days; 95%
CI,
1.0 to 5.3 days; P = 0.005).
29,35,37
However, there was
evidence of statistical heterogeneity (
χ
2
P = 0.02). During
renal recovery, one trial described higher measures in both
serum creatinine and urea levels in those allocated loop
diuretics compared with control.
35
Adverse or toxic effects were reported in four
trials
29,30,35,37
(Table 5). However, reporting was inconsistent,
and valid estimates of occurrence could not be determined.
While two trials described transient episodes of tinnitus,
deafness and vertigo following large intravenous doses of
loop diuretics, neither provided data on rates of occur-
rence.
29,37
In the two largest studies (n = 422), four episodes
of deafness were described, but tinnitus was either not
described or did not occur.
30,35
Additional effects described included a higher occurrence
of seizures,
30
polyuria
35
and trend toward hypokalaemia
35
in
patients receiving loop diuretics. No trial described changes
in acid–base status. Likewise, no trial described the impact
of diuretics on oxygenation or on the need for, or weaning
from, mechanical ventilation. Finally, insufficient data were
available for analysis or comment on length of hospitalisa-
tion or health care costs.
Discussion
We performed a systematic review and meta-analysis of all
randomised controlled trials describing the use of loop
diuretics compared with control in the management of ARF.
Our primary objective was to determine whether, in estab-
lished ARF, loop diuretics influence mortality, the need for
RRT or the rate of renal recovery. We found that loop
diuretics failed to show a significant reduction in mortality or
increase in renal recovery, while assessment of the need for
RRT was limited, and this need could not be reliably deter-
mined. On the other hand, our findings suggest that loop
diuretics may be associated with a reduction in time to
spontaneous decline in serum creatinine or urea level,
reduced duration of RRT and promotion of a modest increase
in urine output. However, these findings require careful
interpretation as we also identified a number of limitations to
the available literature on this topic. Specifically, few studies
fulfilled our eligibility criteria, and those that did were
generally small, several were published over three decades
ago, few enrolled critically ill patients, and all had noteworthy
methodological weaknesses or shortcomings.
The observed 40% hospital mortality across these trials
is considerably lower than the more recent estimates of
60% mortality for ARF in critically ill patients.
1,38,39
This
may indicate that patients in these trials had a lower
severity of illness, or isolated renal injury rather than ARF
in the setting of critical illness and multiorgan system
failure (MOSF), as is now more commonly seen.
1
Impor-
tantly for critical care physicians, the trials published over
three decades ago did not enrol critically ill patients. The
remaining two trials did not report the proportion of
enrolled patients admitted to an ICU.
30,35
Furthermore, it is
plausible to consider that the use of loop diuretics alone
will by no means result in a direct increase in survival in
ARF in the setting of severe sepsis or MOSF. Survival may
not be an appropriate primary outcome measure for loop
diuretics; a more relevant outcome might rather be abso-
lute need for, and duration and timing of RRT.
While 85% of patients included in this meta-analysis
received RRT, most in the form of conventional intermit-
tent haemodialysis, it should be highlighted that nearly all
Table 5. Summary of reported adverse effects
Adverse effect Cantarovich
29
Karayannopoulos
36
Kleinknecht
37
Shilliday
30
* Cantarovich
35
*
Tinnitus Yes
na Yes
na No
Hypoacusia na na Yes
Loop (1) v control (0) Loop (3) v control (1)
Vertigo na na Yes
na No
Allergic reaction na na na na Loop (1) v control (2)
Seizures na na na Loop (12) v control (1) No
Polyuria na na No na Loop (7) v control (0)
Hypokalaemia na na na na Loop (11) v control (4)
Agranulocytosis na na na na Loop (1) v control (0)
* Reported as absolute number(s) with adverse effect. † Tinnitus was reported, but no data were presented on rate of occurrence. It was reported as common
in patients receiving intravenous frusemide 3200 mg in less than 4 h. Toxicity resolved without sequelae.
Tinnitus, deafness and vertigo were reported, but no data were presented on rate of occurrence. Symptoms were reported to occur temporally a few hours
after frusemide administration. Ototoxicity may have been confounded by concomitant use of aminoglycosides. No long-term sequelae were reported.
na= not reported or not available.
REVIEWS
Critical Care and Resuscitation Volume 9 Number 1 March 2007
66
patients were receiving RRT at the time of randomisation.
This is of concern because early intervention would likely
be key to deriving clinical benefit from any therapy.
Moreover, we are unable to conclude from our findings
whether loop diuretics are an appropriate alternative to
the timely implementation of RRT in ARF. Available evi-
dence shows that delay in the provision of RRT can result
in higher mortality and delayed or lower likelihood for
recovery of renal function.
38,40
Several studies included in
this meta-analysis described prolonged durations of oligo–
anuria or delay after the diagnosis of ARF before initiation
of RRT, which would not occur in modern ICU practice. On
the other hand, patients who require RRT because of
severe ARF have been identified as having a higher
mortality.
41,42
However, whether loop diuretics can obviate
the need for RRT, and thus have the potential to reduce
morbidity (eg, need for catheter insertion and anticoagu-
lation) and mortality cannot be reliably determined from
our findings. This may, in part, account for the association
between loop diuretics and higher mortality found in a
large observational study.
21
The use of loop diuretics to
avert initiation of RRT when the latter is indicated may
increase renal injury, metabolic derangement and fluid
balance complications, and contribute to higher mortality
and non-recovery of renal function.
21
Overall, these trials
failed to adequately address whether loop diuretics can
obviate or delay the need for RRT.
Likewise, the assorted definitions of renal recovery after
ARF resulted in significant uncertainty about the influence
of loop diuretics on this outcome. For example, while the
pooled estimate of the proportion of patients recovering
function suggested no benefit, use of loop diuretics was
found to be associated with shorter duration and absolute
number of sessions of RRT, and shorter time to achieve a
spontaneous decline in surrogate markers of renal function
in patients not yet receiving RRT.
Our study also examined several secondary outcomes of
interest. Specifically, loop diuretics were found to promote
both an absolute increase in urine output from baseline and
a shorter time to achieve diuresis. We were unable to
establish whether these physiological changes carried any
non-renal clinical benefits, such as shorter duration of
mechanical ventilation or ICU stay. While toxicity was
described with use of loop diuretics, most often tinnitus and
temporary deafness, no reliable estimates of occurrence
could be determined. Toxicity was generally uncommon and
appeared restricted to patients receiving large intravenous
doses in short periods of time. Likewise, no data were
available on the effect of loop diuretics on length of
hospital stay or associated health care costs.
Two previous systematic reviews evaluated diuretics in
ARF.
22,43
One review focused largely on dopamine in ARF
and was unable to include more recent publications.
43
In
the recent meta-analysis by Ho and Sheridan, there were
several concerns: study selection, the pooled analysis and
the global inferences based on their data.
22
In particular,
this meta-analysis pooled data from trials assessing loop
diuretics for both prevention and treatment of ARF.
Further and importantly, the global conclusions from this
meta-analysis do not appear firmly supported by the
available data because of limitations in individual trial
quality and poor reporting.
22
Our study highlights the
weakness of the available body of evidence and the
dubious nature of any firm conclusions on the efficacy of
loop diuretics in ARF.
Two large observational studies have now reported differ-
ing conclusions as to the effect of loop diuretics in critically
ill patients with ARF. One study suggested the use of
frusemide resulted in an increased risk of death, but this
finding was not corroborated by the second larger study.
6,21
The contrasting findings in these prospective cohort studies
may be the consequence of residual confounding, despite
propensity-adjusted and multivariate analysis, or informa-
tion bias, or may simply reflect significant differences in
patterns of practice. These conflicting findings suggest
there is evidence of genuine equipoise on whether loop
diuretics have a role in the management of ARF. As such,
our findings support a very different and more careful
global conclusion when compared with the meta-analysis
of Ho and Sheridan. Specifically, we can only comment that,
based on the available data, it is not possible to draw any
strong conclusions about how loop diuretics may modify
outcome in ARF. Two small pilot RCTs are underway that will
assess the impact of a frusemide infusion on kidney func-
tion early in ARF and after discontinuation of RRT, both in
critically ill patients. It is hoped these studies will provide
further insight on the potential role of diuretics in ARF and
a rationale for a larger study.
Our study had limitations. First, there were few trials,
each with methodological and quality concerns. Second,
several outcomes were variably defined or inconsistently
reported, which could unduly influence event rates and
effect estimates. Third, patients included in these trials
generally had low severity of illness, and few were admitted
to an ICU. Fourth, findings from these trials are likely
confounded by concomitant co-interventions, such as man-
nitol or dopamine. Fifth, loop diuretics were predominantly
administered by large intravenous boluses rather than
continuous infusion and, in several trials, were not titrated
to a physiological end-point or target urine output. Finally,
many trials had evidence of a prolonged period of oliguria
before intervention with loop diuretics. In total, these
observations strongly suggest a lack of generalisability to
critically ill patients.
REVIEWS
Critical Care and Resuscitation Volume 9 Number 1 March 2007
67
To overcome these limitations and obtain high-quality
evidence on whether loop diuretics have a role in the
management of ARF, a suitably powered multicentre RCT is
needed. This trial should ideally incorporate clinically rele-
vant and patient-centred outcomes, such as need for RRT,
or renal recovery, as well as important secondary outcomes
focused on issues of harm, dose response and physiological
end-points (eg, urinary output). The trial should also include
stratification for the presence of septic ARF, early interven-
tion specifically in patients at risk or with evidence of acute
renal injury,
44
and continuous and physiologically targeted
infusion of loop diuretic or placebo with close hourly
monitoring of urine output and fluid balance. Ideally, the
study would focus on critically ill patients. If such a trial
were to use need for RRT as a primary outcome, assuming
an estimated 15% need for RRT in such patients,
45
with a
90% power to detect a 5% absolute reduction, about 2000
patients would need to be enrolled. Considering the low
cost of loop diuretics and the high cost of RRT, such a
difference would likely have clinical relevance.
Conclusions
In summary, there is a paucity of reliable and high-quality
data assessing the value of loop diuretics in the manage-
ment of ARF. As a result, the findings in this systematic
analysis have uncertain significance and require cautious
interpretation. Loop diuretics failed to improve mortality or
RRT independence. Yet, use of loop diuretics was associated
with a modest increase in urine volume and a shorter
duration of RRT. Importantly for ICU clinicians, these studies
have limited relevance to critically ill patients, who were not
included in many of these trials, and who are both unlikely
to receive such delayed intervention, and more likely to
receive loop diuretics by continuous infusion. This uncer-
tainty surrounding a seemingly common intervention sug-
gests the urgent need for a suitably powered RCT focused
on a critically ill population.
Acknowledgements
Sean Bagshaw is supported by Clinical Fellowships from the
Canadian Institutes for Health Research and the Alberta Heritage
Foundation for Medical Research (AHFMR), and a Royal College of
Physicians and Surgeons of Canada Detweiler Traveling Fellow-
ship. Michael Haase is supported by a Feodor-Lynen Research
Fellowship from the Alexander von Humboldt Foundation. William
Ghali is supported by a Government of Canada Research Chair in
Health Services Research and by a Health Scholar Award from the
AHFMR.
Author details
Sean M Bagshaw, Research Fellow
1
Anthony Delaney, Intensive Care Specialist,
2
and Senior Lecturer
3
Michael Haase, Research Fellow
1
William A Ghali, Internist and Professor of Medicine
4
Rinaldo Bellomo, Intensive Care Specialist and Professor of Medicine
1
1 Department of Intensive Care, Austin Hospital, Melbourne, VIC.
2 Department of Intensive Care, Royal North Shore Hospital, Sydney, NSW.
3 Northern Clinical School, University of Sydney, Sydney, NSW.
4 Departments of Medicine and Community Health Sciences, and
Centre for Health and Policy Studies, Calgary Health Region and
University of Calgary, Calgary, Alberta, Canada.
Correspondence: sean.bagshaw@austin.org.au
References
1 Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure in critically
ill patients: a multinational, multicenter study. JAMA 2005; 294:
813-8.
2 Korkeila M, Ruokonen E, Takala J. Costs of care, long-term prognosis
and quality of life in patients requiring renal replacement therapy
during intensive care. Intensive Care Med 2000; 26: 1824-31.
3 Liano F, Junco E, Pascual J, et al. The spectrum of acute renal failure
in the intensive care unit compared with that seen in other settings.
The Madrid Acute Renal Failure Study Group. Kidney Int Suppl
1998; 66: S16-24.
4 Manns B, Doig CJ, Lee H, et al. Cost of acute renal failure requiring
dialysis in the intensive care unit: clinical and resource implications
of renal recovery. Crit Care Med 2003; 31: 449-55.
5 Morgera S, Kraft AK, Siebert G, et al. Long-term outcomes in acute
renal failure patients treated with continuous renal replacement
therapies. Am J Kidney Dis 2002; 40: 275-9.
6 Uchino S, Doig GS, Bellomo R, et al. Diuretics and mortality in acute
renal failure. Crit Care Med 2004; 32: 1669-77.
7 Kramer HJ, Schuurmann J, Wassermann C, Dusing R. Prostaglan-
din-independent protection by furosemide from oliguric ischemic
renal failure in conscious rats. Kidney Int 1980; 17: 455-64.
8 DeTorrente A, Miller PD, Cronin RE, et al. Effects of furosemide and
acetylcholine in norepinephrine-induced acute renal failure. Am J
Physiol 1978; 235: F131-6.
9 Beaufils F, de Myttenaere S, Rohan J, Chapman A. [Effects of
furosemide in patients with acute renal failure] [French]. Decrease
of resort to dialysis methods. Nouv Presse Med 1972; 1: 1073-8.
10 Cantarovich F, Galli C, Benedetti L, et al. High dose frusemide in
established acute renal failure. BMJ 1973; 4: 449-50.
11 Chandra M, Agarwal SS, Mitra NK, Gupta NN. Massive intravenous
frusemide therapy in acute renal failure. J Indian Med Assoc 1975;
64: 47-8.
12 Humbert G, Fillastre JP, Leroux G. [Treatment of anuria with
furosemide] [French]. Nouv Presse Med 1972; 1: 526.
13 Kjellstrand CM. Ethacrynic acid in acute tubular necrosis. Indica-
tions and effect on the natural course. Nephron 1972; 9: 337-48.
14 Lemieux G, Talbot P Jr, Laurent G. [Use of ethacrynic acid for acute
renal failure]. Union Med Can 1971; 100: 1539-44.
15 Lucas CE, Zito JG, Carter KM, et al. Questionable value of
furosemide in preventing renal failure. Surgery 1977; 82: 341-20.
REVIEWS
Critical Care and Resuscitation Volume 9 Number 1 March 2007
68
16 Seedat YK. High-dose furosemide (Lasix) in renal insufficiency. S Afr
Med J 1972; 46: 1371-4.
17 Seedat YK. High dose frusemide in established acute renal failure
[letter]. BMJ 1973; 4: 784.
18 Sirivella S, Gielchinsky I, Parsonnet V. Mannitol, furosemide, and
dopamine infusion in postoperative renal failure complicating
cardiac surgery. Ann Thorac Surg 2000; 69: 501-6.
19 Weinstein JM, Heyman S, Brezis M. Potential deleterious effect of
furosemide in radiocontrast nephropathy. Nephron 1992; 62: 413-5.
20 Zito J, Cortez A, Lucas CE, et al. Distribution of renal blood flow
after furosemide in critically ill patients. Surg Forum 1976; 27: 27-9.
21 Mehta RL, Pascual MT, Soroko S, et al. Diuretics, mortality, and
nonrecovery of renal function in acute renal failure. JAMA 2002;
288: 2547-53.
22 Ho KM, Sheridan DJ. Meta-analysis of frusemide to prevent or treat
acute renal failure. BMJ 2006; 333: 420.
23 Lefebvre C, Clarke M. Identifying randomized trials. In: Egger M,
Smith G, Altman D, editors. Systematic reviews in health care. 2nd
ed. London: BMJ Publishing, 2001: 69-86.
24 Robinson KA, Dickersin K. Development of a highly sensitive search
strategy for the retrieval of reports of controlled trials using
PubMed. Int J Epidemiol 2002; 31: 150-3.
25 Dickersin K, Scherer R, Lefebvre C. Identifying relevant studies for
systematic reviews. BMJ 1994; 309: 1286-91.
26 Moher D, Cook DJ, Jadad AR, et al. Assessing the quality of reports
of randomised trials: implications for the conduct of meta-analyses.
Health Technol Assess 1999; 3: i-iv, 1-98.
27 Shea B, Dube C, Moher D. Assessing the quality of reports of
systematic review: the QUOROM statement compared with other
tools. In: Egger M, Smith G, Altman D, editors. Systematic reviews
in health care. 2nd ed. London: BMJ Publishing, 2001: 122-39.
28 Herbison P, Hay-Smith J, Gillespie WJ. Adjustment of meta-analyses
on the basis of quality scores should be abandoned. J Clin
Epidemiol 2006; 59: 1249-56.
29 Cantarovich F, Fernandez JC, Locatelli A, Perez Loredo J. Frusemide
in high doses in the treatment of acute renal failure. Postgrad Med
J 1971; 47 Suppl: 13-7.
30 Shilliday IR, Quinn KJ, Allison ME. Loop diuretics in the manage-
ment of acute renal failure: a prospective, double-blind, placebo-
controlled, randomized study. Nephrol Dial Transplant 1997; 12:
2592-6.
31 Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-
analysis. Stat Med 2002; 21: 1539-58.
32 DerSimonian R. Combining evidence from clinical trials. Anesth
Analg 1990; 70: 475-6.
33 Dear K, Begg C. An approach for assessing publication bias before
performing a meta-analysis. Stat Sci 1992; 7: 237-45.
34 Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-
analysis detected by a simple, graphical test. BMJ 1997; 315: 629-34.
35 Cantarovich F, Rangoonwala B, Lorenz H, et al. High-dose furosem-
ide for established ARF: a prospective, randomized, double-blind,
placebo-controlled, multicenter trial. Am J Kidney Dis 2004; 44:
402-9.
36 Karayannopoulos S. High-dose frusemide in renal failure [letter].
BMJ 1974; 2: 278-9.
37 Kleinknecht D, Ganeval D, Gonzalez-Duque LA, Fermanian J.
Furosemide in acute oliguric renal failure. A controlled trial. Neph-
ron 1976; 17: 51-8.
38 Bagshaw SM, Laupland KB, Doig CJ, et al. Prognosis for long-term
survival and renal recovery in critically ill patients with severe acute
renal failure: a population-based study. Crit Care 2005; 9: R700-9.
39 Ympa YP, Sakr Y, Reinhart K, Vincent JL. Has mortality from acute
renal failure decreased? A systematic review of the literature. Am J
Med 2005; 118: 827-32.
40 Gettings LG, Reynolds HN, Scalea T. Outcome in post-traumatic
acute renal failure when continuous renal replacement therapy is
applied early vs. late. Intensive Care Med 1999; 25: 805-13.
41 Bagshaw SM, Mortis G, Doig CJ, et al. One-year mortality in
critically ill patients by severity of kidney dysfunction: a population-
based assessment. Am J Kidney Dis 2006; 48: 402-9.
42 Metnitz PG, Krenn CG, Steltzer H, et al. Effect of acute renal failure
requiring renal replacement therapy on outcome in critically ill
patients. Crit Care Med 2002; 30: 2051-8.
43 Kellum JA. The use of diuretics and dopamine in acute renal failure:
a systematic review of the evidence. Crit Care 1997; 1: 53-9.
44 Bellomo R, Ronco C, Kellum JA, et al. Acute renal failure –
definition, outcome measures, animal models, fluid therapy and
information technology needs: the Second International Consensus
Conference of the Acute Dialysis Quality Initiative (ADQI) Group.
Crit Care 2004; 8: R204-12.
45 Hoste EA, Clermont G, Kersten A, et al. RIFLE criteria for acute
kidney injury are associated with hospital mortality in critically ill
patients: a cohort analysis. Crit Care 2006; 10: R73.
... Therefore, furosemide is commonly used to promote diuresis in the postoperative and critically ill patients with uid overload, regardless of whether the patient has AKI or not. However, human and animal studies have shown that furosemide is not effective in preventing AKI in these settings [9,10]. Furthermore, it was recently demonstrated that diuretic use was independently associated with higher AKI risk in patients hospitalized with Covid-19, which is a hyperin ammatory disease per se [11]. ...
Preprint
Full-text available
Background: Perioperative Acute Kidney Injury (AKI) caused by Ischemia-Reperfusion (IR) is a significant contributor to mortality and morbidity after major surgery. Furosemide is commonly used in postoperative patients to promote diuresis and reduce tissue edema. However, its effects on renal microcirculation, oxygenation and function after AKI are poorly understood. Herein, we investigated the effects of furosemide in rats subjected to IR insult. Methods: 24 Wistar albino rats were divided into 4 groups, with 6 in each; Sham-operated Control (C), Control + Furosemide (C+F), ischemia/reperfusion (IR), and IR+F. After induction of anaesthesia (BL), supra-aortic occlusion was applied to IR and IR+F groups for 45 minutes followed by ongoing reperfusion for 15 minutes (T1) and 2 hours(T2). Furosemide infusion was initiated simultaneously in the intervention groups after ischemia. Renal blood flow (RBF), vascular resistance (RVR), oxygen delivery (DO2ren) and consumption (VO2ren), sodium reabsorption (TNa+), oxygen utilization efficiency (VO2/TNa+), cortical (CμO2) and medullar (MμO2) microvascular oxygen pressures, urine output (UO) and creatinine clearance (Ccr) were measured. Biomarkers of inflammation, oxidative and nitrosative stress were measured and kidneys were harvested for histological analysis. Results: IR significantly decreased RBF, mainly by increasing RVR, which was exacerbated in the IR+F group at T2 (2198 vs 4223 dyne/s/cm5, p<0.05). CμO2 (61.65±6.8 vs 86.02±6.67 mmHg) and MμO2 (51.18±4.16 vs 68.79±4.98 mmHg, p<0.05) were both reduced after IR and did not improve by Furosemide. Moreover, VO2/TNa+ deteriorated in the IR+F compared to IR (78,5±63% vs IR+F:422,5±707% compared to BL values, p=0,07) suggesting a possible harm. Ccr did not change and plasma creatinine increased significantly in IR+F. Histopathology revealed widespread damage both in the cortex and medulla in IR+F and C+F groups. Conclusion: Renal oxygen delivery, medullar and cortical oxygenation and oxygen utilization all declined by furosemide administration after IR insult. Our study suggests that furosemide may cause additional structural and functional impairment to the kidney following ischemic injury and should be used with caution.
Chapter
In recent years our understanding of molecular mechanisms of drug action and interindividual variability in drug response has grown enormously. Meanwhile, the practice of anesthesiology has expanded to the preoperative environment and numerous locations outside the OR. Anesthetic Pharmacology: Basic Principles and Clinical Practice, 2nd edition, is an outstanding therapeutic resource in anesthesia and critical care: Section 1 introduces the principles of drug action, Section 2 presents the molecular, cellular and integrated physiology of the target organ/functional system and Section 3 reviews the pharmacology and toxicology of anesthetic drugs. The new Section 4, Therapeutics of Clinical Practice, provides integrated and comparative pharmacology and the practical application of drugs in daily clinical practice. Edited by three highly acclaimed academic anesthetic pharmacologists, with contributions from an international team of experts, and illustrated in full colour, this is a sophisticated, user-friendly resource for all practitioners providing care in the perioperative period.
Article
Full-text available
Objectives The study aimed to investigate the incidence and risk factors of acute kidney injury (AKI) in elderly patients (aged ≥ 75 years) undergoing major nonvascular abdominal surgery. Methods The study was a retrospective study that evaluated the incidence of AKI in patients within 48 h after major abdominal surgeries. Patients' preoperative characteristics and intraoperative management, including the use of nephrotoxic medications, were evaluated for associations with AKI using a logistic regression model. Results A total of 573 patients were included in our analysis. A total of 33 patients (5.76%) developed AKI, and 30 (90.91%), 2 (6.06%) and 1 (3.03%) reached the AKI stages 1, 2 and 3, respectively. Older age (adjusted OR, aOR 1.112, 95% confidence interval, CI 1.020–1.212), serum albumin (aOR 0.900, 95% CI 0.829–0.977), baseline eGFR (aOR 3.401, 95% CI 1.479–7.820), the intraoperative occurrence of hypotension (aOR 3.509, 95% CI 1.553–7.929), and the use of hydroxyethyl starch in combination with nonsteroidal anti-inflammatory drugs (aOR 3.596, 95% CI 1.559–8.292) or furosemide (aOR 5.724, 95% CI 1.476–22.199) were independent risk factors for postoperative AKI. Conclusions Several risk factors, including intraoperative combined administration of HES and furosemide, are independent factors for AKI during abdominal surgeries. Anesthesiologists and surgeons should take precautions in treating at-risk patients.
Preprint
Full-text available
Background: Diuretic use in acute kidney injury (AKI) is debatable. Contrasting outcomes have been reported regarding its efficacy in AKI. Sequential nephron blockade (SNB) has never been studied in a diverse ICU population admitted with AKI. Methods: Consecutive patients admitted to medical ICU with oliguric AKI over 1-year period were recruited. SNB as a diuretic strategy was administered. Responders (R) and non-responders (NR) were compared. Demographic data, clinical profile, treatment and outcomes were compared between the groups using t-test and Chi-square test as appropriate. Predictors of mortality were explored using bivariate and multivariate logistic regression analysis and expressed as Odds ratio (OR) with 95% confidence interval (CI). Results: Of the 331 ICU patients admitted with AKI, 312 patients received SNB. The mean (SD) age was 64.14±14·79 years with male preponderance (196:116). Admission SOFA score was significantly (p<0·001) higher (9.6±3.8) in non-responders than responders (6·3±3.6). Overall, 80.13% patients were responders and 19.6% patients received RRT. Significant difference (p<0·001) for diuretic response and dialysis were seen with pre-existing diabetes, higher KDIGO stage and SOFA score (>9) at ICU admission, sepsis, shock, mechanical ventilation, positive fluid balance, lengthy ICU stay and mortality. On multivariate logistic regression analysis, SOFA score >9(OR 4.50; 95%CI 2.30-8.80), negative diuretic response (OR 0.45; 95%CI 0.34-0.59), RRT need (OR 1.78; 95%CI 1.25-2.19) and a higher positive fluid balance (OR 2.82; 95%CI 1.76-3.89) were independently associated with death. Conclusions: SNB in oliguric AKI patients resulted in lesser need for dialysis and better outcomes. A negative response, need for dialysis and a higher positive fluid balance were independently associated with mortality.
Preprint
Full-text available
Background: Diuretic use in acute kidney injury (AKI) is debatable. Contrasting outcomes have been reported regarding its efficacy in AKI. Sequential nephron blockade (SNB) has never been studied in a diverse ICU population admitted with AKI. Methods: Consecutive patients admitted to medical ICU with oliguric AKI over 1-year period were recruited. SNB as a diuretic strategy was administered. Responders (R) and non-responders (NR) were compared. Demographic data, clinical profile, treatment and outcomes were compared between the groups using t-test and Chi-square test as appropriate. Predictors of mortality were explored using bivariate and multivariate logistic regression analysis and expressed as Odds ratio (OR) with 95% confidence interval (CI). Results: Of the 331 ICU patients admitted with AKI, 312 patients received SNB. The mean (SD) age was 64.14±14·79 years with male preponderance (196:116). Admission SOFA score was significantly (p<0·001) higher (9.6±3.8) in non-responders than responders (6·3±3.6). Overall, 80.13% patients were responders and 19.6% patients received RRT. Significant difference (p<0·001) for diuretic response and dialysis were seen with pre-existing diabetes, higher AKIN stage and SOFA score (>9) at ICU admission, sepsis, shock, mechanical ventilation, positive fluid balance, lengthy ICU stay and mortality. On multivariate logistic regression analysis, SOFA score >9(OR 4.50; 95%CI 2.30-8.80), negative diuretic response (OR 0.45; 95%CI 0.34-0.59), RRT need (OR 1.78; 95%CI 1.25-2.19) and a higher positive fluid balance (OR 2.82; 95%CI 1.76-3.89) were independently associated with death. Conclusions: SNB in oliguric AKI patients resulted in lesser need for dialysis and better outcomes. A negative response, need for dialysis and a higher positive fluid balance were independently associated with mortality.
Article
Extracorporeal life support (ECLS) also known as extracorporeal membrane oxygenation (ECMO) is increasingly employed in post-cardiotomy (PC) shock to facilitate a bridge to sustained recovery, long-term mechanical support or heart transplantation. Given increasing prevalence and complexity of PC-ECLS, a joint expert consensus guideline was created in 2020 for management of adult PC-ECLS by The European Association for Cardio-Thoracic Surgery (EACTS), The Extracorporeal Life Support Organization (ELSO), The Society of Thoracic Surgeon (STS) and The American Association of Thoracic Surgery (AATS). The aim of this analysis is to comprehensively review the expert consensus guidelines with particular emphasis on PC-ECLS candidacy, timing, cannula configuration, left ventricular distension, anticoagulation, ECLS weaning and ICU complications. This analysis finds the expert consensus guideline to be timely, pertinent and clinically valuable although there remains the need for larger clinical trials to codify best practices.
Article
Background: Continuous renal replacement therapy (CRRT) is the standard treatment for severe acute kidney injury in critically ill patients. However, a practical consensus for discontinuing CRRT is lacking. We aimed to develop a prediction model with simple clinical parameters for successful discontinuation of CRRT. Methods: Adult patients who received CRRT at Samsung Medical Center from 2007 to 2017 were included. Patients with preexisting ESRD and patients who progressed to ESRD within 1 year or died within 7 days after CRRT were excluded. Successful discontinuation of CRRT was defined as no requirement for renal replacement therapy for 7 days after discontinuing CRRT. Patients were assigned to either a success group or failure group according to whether discontinuation of CRRT was successful or not. Results: A total of 1,158 patients were included in the final analyses. The success group showed greater urine output on the day before CRRT discontinuation (D-1) and the discontinuation day (D0). Multivariable analysis identified that urine output ≥300 mL on D-1, and mean arterial pressure 50∼78 mm Hg, serum potassium <4.1 mmol/L, and BUN <35 mg/dL (12.5 mmol/L) on D0 were predictive factors for successful discontinuation of CRRT. A scoring system using the 4 variables above (area under the receiver operating curve: 0.731) was developed. Conclusions: Scoring system composed of urine output ≥300 mL/day on D-1, and adequate blood pressure, serum potassium <4.1 mmol/L, and BUN <35 mg/dL (12.5 mmol/L) on D0 was developed to predict successful discontinuation of CRRT.
Chapter
Acute kidney injury (AKI) remains a frequent and serious complication of surgical procedures and critical illness that is consistently associated with worse outcomes and increased long-term morbidity and mortality. Much work has gone into finding kidney protective measures with disappointingly few therapeutic options available to prevent or to treat AKI. Research has defined some effective kidney protective practices and the purpose of this chapter is to help the clinician to differentiate better clinical practices that are either ineffective, detrimental, or protective to the kidney. The chapter will first provide an introduction to kidney physiology with particular focus on areas that make it vulnerable to injury. The chapter will then shift focus to diagnosis of kidney injury including definitions and early biomarkers that can help risk stratify patients. Then, specific mechanisms of postoperative and critical illness-associated kidney injury will be outlined. The rest of the chapter will review evidence from renal-protective research studies pertinent to the preoperative, intraoperative, and postoperative periods. After reading this chapter, the clinician should have a more robust framework for adoption (mostly avoidance) of renal protective practices (stratified based on level of evidence available) to guide their clinical practice.
Article
Full-text available
Introduction The lack of a standard definition for acute kidney injury has resulted in a large variation in the reported incidence and associated mortality. RIFLE, a newly developed international consensus classification for acute kidney injury, defines three grades of severity – risk (class R), injury (class I) and failure (class F) – but has not yet been evaluated in a clinical series.
Article
Full-text available
Objective: Funnel plots (plots of effect estimates against sample size) may be useful to detect bias in meta-analyses that were later contradicted by large trials. We examined whether a simple test of asymmetry of funnel plots predicts discordance of results when meta-analyses are compared to large trials, and we assessed the prevalence of bias in published meta-analyses. Design: Medline search to identify pairs consisting of a meta-analysis and a single large trial (concordance of results was assumed if effects were in the same direction and the meta-analytic estimate was within 30
Article
Context Acute renal failure is associated with high mortality and morbidity. Diuretic agents continue to be used in this setting despite a lack of evidence supporting their benefit. Objective To determine whether the use of diuretics is associated with adverse or favorable outcomes in critically ill patients with acute renal failure. Design Cohort study conducted from October 1989 to September 1995. Patients and Setting A total of 552 patients with acute renal failure in intensive care units at 4 academic medical centers affiliated with the University of California. Patients were categorized by the use of diuretics on the day of nephrology consultation and, in companion analyses, by diuretic use at any time during the first week following consultation. Main Outcome Measures All-cause hospital mortality, nonrecovery of renal function, and the combined outcome of death or nonrecovery. Results Diuretics were used in 326 patients (59%) at the time of nephrology consultation. Patients treated with diuretics on or before the day of consultation were older and more likely to have a history of congestive heart failure, nephrotoxic (rather than ischemic or multifactorial) origin of acute renal failure, acute respiratory failure, and lower serum urea nitrogen concentrations. With adjustment for relevant covariates and propensity scores, diuretic use Was associated with a significant increase in the risk of death or nonrecovery of renal function (odds ratio, 1.77; 95% confidence interval, 1.14-2.76). The risk was magnified (odds ratio, 3.12; 95% confidence interval, 1.73-5.62) when patients who died within the first week following consultation were excluded. The increased risk was borne largely by patients who were relatively unresponsive, to diuretics. Conclusions The use of diuretics in critically ill patients with acute renal failure was associated with an increased risk of death and nonrecovery of renal function. Although observational data prohibit causal inference, it is unlikely that diuretics afford any material benefit in this clinical setting. In the absence of compelling contradictory data from a randomized, blinded clinical trial, the widespread use of diuretics in critically ill patients with acute renal failure should be discouraged.
Chapter
This chapter contains section titled: Summary points A Systematic Review of Published Checklists and Scales The QUOROM Statement Comparison of QUOROM to Other Checklists and Scales Discussion Acknowledgements Appendix 1 Appendix 2
Article
A semi-parametric method is developed for assessing publication bias prior to performing a meta-analysis. Summary estimates for the individual studies in the meta-analysis are assumed to have known distributional form. Selective publication is modeled using a nonparametric weight function, defined on the two-sided p-value scale. The shape of the estimated weight function provides visual evidence of the presence of bias, if it exists, and observed trends may be tested using rank order statistics or likelihood ratio tests. The method is intended as an exploratory technique prior to embarking on a standard meta-analysis.
Article
The influence of intrarenal acetylcholine (ACh) or intravenous furosemide (F) on the course of norepinephrine (NE)-induced acute renal failure (ARF) was investigated in the dog. All groups received a 40-min intrarenal infusion of NE (0.75 μg/kg per min). Group I received NE alone; group II received ACh (20 μg/min) intrarenally which was started 30 min before NE infusion and continued throughout the study; and group III received F (10 mg/kg i.v.) before NE infusion followed by a constant infusion of F (10 mg/kg per h) throughout the study. In group IV F was started immediately after completion of NE infusion. All groups were studied before, during, and 3 h after NE infusion. Urine losses were replaced with a 0.9% sodium chloride solution. Before NE, the mean glomerular filtration rate (GFR) in the infused kidney of groups I-IV was 54.1 ± 6.5, 44.2 ± 1.3, 50.2 ± 5.3, and 57.6 ± 7.3 ml/min, respectively. During NE, GFR and renal blood flow (RBF) were undetectable in the infused kidneys of all four groups. Three hours after NE GFR was 1.3 ± 1.3 in group I and 5.9 ± 4.9 ml/min in group II. In contrast, GFR in groups III and IV was 28.4 ± 5.4 and 15.40 ± 2.4 ml/min, respectively (P < 0.01 compared to group I). Before the NE infusion RBF was comparably and significantly increased from control in groups II and III. However, osmolar clearance C(osM) was increased before NE most profoundly in group III (9.04 ± 1.25 ml/min, P < 0.01 compared to groups I and II). In group IV, C(osM) increased to 7.13 ± 1.37 ml/min (P < 0.01 compared to group I). We conclude that F exerts a significant protective effect in this ischemic model of ARF, which could not be mimicked by ACh despite comparable increases in RBF. The higher C(osM) with F may be an important intratubular factor in the protection.