Journal of Critical Care 29 (2014) 10–17

Contents lists available at ScienceDirect

Journal of Critical Care

journal homepage: www.jccjournal.org

A meta-analysis of continuous vs intermittent infusion of loop diuretics in

hospitalized patients
Fahad Alqahtani, MD a, Ioannis Koulouridis, MD, MS a, b, Paweena Susantitaphong, MD a, b, c,
Khagendra Dahal, MD a, Bertrand L. Jaber, MD, MS a, b,⁎
a
Division of Nephrology, Department of Medicine, Kidney and Dialysis Research Laboratory, St Elizabeth's Medical Center, Boston, MA, USA
b
Department of Medicine, Tufts University School of Medicine, Boston, MA, USA
c
Division of Nephrology, Department of Medicine, Extracorporeal Multiorgan Support Dialysis Center, King Chulalongkorn Memorial Hospital, Faculty of Medicine,
Chulalongkorn University, Bangkok, Thailand

a r t i c l e i n f o a b s t r a c t

Keywords: Background: Several studies have examined the potential benefits of continuous vs intermittent (bolus)
Meta-analysis intravenous loop diuretic administration in hospitalized patients with conflicting results. We conducted a
Loop diuretics meta-analysis to compare the efficacy of these 2 strategies in hospitalized adults and children with
Infusion
extracellular fluid volume expansion.
Continuous
Methods: We searched MEDLINE (through October 2012) and prior meta-analyses for randomized controlled
Intermittent
Urine output
trials comparing the efficacy of continuous vs intermittent infusion of loop diuretics. Random-effects model
meta-analyses were performed to examine several outcomes, including net change in urine output and
body weight.
Results: We identified 7 crossover and 11 parallel-arm randomized controlled trials (936 patients) of adults
and children. In the 15 studies of adults, continuous loop diuretic infusion resulted in a nonsignificant net
increase in daily urine output of 334 mL (95% confidence interval [CI], − 74 to 742; P = .11) relative to the
bolus infusion. In the 8 studies that used a loading dose, continuous loop diuretic infusion resulted in a
significant net increase in daily urine output of 294 mL (95% CI, 31-557; P = .03) relative to the intermittent
infusion. There was also a significant net decrease in body weight of 0.78 kg (95% CI, − 1.54 to − 0.03; P = .04)
in the continuous relative to the intermittent loop diuretic infusion. In the 3 studies of children, there was no
demonstrable effect on daily urine output or body weight.
Conclusion: Continuous infusion of loop diuretics preceded by a loading dose results in greater diuresis in
hospitalized adults with extracellular fluid volume expansion compared with intermittent dosing
regimens. Further studies are required to examine whether these benefits translate into improved
clinical outcomes.
© 2014 Elsevier Inc. All rights reserved.

1. Introduction (bolus) or a continuous infusion. Several studies comparing the

Loop diuretics are the mainstay of treatment for disorders
associated with extracellular fluid volume expansion. By inhibiting
the active cotransport of sodium and chloride ions out of the tubular
lumen in the thick ascending limb of the loop of Henle, loop diuretics
produce potent natriuresis and diuresis [1,2]. Loop diuretics are
commonly used in hospitalized critically and noncritically ill patients
for the symptomatic treatment of edematous states, including acute
decompensated heart failure [3–8], acute kidney injury [9–11],
chronic kidney disease [12,13], and following cardiac surgery [14,15].
To induce rapid diuresis in hospitalized patients, loop diuretics
are routinely administered intravenously either as an intermittent
⁎ Corresponding author. St Elizabeth's Medical Center, Boston, MA 02135, USA. Tel.: +1
617 562 7830; fax: +1 617 562 7797.
E-mail address: bertrand.jaber@steward.org (B.L. Jaber).
efficacy of these 2 intravenous administration routes, including 3
meta-analyses [16–18], have demonstrated contradictory results in
term of net gains in diuresis as well as other more meaningful
clinical end points [3-8,12-15,19–23]. To provide an update on this
controversy, we conducted a meta-analysis of all randomized
controlled trials published to date, to examine the efficacy and
potential harm of continuous vs intermittent infusion of loop
diuretics in hospitalized patients with disorders associated with
extracellular fluid volume expansion.
2. Methods
2.1. Data sources and selection
We conducted a literature search in MEDLINE (1965 to October
2012) to identify eligible studies using the Medical Subject

0883-9441/$ – see front matter © 2014 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcrc.2013.03.015
 F. Alqahtani et al. / Journal of Critical Care 29 (2014) 10–17
Fig. 1. Literature search and selection.
Headings search terms “furosemide” or “torsemide” or “bumeta-
nide” or “ethacrynic acid,” and “randomized controlled trial” or
“controlled clinical trial” or “randomized controlled trials” or
“random allocation” or “double blind method” or “single blind
method” or “clinical trial(s)” or “random” or “trial” or “randomized
control trial” or “clinical control trial” or “comparative study” or
“cross over study.” The search was limited to human studies with
no language restrictions. We also searched ClinicalTrials.gov for
completed trials using similar search terms and reviewed scientific
abstracts from the annual meetings of the American Society of
Nephrology (2003-2012). Two of the authors (FA and KD) screened
independently titles and abstracts to identify potentially relevant
articles. We included parallel-arm and cross-over randomized
controlled trials that compared the efficacy of continuous vs
intermittent intravenous infusion of loop diuretics in hospitalized
children or adults with disorders associated with extracellular fluid
volume expansion. We imposed a minimum 4-hour loop diuretic
infusion criterion in consideration of the 1-hour half-life of
furosemide, and the fact that it takes 4 half-lives for the drug to
reach a steady state [24]. For cross-over trials, we only included the
initial phase of the study before the cross-over due to concerns
regarding the absence of a washout period and its potential effect
on the outcomes of interest [25]. For duplicate publications, data
from the most inclusive report were used.
11
2.2. Data extraction and quality assessment
The data were independently extracted in duplicate from full-text
articles by 2 of the authors (FA and KD). Where indicated, the G3data
graph analyzer (version 1.5.3; GNU; General Public License; http://
www.frantz.fi/software/g3data.php) was used to extract data from
graphs. Disagreements were resolved through consensus and arbi-
tration by a third author (BLJ).
The following trial characteristics were extracted: country of origin,
year of publication, study design, population setting (heart failure,
chronic kidney disease, critical illness, and cardiac surgery), total
number of patients, baseline demographic and clinical characteristics
(sex and mean age, weight, serum creatinine, sodium, and potassium),
loop diuretic name, use of a loading dose before the continuous
infusion, prescribed dose, duration of intervention, and study follow-up.
Using a dose conversion factor, a daily furosemide equivalent dose
was calculated whenever the loop diuretic dose was documented. In
brief, a 40 mg intravenous dose of furosemide was considered to have
equivalent therapeutic efficacy to 20 mg of torsemide, 1 mg of
bumetanide, and 40 mg of ethacrynic acid [26,27].
Our main outcome of interest was the change in absolute
daily (and hourly) urine output. Other outcomes of interest
were changes in body weight, serum creatinine, urine sodium and
potassium excretion rate, development of electrolyte imbalances,
12 F. Alqahtani et al. / Journal of Critical Care 29 (2014) 10–17

tinnitus or hearing loss, hospital length of stay, and in-hospital
mortality. Study quality was assessed using the Jadad scale, based on
adequacy of randomization, blinding, and attrition rate [28].
2.3. Data synthesis and analysis
Because of differences in the reported units of urine output, we
performed random-effects model meta-analyses separately for stud-
ies of adults and children, to examine net changes in absolute daily
(and hourly) urine output, in the continuous relative to the
intermittent loop diuretic infusion group [29]. Because most of the
studies reported the outcome on a daily timeframe, we assumed
the 24-hour period to be a good approximation and, therefore,
adopted this approach to reflect everyday practice. We conducted
similar meta-analyses in studies of adults to examine net changes in
body weight; daily urine sodium and potassium excretion rate; serum
sodium, potassium, and creatinine; and hospital length of stay.
Because of the low number of events for the outcome of mortality,
for studies of adults with no events in either group, we added a fudge
factor of 0.5 to the number of events and nonevents and performed a
random-effects model meta-analysis to generate relative risk for
mortality in the continuous relative to the intermittent loop diuretic
infusion group.
Heterogeneity among individual study effect sizes was examined
using the I 2 index and the Q test P value. The I 2 index describes the
percentage of total variation across trials due to true heterogeneity
rather than chance, with a value of 75% or greater indicating
medium-to-high heterogeneity [30].
Potential causes of heterogeneity were explored by performing
subgroup analyses for the main outcome of interest based on a priori
selected study characteristics, including year of publication, study size
(≤ vs N median), study design (cross-over vs parallel-arm design),
clinical settings (ie, heart failure, cardiac surgery, intensive care unit,
chronic kidney disease), use of loading dose, study quality, and study
duration (b vs ≥24 hours). We also performed our main analysis by
removing the 2 most extreme outliers among the included studies.
Meta-regression analyses were also performed to explore the relation
between the prescribed furosemide-equivalent dose (in milligrams per
day) and the daily urine output. Finally, publication bias was formally
assessed using funnel plots and the Egger's linear regression test of
funnel plot asymmetry [31].

Table 1
Characteristics of the randomized controlled trials included in the meta-analysis

Author Year Population Clinical setting Study No. of Mean Loop Duration of Loading Prescribed furosemide Jadad
design patients age (y) diuretic intervention dosea (or equivalent) score
(h) dose (mg/d)

Continuous Intermittent
infusion infusion

Copeland 1983 Adults Cardiac surgery Parallel-arm 18 NR Furosemide 12 No 90b 1

et al [14]
b
Rudy 1991 Adults Chronic kidney disease Cross-over 8 40.8 Bumetanide 12 Yes 960 1
et al [12]
Singh 1992 Children Cardiac surgery Parallel arm 20 1.9 Furosemide 24 Yes 4.9c 6.2c 1
et al [34]
Lahav 1992 Adults Heart failure Cross-over 9 74.1 Furosemide 48 Yes 90-120b 1
et al [20] (classes III and IVd)
b
Dormans 1996 Adults Heart failure Cross-over 20 71.0 Furosemide 8 Yes 690 1
et al [5] (classes III and IVd)
Kramer 1996 Adults Heart failure Cross-over 8 53.4 Torsemide 24 Yes 200b 1
et al [19] (classes II and IIId)
Luciani 1997 Children Cardiac surgery Parallel arm 26 0.3 Furosemide 24 Yes 2.5c 6.8c 1
et al [32]
c c
Klinge 1997 Children Cardiac surgery Parallel arm 46 2.8 Furosemide 72 No 2.1 1.6 2
et al [33]
b
Aaser 1997 Adults Heart failure Cross-over 8 54.0 Furosemide 24 No 145 1
et al [3] (classes II and IIId)
Schuller 1997 Adults Medical intensive Parallel arm 33 64.0 Furosemide 24 Yes 320 320 1
et al [23] care unit
Pivac 1998 Adults Heart failure Cross-over 20 62.2 Furosemide 24 No 80b 1
et al [7] (classes III and IVd)
Mojtahedzadeh 2004 Adults Medical intensive Parallel arm 22 NR Furosemide 36 Yes 320 320 1
et al [21] care unit
Ostermann 2007 Adults Medical and Parallel arm 56 64.0 Furosemide 72 Yes 221 576 5
et al [22] surgical intensive
care unit
Sanjay 2008 Adults Chronic kidney disease Cross-over 42 53.6 Furosemide 4 Yes 360-1440b 2
et al [13]
b
Kunt 2009 Adults Cardiac surgery Parallel arm 100 65.6 Furosemide 24 No 240 4
et al [15]
Allen 2010 Adults Acute decompensated Parallel arm 41 59.5 Furosemide 24 No 162b 3
et al [4] heart failure
Thomson 2010 Adults Acute decompensated Parallel arm 56 55.5 Furosemide 86-112 No 197 172 3
et al [8] heart failure
Felker 2011 Adults Acute decompensated Parallel arm 308 66.0 Furosemide 72 No 160 198 5
et al [6] heart failure

NR indicates not reported.

a
Loading dose before administration of continuous loop diuretic infusion.
b
Distinction between the 2 arms is not reported.
c
Dose in milligrams per kilogram per day.
d
Heart failure in accordance with the New York Heart Association classification system.
 F. Alqahtani et al. / Journal of Critical Care 29 (2014) 10–17 13

The meta-analyses were performed using Comprehensive for the subgroup analyses were generated using the R system software
Meta-Analysis version 2.0 (Biostat, www.meta-analysis.com) and version 2.13.0 (cran.rproject.org/bin/windows/base/old/2.13.0).
Open_Meta (http://tuftscaes.org/open_meta/download. html). Figures All pooled estimates are displayed with a 95% confidence interval

Fig. 2. Forest plot displaying the effect of continuous vs intermittent (bolus) loop diuretic infusion on daily (A) or hourly (B) urine output in trials of adults and daily (C) urine output
in trials of children.
14 F. Alqahtani et al. / Journal of Critical Care 29 (2014) 10–17

Fig. 3. Subgroup analyses in trials of adults displaying the effect of continuous vs intermittent (bolus) loop diuretic infusion on urine output according to publication year, sample
size, study design, clinical setting, use of loading dose, and study quality.

(CI) [32]. P values were considered statistically significant at less (95% CI, − 3 to 31; P = .11; Fig. 2B) compared with the intermittent
than .05. (bolus) infusion. The test for heterogeneity was significant (Q test, P b
.001; I 2 = 84%) [5,12–14]. In the 3 studies of children undergoing
3. Results cardiac surgery that reported on changes in urine output (92
patients), by meta-analysis, the continuous loop diuretic infusion
3.1. Study characteristics and quality resulted in a nonsignificant net decrease in urine output of 6.3 mL/kg
per day (95% CI, − 16.4 to 3.7; P = .22; Fig. 2C) compared with the
A total of 3824 potentially relevant citations were identified and intermittent infusion.
screened (Fig. 1). We retrieved 84 full-text articles for evaluation, of On subgroup analyses (Fig. 3), in studies that used a loop diuretic
which 17 satisfied the selection criteria. In addition, 6 relevant loading dose, the continuous infusion strategy resulted in a
systematic reviews were identified (from a total of 22) and evaluated significant net increase in urine output of 294 mL/d (95% CI, 31-
in full text. Fifty-four potentially relevant citations were identified 556; P = .03) compared with the intermittent infusion strategy.
from the references of these systematic reviews, of which 9 satisfied There was no demonstrable benefit of the continuous relative to the
the selection criteria. After removal of duplicate reports, 18 trials of intermittent infusion strategy in studies that did not use a loading
adults and children were included [3-8,12-15,19-23,32–34]. All dose (net increase in urine output of 340 mL/d [95% CI, − 416 to
eligible studies were in the English language. 1096; P = .38; Fig. 3]). Of note, after excluding 4 studies with a
Table 1 summarizes the characteristics of the studies. Published duration of less than 24 hours [5,12–14], the continuous infusion
over 28 years, there were 7 cross-over and 11 parallel-arm
randomized controlled trials. The 18 trials enrolled a total of 936
patients. Sample sizes ranged from 8 to 308 patients. In trials of adults,
mean age ranged from 41 to 74 years, and body weight, from 64.0 to
85.4 kg. In trials of children, mean age ranged from 3.6 to 33.6 months,
and body weight, from 4.3 to 14 kg. Eight trials were restricted to
patients with heart failure [3-8,19,20]; 5 trials, to patients undergoing
cardiac surgery [14,15,32–34]; 3 trials, to critically ill patients in the
intensive care unit [21–23]; and 2 trials, to patients with chronic
kidney disease [12,13].
The loop diuretic dosing parameters for each trial are summarized
in Table 1. In brief, furosemide was used in 16 studies, and 10 of the 18
studies used a loading dose [5,12,13,19–23,32,34]. The randomization
procedure was described in 5 trials [4,6,8,15,22], with the blinding
procedure being adequately documented in only 1 study [6]. As
shown in Table 1, the Jadad score ranged from 1 to 5 (median of 1),
indicating overall low quality of the studies.

3.2. Effect of continuous vs intermittent loop

diuretic infusion on urine output

In the 15 studies of adults, 14 reported on daily urine output (811

patients). As shown in Fig. 2, by meta-analysis, the continuous loop
diuretic infusion resulted in a nonsignificant net increase in the urine Fig. 4. Funnel plot of mean difference in urine output against the standard error of the
output of 334 mL/d (95% CI, −74 to 742; P = .11; Fig. 2A) or 14 mL/h treatment estimate in trials of adults (Egger's test, P = .6).
 F. Alqahtani et al. / Journal of Critical Care 29 (2014) 10–17 15

Table 2
Effect of continuous vs intermittent (bolus) loop diuretic infusion on other outcomes in trials of adults

Continuous outcome No. of No. of Difference 95% CI P Heterogeneity

studies patients in means
Q test P value I2 index (%)

Urinary sodium excretion, mmol/d 8 228 21.18 −10.50 52.86 .190 .420 1.27
Urinary potassium excretion, mmol/d 5 178 5.38 −4.00 14.77 .261 .767 0.00
Change in serum creatinine, mg/dL 7 600 −0.07 −0.24 0.11 .454 b.001 90.18
Change in serum potassium, mmo/L 4 121 0.39 −0.12 0.90 .138 .132 46.52
Change in serum sodium, mmol/L 4 121 0.47 −0.29 1.22 .226 .009 74.20
Hospital length of stay, d 5 464 −0.72 −2.25 0.81 .359 .025 64.16
Change in body weight, kg 3 405 −0.78 −1.54 −0.03 .041 .712 0.00

Binary outcome No. of No. of Risk 95% CI P Heterogeneity

studies patients ratio
Q test P value I2 index (%)

All-cause mortality 6 560 0.88 0.45 1.72 .783 .289 19.12

strategy resulted in a nonsignificant net increase in urine output of
384 mL/d (95% CI, − 111 to 879; P = .13) compared with the
intermittent infusion strategy. After excluding the 2 studies
considered to be outliers due to their most prominent effect size
[14,15], there was no substantial heterogeneity (Q test, P = .4; I 2 =
9%), and there was a nonsignificant trend toward a net increase in
urine output of 150 mL/d (95% CI, − 20 to 319; P = .08) in the
continuous relative to intermittent infusion strategy. Other subgroup
analyses stratified according to year of publication, study design,
sample size, clinical setting, and study quality did not significantly
influence net changes in daily urine output. Finally, by meta-
regression, there was no relation between the furosemide-equiva-
lent daily dose and daily urine output (P = .45).
As shown in Fig. 4, the funnel plot for urine output in studies of
adults was symmetric, and the Egger's linear regression test of funnel
plot asymmetry was not significant (β coefficient for bias, .81; P = .6),
suggesting less susceptibility to publication bias. Because of the
small number of studies of children, assessment of publication bias
was not entertained.
3.3. Effect of continuous vs intermittent loop
diuretic infusion on other outcomes
The results of the meta-analyses examining the effect of the
continuous vs intermittent loop diuretic infusion strategy on other
outcomes are summarized in Table 2. In the 3 studies that reported on
changes in body weight (405 patients), the continuous loop diuretic
infusion resulted in a significant net decrease in body weight of 0.78 kg
(95% CI, − 1.54 to − 0.03; P = .04) compared with the intermittent
infusion. However, there was no significant effect on net change in
urinary sodium and potassium excretion rates and serum creatinine,
sodium, and potassium level. There was a nonsignificant reduction
in hospital length of stay of 0.72 day (95% CI, −2.25 to 0.81; P = .40) and
a nonsignificant 12% in-hospital mortality risk reduction (0.88; 95% CI,
0.45-1.72; P = .78).
4. Discussion
The present meta-analysis of randomized controlled trials sug-
gests that the use of a continuous loop diuretic infusion preceded by a
loading dose results in greater diuresis in hospitalized adults with
disorders associated with extracellular fluid volume expansion
compared with an intermittent (bolus) dosing regimen. Body weight
loss was also significantly enhanced by the adoption of this strategy.
No differences were observed in other surrogate outcomes, including
changes in serum creatinine, urinary sodium, and potassium excre-
tion; hospital length of stay; and in-hospital mortality between these
2 intravenous loop diuretic administration routes.
Following an intermittent bolus administration of a loop diuretic,
diuresis reaches a peak within 2 hours, which is followed by a gradual
decline in the urine output. By contrast, the continuous infusion of a
loop diuretic results in a gradual increase in the plasma drug level,
which peaks several hours after the initiation of the infusion,
remaining constant throughout the infusion period [14]. Several
studies have used loading doses of loop diuretics to achieve
immediate peak effective levels, which may be critical in patients
with severe volume overload in whom rapid and sustained diuresis is
warranted [5,12,13,19-23,34].
Three meta-analyses comparing the efficacy of continuous vs
intermittent infusion of loop diuretics have previously been published
(Table 3) [16–18]. Two meta-analyses were restricted to hospitalized
adults with heart failure [16,17] and 1 meta-analysis to children
undergoing cardiac surgery [18]. Both meta-analyses of adults with
heart failure demonstrated a benefit of using a continuous loop

Table 3
Comparison of the results of the present meta-analysis with previously published reports

Author Year Population Intervention Comparator Setting No. of Outcome Conclusion

trials

Salvador et al [16] 2005 Adult Continuous Bolus Heart failure 8 Urine output (mL/d) Significant increase
in urine output
Amer et al [17] 2011 Adult Continuous Bolus Heart failure 10 Urine output Significant increase
(mL/d/100 mg furosemide) in urine output
Zangrillo et al [18] 2012 Children Continuous Bolus Cardiac surgery 3 Urine output (mL/kg/d) No significant increase
in urine output
Alqahtani et al 2013 Adults and children Continuous Bolus Heart failure; cardiac 18 Urine output (mL/d) No significant increase
surgery; chronic in urine output (except
kidney disease; for when loading dose
intensive care unit administered
16 F. Alqahtani et al. / Journal of Critical Care 29 (2014) 10–17
diuretic infusion in term of greater urine output, compared with the
intermittent infusion strategy [16,17]. However, the first meta-
analysis of 8 trials published in 2005 [16] included 2 trials that used
continuous loop diuretic infusions of shorter duration (30-60 mi-
nutes) [35,36], as compared with our analyses that mandated an a
priori minimum 4-hour continuous infusion in consideration for the 4
half-lives that a drug takes to reach a steady state [24]. These authors
also included 1 trial of patients with heart failure when in fact the
study had enrolled critically ill patients in pulmonary edema not
necessarily due to heart failure [23]. In the second meta-analysis of 10
trials published in 2012 [17], the authors excluded loop diuretics
other than furosemide, divided 1 crossover trial into 2 independent
studies [7], and included a duplicate study [18]. In addition, these
authors included 2 trials of critically ill patients who did not have a
diagnosis of heart failure [21–23]. The discordance observed between
these 2 meta-analyses and our subgroup analysis restricted to trials
of adults with heart failure might be explained by these aforemen-
tioned differences.
In the third meta-analysis of children undergoing cardiac surgery,
there was no significant effect of the continuous loop diuretic infusion
strategy on urine output compared with the intermittent infusion
strategy [18], which is consistent with our findings. However, we did
observe a paradoxical nonsignificant increase in urine output in the
intermittent compared with the continuous loop diuretic infusion
strategy, which was driven by the use of higher doses of loop diuretics
in the intermittent (bolus) group in 2 of the 3 trials included in our
analysis [32,34].
Adverse effects such as ototoxicity and hypotension were not
observed in patients receiving continuous loop diuretic infusions. In
fact, hearing loss, tinnitus, or both were more commonly reported in
patients receiving bolus injections, although these side effects were
transient and did not necessitate discontinuation of the drug.
To our knowledge, this is the largest meta-analysis on the topic,
which included 18 trials of children and adults. We used subgroup
analyses to explore heterogeneity. Moreover, we combined studies
that used loop diuretics other than furosemide, including torsemide
[19] and bumetanide [12], which allowed us to meta-analyze the
prescribed furosemide-equivalent dose. There are, however, several
important limitations to consider. There was significant heterogeneity
among individual study effect sizes inserting potential bias into the
pooled estimates. In addition, the inclusion of crossover trials that
lacked optimal washout periods [3,7] due to concerns
with hemodynamic stability in patients with disorders of extracellular
fluid volume expansion such as heart failure prompted us to remove
from our analyses the data generated after the crossover in an attempt
to account for inadequate washout periods. Such action might have
inserted observation bias. Nevertheless, there is no indication that this
bias exerted a differential effect on our pooled estimates.
In conclusion, our updated meta-analysis of 18 trials demonstrates
that the continuous infusion of loop diuretics preceded by a loading
dose results in greater diuresis in hospitalized adults with extracel-
lular fluid volume expansion compared with intermittent dosing
regimens. However, there were no observed benefits on other more
meaningful clinical end points. Further studies are required to
examine whether this strategy does confer a true clinical benefit.
5. Role of the funding source
This work was supported by the authors' own funds. The content is
solely the responsibility of the authors.
Acknowledgments
This work was made possible in part through Dr Susantitaphong's
International Society of Nephrology–funded fellowship.
References
[1] Brater DC. Clinical pharmacology of loop diuretics. Drugs 1991;41(Suppl 3):14–22.
[2] Wittner M, Di Stefano A, Wangemann P, et al. How do loop diuretics act? Drugs
1991;41(Suppl 3):1–13.
[3] Aaser E, Gullestad L, Tollofsrud S, et al. Effect of bolus injection versus continuous
infusion of furosemide on diuresis and neurohormonal activation in patients with
severe congestive heart failure. Scand J Clin Lab Invest 1997;57:361–7.
[4] Allen LA, Turer AT, Dewald T, et al. Continuous versus bolus dosing of Furosemide
for patients hospitalized for heart failure. Am J Cardiol 2010;105:1794–7.
[5] Dormans TP, van Meyel JJ, Gerlag PG, et al. Diuretic efficacy of high dose
furosemide in severe heart failure: bolus injection versus continuous infusion.
J Am Coll Cardiol 1996;28:376–82.
[6] Felker GM, Lee KL, Bull DA, et al. Diuretic strategies in patients with acute
decompensated heart failure. N Engl J Med 2011;364:797–805.
[7] Pivac N, Rumboldt Z, Sardelic S, et al. Diuretic effects of furosemide infusion versus
bolus injection in congestive heart failure. Int J Clin Pharmacol Res 1998;18:121–8.
[8] Thomson MR, Nappi JM, Dunn SP, et al. Continuous versus intermittent infusion of
furosemide in acute decompensated heart failure. J Card Fail 2010;16:188–93.
[9] Brown CB, Ogg CS, Cameron JS. High dose frusemide in acute renal failure: a
controlled trial. Clin Nephrol 1981;15:90–6.
[10] Cantarovich F, Rangoonwala B, Lorenz H, et al. High-dose furosemide for
established ARF: a prospective, randomized, double-blind, placebo-controlled,
multicenter trial. Am J Kidney Dis 2004;44:402–9.
[11] Shilliday IR, Quinn KJ, Allison ME. Loop diuretics in the management of acute renal
failure: a prospective, double-blind, placebo-controlled, randomized study. Nephrol
Dial Transplant 1997;12:2592–6.
[12] Rudy DW, Voelker JR, Greene PK, et al. Loop diuretics for chronic renal
insufficiency: a continuous infusion is more efficacious than bolus therapy. Ann
Intern Med 1991;115:360–6.
[13] Sanjay S, Annigeri RA, Seshadri R, et al. The comparison of the diuretic and
natriuretic efficacy of continuous and bolus intravenous furosemide in patients
with chronic kidney disease. Nephrology (Carlton) 2008;13:247–50.
[14] Copeland JG, Campbell DW, Plachetka JR, et al. Diuresis with continuous infusion
of furosemide after cardiac surgery. Am J Surg 1983;146:796–9.
[15] Kunt AT, Akgun S, Atalan N, et al. Furosemide infusion prevents the requirement of
renal replacement therapy after cardiac surgery. Anadolu Kardiyol Derg 2009;9:
499–504.
[16] Salvador DR, Rey NR, Ramos GC, et al. Continuous infusion versus bolus injection
of loop diuretics in congestive heart failure. Cochrane Database Syst Rev
2005;20(3):CD003178.
[17] Amer M, Adomaityte J, Qayyum R. Continuous infusion versus intermittent bolus
furosemide in ADHF: an updated meta-analysis of randomized control trials.
J Hosp Med 2011;7:270–5.
[18] Zangrillo A, Cabrini L, Biondi-Zoccai GGL, et al. Continuous infusion versus bolus
injection of furosemide in pediatric patients after cardiac surgery: a meta-analysis
of randomized studies. Signa Vitae 2012;7:17–22.
[19] Kramer WG, Smith WB, Ferguson J, et al. Pharmacodynamics of torsemide
administered as an intravenous injection and as a continuous infusion to patients
with congestive heart failure. J Clin Pharmacol 1996;36:265–70.
[20] Lahav M, Regev A, Ra'anani P, et al. Intermittent administration of furosemide vs
continuous infusion preceded by a loading dose for congestive heart failure. Chest
1992;102:725–31.
[21] Mojtahedzadeh M, Salehifar E, Vazin A, et al. Comparison of hemodynamic and
biochemical effects of furosemide by continuous infusion and intermittent bolus
in critically ill patients. J Infus Nurs 2004;27:255–61.
[22] Ostermann M, Alvarez G, Sharpe MD, et al. Frusemide administration in critically
ill patients by continuous compared to bolus therapy. Nephron Clin Pract
2007;107:c70–6.
[23] Schuller D, Lynch JP, Fine D. Protocol-guided diuretic management: comparison of
furosemide by continuous infusion and intermittent bolus. Crit Care Med 1997;25:
1969–75.
[24] Brunton L, Chabner B, Knollman B. Goodman and Gilman's The Pharmacological Basis
of Therapeutics. 12th ed. New York: McGraw-Hill Professional; 2010 .
[25] Elbourne DR, Altman DG, Higgins JP, et al. Meta-analyses involving cross-over
trials: methodological issues. Int J Epidemiol 2002;31:140–9.
[26] Brater DC, Day B, Burdette A, et al. Bumetanide and furosemide in heart failure.
Kidney Int 1984;26:183–9.
[27] Brater DC, Voelker JR. Use of diuretics in patients with renal disease. In:
Contemporary Issues in Nephrology. In: Bennett W, McCarron D, editors.
Pharmacotherapy of Renal Disease and Hypertension. New York: Churchill
Livingstone; 1987. p. 115–27.
[28] Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized
clinical trials: is blinding necessary? Control Clin Trials 1996;17:1–12.
[29] DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:
177–88.
[30] Huedo-Medina TB, Sanchez-Meca J, Marin-Martinez F, et al. Assessing heterogeneity
in meta-analysis: Q statistic or I2 index? Psychol Methods 2006;11:193–206.
[31] Sterne JAC, Egger M. Funnel plots for detecting bias in meta-analysis: guidelines
on choice of axis. J Clin Epidemiol 2001;54:1046–55.
[32] Luciani GB, Nichani S, Chang AC, et al. Continuous versus intermittent furosemide
infusion in critically ill infants after open heart operations. Ann Thorac Surg
1997;64:1133–9.
[33] Klinge JM, Scharf J, Hofbeck M, et al. Intermittent administration of furosemide
versus continuous infusion in the postoperative management of children
following open heart surgery. Intensive Care Med 1997;23:693–7.
 F. Alqahtani et al. / Journal of Critical Care 29 (2014) 10–17 17

[34] Singh NC, Kissoon N, Mofada SA, et al. Comparison of continuous versus
intermittent furosemide administration in postoperative pediatric cardiac
patients. Crit Care Med 1992;20:17–21.
[35] Licata G, Di Pasquale P, Parrinello G, et al. Effects of high-dose furosemide and small-
volume hypertonic saline solution infusion in comparison with a high dose of
furosemide as bolus in refractory congestive heart failure: long-term effects. Am
Heart J 2003;145:459–66.
[36] Bagatin J, Sardelic S, Gancevic I. Diuretic efficiency of furosemide in continuous
intravenous infusion vs. bolus injection in congestive heart failure: results of a
pilot study. Pharmaca 1993;31:279–86.