Randomized clinical trial

Randomized clinical trial of goal-directed fluid therapy within

an enhanced recovery protocol for elective colectomy
S. Srinivasa1 , M. H. G. Taylor2 , P. P. Singh1 , T.-C. Yu1 , M. Soop3 and A. G. Hill1
1
Department of Surgery, South Auckland Clinical School, Middlemore Hospital, University of Auckland, 2 Department of Anaesthesia, Middlemore
Hospital, and 3 Department of Surgery, University of Auckland, Auckland, New Zealand
Correspondence to: Dr S. Srinivasa, PO Box 93311, Otahuhu, Auckland, New Zealand (e-mail: [email protected])

Background: Goal-directed fluid therapy (GDFT) has been compared with liberal fluid administration
in non-optimized perioperative settings. It is not known whether GDFT is of value within an enhanced
recovery protocol incorporating fluid restriction. This study evaluated GDFT under these circumstances
in patients undergoing elective colectomy.
Methods: Patients undergoing elective laparoscopic or open colectomy within an established enhanced
recovery protocol (including fluid restriction) were randomized to GDFT or no GDFT. Bowel
preparation was permitted for left colonic operations at the surgeon’s discretion. Exclusion criteria
included rectal tumours and stoma formation. The primary outcome was a patient-reported surgical
recovery score (SRS). Secondary endpoints included clinical outcomes and physiological measures of
recovery.
Results: Eighty-five patients were randomized, and there were 37 patients in each group for analysis.
Nine patients in the GDFT and four in the fluid restriction group received oral bowel preparation for
either anterior resection (12) or subtotal colectomy (1). Patients in the GDFT group received more
colloid during surgery (mean 591 versus 297 ml; P = 0·012) and had superior cardiac indices (mean
corrected flow time 374 versus 355 ms; P = 0·018). However, no differences were observed between the
GDFT and fluid restriction groups with regard to surgical recovery (mean SRS after 7 days 47 versus 46
respectively; P = 0·853), other secondary outcomes (mean aldosterone/renin ratio 9 versus 8; P = 0·898),
total postoperative fluid (median 3750 versus 2400 ml; P = 0·604), length of hospital stay (median 6
versus 5 days; P = 0·570) or number of patients with complications (26 versus 27; P = 1·000).
Conclusion: GDFT did not provide clinical benefit in patients undergoing elective colec-
tomy within a protocol incorporating fluid restriction. Registration number: NCT00911391
(http://www.clinicaltrials.gov).

Paper accepted 16 August 2012

Published online 6 November 2012 in Wiley Online Library (www.bjs.co.uk). DOI: 10.1002/bjs.8940

Introduction Excessive fluid ‘restriction’ can risk hypovolaemia

and tissue hypoperfusion8 , so individualized management
Modern perioperative care in colorectal surgery – termed
guided by cardiovascular measures of responsiveness,
enhanced recovery after surgery (ERAS) – has encouraged
termed goal-directed fluid therapy (GDFT), has been
renewed scrutiny of intravenous fluid administration.
proposed9 . Trials and a meta-analysis of GDFT have
Several randomized trials have demonstrated that fluid
excess contributes to surgical morbidity1 – 4 . As a result, shown clinical benefits over traditional liberal fluid
fluid restriction has been incorporated into ERAS protocols administration10 – 14 . These studies were not conducted
for patients undergoing elective colectomy5,6 . Although within an ERAS environment, and the control groups
the term is defined heterogeneously7 , fluid restriction is received the liberal perioperative fluid therapy of the time.
essentially the avoidance of fluid overload, the latter being Thus the role of GDFT, within an otherwise optimized
a historical norm following extrapolation of resuscitation environment in which patients are usually subjected to fluid
practices from trauma care at the time of the Korean war. restriction, is undefined.

 2012 British Journal of Surgery Society Ltd British Journal of Surgery 2013; 100: 66–74
Published by John Wiley & Sons Ltd
Goal-directed fluid therapy within an enhanced recovery protocol for elective colectomy 67

Methods Ensure satisfactory probe position

This was a prospective double-blind single-centre ran-

Monitor FTc and SV
domized clinical trial with a 1 : 1 patient allocation ratio.
The study received both regional and institutional eth-
FTc < 350 ms No Monitor FTc and SV
ical approval before commencement and was registered
prospectively at http://www.clinicaltrials.gov (trial identi-
fier NCT00911391). Yes

Colloid challenge
FTc < 350 ms or SV
Participants 7 ml/kg first bolus (if low FTc)
decreased > 10%
3 ml/kg subsequent bolus
(or initial low SV)
Consecutive consenting patients undergoing elective open
or laparoscopic colectomy for any indication were enrolled,
Yes FTc < 350 ms
and randomized to either GDFT or fluid restriction.
Exclusion criteria were: severe oesophageal disease, recent
No
oesophageal or upper airway surgery, moderate or severe
aortic valve disease on echocardiography, bleeding diathe- FTc > 400 ms Yes Monitor FTc and SV
sis, regular use of corticosteroids or mineralocorticoids,
cognitive impairment, American Society of Anesthesiolo-
No
gists grade IV or V, rectal tumour (less than 15 cm from
the anal verge), stoma formation and patient choice.
SV increased > 10% since previous
No
bolus or measurement

Intervention Fig. 1Fluid administration protocol. Colloid comprised

succinylated gelatine solution. FTc, corrected flow time; SV,
The intervention was implemented in the intraoperative stroke volume
phase. Any time before this was defined as the preoperative
period and any subsequent period (including in the cycles15 . The algorithm for fluid administration based on
postanaesthetic care unit) was defined as the postoperative ODM measurements is shown in Fig. 1.
phase. In both groups an extra 500 ml crystalloid was allowed
Patients randomized to fluid restriction were allowed every hour if the operation extended beyond 3 h. A
to receive up to 1500 ml crystalloid solution (Plasma- consultant anaesthetist (1 of 10) was present for every
TM
Lyte 148; Baxter Healthcare, Sydney, New South Wales, operation. Vasopressors were permitted at discretion of
Australia) during surgery. They were also permitted to the anaesthetist in both groups.
receive a total of 500 ml succinylated gelatine colloid
solution (Gelofusine ; Braun, Sydney, New South Wales,
Australia) titrated by heart rate, blood pressure, urine Preoperative fluid administration
output and invasive measures (arterial lines) when used. Oral bowel preparation was used at the discretion of the
Blood loss could be corrected for in a 1 : 1 ratio using operating surgeon in patients having left-sided colonic
colloid, and hospital transfusion guidelines (haemoglobin operations (Fleet Phospho-soda ; Pharmaco NZ), but
level less than 10 g/dl in patients with cardiac co- was otherwise avoided. Patients who received bowel
morbidities, and below 7 g/dl in those without cardiac preparation received 1 litre crystalloid before surgery.
disease) were used to determine whether blood products
were necessary in either group.
Patients randomized to GDFT were also treated with
Postoperative fluid administration
baseline fluid restriction and a limit of 1500 ml crystalloid All intravenous fluids were stopped by default when
solution. A weight-based bolus of colloid was permitted patients arrived on the ward (typically 1–3 h after surgery),
based on cardiac function measured by means of an and oral intake of food, fluids and supplements was
TM
oesophageal Doppler monitor (ODM) (CardioQ , DP12 encouraged. The patient was formally assessed by a
probe; Pharmaco NZ, Auckland, New Zealand). The blinded ward doctor to decide whether any intravenous
machine was calibrated to provide data averaged over ten fluid was necessary based on the criteria described below.

 2012 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2013; 100: 66–74
Published by John Wiley & Sons Ltd
68 S. Srinivasa, M. H. G. Taylor, P. P. Singh, T.-C. Yu, M. Soop and A. G. Hill

Clinicians were required to see the patient and document general anaesthesia and mid or low thoracic epidural
their findings, and were not allowed to make decisions analgesia (combination of local anaesthetic and opioid)
over the telephone. Examination findings consistent with unless contraindicated. Successful epidural analgesia was
volume deficit were required for intravenous fluid to be defined as that not requiring an intravenous morphine
prescribed. Intravenous fluid was administered if patients pump for the first 48 h after surgery. Eight milligrams
were oliguric (less than 0·5 ml per kg per h averaged of intravenous dexamethasone (DBL Dexamethasone
over 4 h) or had deranged physiological parameters Sodium Phosphate Injection; Hospira NZ, Wellington,
suggestive of volume deficit: tachycardia exceeding 90 New Zealand) was administered at induction17 .
beats per min or low blood pressure (systolic blood Mobilization from 4 h after surgery was encouraged and
pressure below 90 mmHg in the presence of a functioning urinary catheters were removed on the day after surgery.
epidural; less than 100 mmHg without an epidural). Epidural analgesia was continued for 48 h. Simple oral
Intravenous fluid was also administered for resuscitative analgesia was provided regularly; oral or parenteral opioid
purposes in the event of complications, to compensate analgesia was avoided unless required for breakthrough
for losses or for prolonged poor oral intake such as in pain. Non-steroidal analgesia was used from day 2 (20 mg
the event of paralytic ileus. The only colloid used after oral tenoxicam). The ERAS protocol is summarized in
surgery was succinylated gelatine solution. Crystalloid Table 16 .
as described previously was used after operation in
the high-dependency unit, and dextrose–saline solution
TM
(Dex Saline , 0·18 per cent sodium chloride, 4 per cent Outcomes
glucose, 20 mmmol/l potassium; Baxter Healthcare) was
Outside the intraoperative phase all data were collected
used as maintenance crystalloid on the ward, when required
prospectively by a single blinded investigator.
to replenish extracellular fluid and total body water
respectively and to limit sodium overload.
Primary outcome
The primary outcome of the study was the surgical recovery
Perioperative care score (SRS) on day 7 after surgery. This score assesses
Patients were cared for within an established ERAS fatigue, vigour, mental function, impact on patient activity
programme6,16 . Patients received 400 ml oral carbohydrate and activities of daily living. The SRS (range 17–100) is
loading on the morning of surgery up to 2 h before derived from the validated Identity-Consequence Fatigue
TM
operation (preOp ; Nutricia, Auckland, New Zealand). Scale, which measures surgical recovery by incorporating
TM
Nutritional supplementation (Fortisip ; Nutricia) was measures of postoperative fatigue and quality of life, and
also provided before surgery if deemed necessary. Oral has been correlated with postoperative inflammation, as
bowel preparation was used selectively as described evidenced by cytokine release18,19 . The SRS was also
above. Prophylactic nasogastric tubes and abdominal registered before surgery, and on days 1, 3, 14 and 30
drains were not used. All patients received volatile after the procedure.

Table 1 Enhanced recovery after surgery protocol

Preoperative Intraoperative Postoperative

Education Epidural anaesthesia Prophylactic use of antiemetics at

Nutritional assessment Short-acting anaesthetic agents induction
Social situation explored Prevention of hypothermia, warming blanket Avoidance of drains/nasogastric tubes
Milestone and goal setting Laparoscopic/transverse incisions when Conservative fluid regimen
Ward visit/meet nursing staff feasible High inspired oxygen
Careful medical assessment Prophylactic antiemetics
Functional assessment, activities of daily living Early oral feeding with
Avoidance of fasting supplementation
Carbohydrate loading Opioid-sparing analgesia/NSAIDs
Avoidance of bowel preparation except in patients ‘Balanced analgesia’ ladder
with left-sided colonic pathology (see text) Early removal of urinary catheter
Structured nursing pathway Timed removal of epidural
Early mobilization and physiotherapy
Predefined discharge criteria

Details of fluid management are explained in text. NSAID, non-steroidal anti-inflammatory drug.

 2012 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2013; 100: 66–74
Published by John Wiley & Sons Ltd
Goal-directed fluid therapy within an enhanced recovery protocol for elective colectomy 69

Secondary outcomes Randomization

Intraoperative cardiac indices obtained from the ODM
Patients were recruited at the preadmission clinic and gave
were registered at 15-min intervals in both groups.
consent to participate in a private room, after being given
Fluid volumes administered, and intraoperative and early
verbal explanations and information sheets. Consent was
postoperative (first 24 h after surgery) urine output
reaffirmed verbally on the day of surgery and participants
were measured. Vasopressor use was noted. Bodyweight
were free to withdraw consent at any time. Randomization
was recorded before surgery and daily for the first was conducted using random numbers obtained from an
3 days after operation. Serum concentrations of brain open-source computer-based random number generator
natriuretic peptide, renin and aldosterone were measured (http://www.random.org). The randomization sequence
before operation and on the day after surgery. Serum was generated by a third party not involved in the
concentrations of sodium and creatinine were measured conduct of the study. Allocation details were concealed
before surgery and daily for the first 3 days after the in opaque envelopes that were opened on the day of
procedure. surgery, when patients were randomized. The allocation
Maximum voluntary grip strength using a hand- was performed by a research assistant after insertion of the
held dynamometer (hydraulic hand dynamometer; Jamar, ODM probe before the start of surgery and before colloid
Chicago, Illinois, USA) and peak expiratory flow (Mini- administration.
Wright peak flowmeter; Clement Clark, Harlow, UK)
were measured in the sitting position before surgery and
Blinding
daily for 3 days after the procedure. These outcomes were
not part of the SRS. The patient, study investigators, surgeon and other medical
Complications within 30 days of surgery were cate- staff responsible for patient care were blinded to patient
gorized according to the Clavien–Dindo classification20 . allocation. All patients had the ODM probe inserted by a
Length of hospital stay was recorded. trained research assistant. The ODM monitor was covered

Assessed for eligibility n = 98

Excluded n = 13
Declined to participate n = 5
Enrolment

Investigators not available n = 5

Non-English speaking n = 1
Acute operation n = 1
Anaesthetist unwilling to participate n = 1

Randomized n = 85

Allocated to GDFT n = 42 Allocated to fluid restriction n = 43

Received allocated intervention n = 37 Received allocated intervention n = 37
Did not receive allocated intervention Did not receive allocated intervention
Allocation

(stoma created) n = 5 (stoma created) n = 6

Stapler misfire n = 1 Unresectable lesion n = 1
Rectal lesion found at operation n = 2 Rectal lesion found at operation n = 4
Poor vascularity of bowel on Poor vascularity of bowel on
clinical assessment n = 2 clinical assessment n = 1
Follow-Up

Lost to follow-up n = 0 Lost to follow-up n = 0

Discontinued intervention n = 0 Discontinued intervention n = 0
Analysis

Analysed n = 37 Analysed n = 37
Excluded from analysis n = 0 Excluded from analysis n = 0

Fig. 2 CONSORT diagram for the trial. GDFT, goal-directed fluid therapy

 2012 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2013; 100: 66–74
Published by John Wiley & Sons Ltd
70 S. Srinivasa, M. H. G. Taylor, P. P. Singh, T.-C. Yu, M. Soop and A. G. Hill

by a custom-designed plastic casing (Acryform, Auckland, for one patient in the GDFT group. Three patients (2
New Zealand) to ensure blinding. fluid restriction, 1 GDFT) had intraoperative protocol
An unblinded research assistant and the consultant violations; the two patients allocated to fluid restriction
anaesthetist were aware of patient allocation. All fluid received an extra 500 ml crystalloid each and the patient in
administration was protocol-driven and delivered by the the GDFT group received a 3-ml/kg bolus of colloid when
unblinded research assistant under the supervision of not indicated by the protocol. All patients were analysed
the anaesthetist. Suggested action based on the ODM on an intention-to-treat basis.
protocol was reconfirmed every 15 min between the The two groups were well matched at baseline (Table 2).
research assistant and anaesthetist for patients in the Nine patients in the GDFT group and four in the fluid
GDFT group. The research assistant was not involved restriction arm received oral bowel preparation. All patients
in any postoperative data collection or perioperative care receiving bowel preparation were undergoing anterior
of the patients. The plastic cover over the monitor resection, except for one patient in the GDFT group
prevented the anaesthetist from looking at the readings undergoing subtotal colectomy.
obtained by the ODM for patients in the fluid restriction
group. Intraoperative adherence to the fluid protocol was
Intraoperative parameters
monitored.
A drape was placed to prevent the surgeons from Data regarding intraoperative management are shown
observing fluid administration, and the research assistant in Table 3. Intraoperative haemodynamic parameters are
was instructed to attach intravenous fluids periodically for recorded in Table 4.
patients in the fluid restriction group without actually Table 2 Baseline characteristics
administering them, to mimic the anticipated practice of
fluid boluses in the GDFT group. Fluid
GDFT restriction
(n = 37) (n = 37) P§
Statistical analysis Age (years)* 69(16) 72(12) 0·261¶
Sex ratio (F : M) 18 : 19 15 : 22 0·479
Based on previous data17 , a priori power calculations ASA grade 0·524
indicated that 37 patients would be required in each group I 5 5
to detect a 20 per cent difference in the SRS with an α of II 20 15
0·05 and β of 0·2. Thus a sample size of 80 was planned. III 12 17
Body mass index (kg/m2 )* 26·9(4·5) 26·4(4·4) 0·622¶
Owing to a higher than anticipated number of exclusions, Preop. haemoglobin (g/dl)* 11·8(1·9) 12·5(2·1) 0·169¶
the sample size was increased to 85 patients after gaining Bowel preparation 9 4 0·220
permission from the ethics committee. Resection
Right hemicolectomy 14 17 0·642
Normality of data was assessed using the Kol- Extended right hemicolectomy 4 5 1·000
mogorov–Smirnov test and visually using the Q-Q plot. High anterior resection 14 14 1·000
Continuous data are presented as mean(s.d.) if distributed Total/subtotal colectomy 5 1 0·200
normally and as median (range or interquartile range) oth- Laparoscopic/HALS 5 6 1·000
Colorectal POSSUM†
erwise. Two-tailed Fisher’s exact test, Mann–Whitney U Physiology score 9 (6–16) 9 (6–15) 0·214#
test and t test were used as appropriate for statistical anal- Operative severity score 7 (7–13) 7 (7–12) 0·925#
ysis. Repeated measures were analysed by ANOVA with Malignancy‡ 33 34 1·000
AJCC stage
Tukey’s correction. No subgroup analyses were planned
I 3 3 1·000
and no post hoc analyses were conducted. P < 0·050 was II 14 15 1·000
considered statistically significant and all data were anal- III 10 12 0·798
ysed using intention-to-treat principles. The results were IV 2 2 1·000

analysed using SPSS version 17.0 (IBM, Armonk, New

No. of lymph nodes excised* 24(10) 21(9) 0·300¶

York, USA).
Values are *mean(s.d.) and †median (range). ‡Includes high-grade
dysplasia and appendiceal carcinoid that were not included in American
Joint Committee on Cancer (AJCC) staging analysis. GDFT,
Results goal-directed fluid therapy; ASA, American Society of Anesthesiologists;
HALS, hand-assisted laparoscopic surgery; POSSUM, Physiological and
This study was conducted between November 2009 and Operative Severity Score for the enUmeration of Mortality and
September 2011. Patient recruitment is summarized in morbidity. §Two-tailed Fisher’s exact test, except ¶t test and
Fig. 2. Consistent ODM measurements were not available #Mann–Whitney U test.

 2012 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2013; 100: 66–74
Published by John Wiley & Sons Ltd
Goal-directed fluid therapy within an enhanced recovery protocol for elective colectomy 71

Table 3 Intraoperative management Table 5 Fluid-related variables

GDFT Fluid restriction P† GDFT Fluid restriction P‡

Time in operating theatre (min)* 133(56) 115(27) 0·488‡ Weight (kg)*

Vasopressor use 31 34 0·478 Preop. 76(16) 76(14) 0·631
Epidural analgesia 30 32 0·750 Day 1 78(16) 78(14) 0·542
Epidural successful 24 27 0·749 Day 2 78(17) 78(14) 0·766
Arterial line inserted 13 13 1·000 Day 3 78(17) 77(13) 0·209
Intraop. urine output (ml 1·03 (0·47–4·29) 0·99 (0·54–2·90) 0·689§
*Values are mean(s.d.). GDFT, goal-directed fluid therapy. †Two-tailed per kg per h)†
Fisher’s exact test, except ‡t test. 24-h urine output (ml per 0·75 (0·31–2·30) 0·77 (0·31–1·81) 0·885§
kg per h)†

Table 4 Intraoperative haemodynamic parameters Values are *mean(s.d.) and †median (range). GDFT, goal-directed fluid
therapy. ‡t test, except §Mann–Whitney U test.
GDFT Fluid restriction P*

Heart rate (beats/min)

First measurement 66(15) 62(14) 0·349 Table 6 Serum electrolytes and vasoactive hormones
Mean during operation 67(11) 64(12) 0·422
Final measurement 65(14) 63(12) 0·381 GDFT Fluid restriction P‡
Cardiac index (l per min
Sodium (mmol/l)*
per m2 )
Preop. 139(3) 138(3) 0·140
First measurement 3·9(1·7) 3·5(1·5) 0·255
Day 1 136(3) 136(3) 0·679
Mean during operation 4·1(1·4) 4·1(1·6) 0·847
Day 2 134(4) 135(4) 0·732
Final measurement 4·0(1·6) 4·2(1·7) 0·771
Day 3 135(5) 135(5) 0·831
Corrected flow time (ms)
Creatinine (µmol/l)*
First measurement 357(44) 341(43) 0·120
Preop. 76(26) 79(21) 0·499
Mean during operation 374(33) 355(30) 0·018
Day 1 77(23) 80(21) 0·592
Final measurement 382(41) 362(39) 0·039
Day 2 78(26) 76(25) 0·812
Day 3 74(30) 77(27) 0·636
Values are mean(s.d.). GDFT, goal-directed fluid therapy. *t test. Albumin (g/l)*
Preop. 40(4) 41(3) 0·201
Day 1 31(4) 32(5) 0·240
Preop. Day 1 Brain natriuretic peptide (pmol/l)†
Intraop. Day 2 Preop. 19 (5–112) 22 (3–217) 0·639§
1600 Day 0 Day 3
Day 1 49 (5–240) 50 (16–289) 0·525§
1400 Aldosterone/renin ratio*
1200
Volume (ml)

Preop. 10(10) 10(12) 0·830

1000 Day 1 9(11) 8(9) 0·898
800
600 Values are *mean(s.d.) and †median (range). GDFT, goal-directed fluid
400 therapy. ‡t test, except §Mann–Whitney U test.
200
0
Crystalloid Colloid Crystalloid Colloid
two groups. A similar number of patients required blood
GDFT Fluid restriction
transfusions (GDFT 13, fluid restriction 12; P = 1·000).
Fig. 3Perioperative fluid volumes before, during and on days Median (interquartile range) total postoperative fluid
0–3 after surgery in goal-directed fluid therapy (GDFT) and administered in the GDFT and fluid restriction groups was
fluid restriction groups 3750 (8808) and 2400 (9513) ml respectively (P = 0·604).
There were no differences in bodyweight, urine output,
Fluid management serum electrolytes and vasoactive hormones (Tables 5
Perioperative fluid administration is shown in Fig. 3. and 6).
Patients randomized to GDFT received significantly more
colloid during surgery than those randomized to fluid Surgical recovery and clinical outcomes
restriction (mean 591(471) versus 297(275) ml; P = 0·012)
and an overall greater volume of intravenous fluid There were no differences between the GDFT and fluid
(1994(590) versus 1614(420) ml; P = 0·010). There were restriction groups at any point in surgical recovery as
no other statistically significant differences between the measured by the SRS (mean SRS after 7 days 47 versus

 2012 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2013; 100: 66–74
Published by John Wiley & Sons Ltd
72 S. Srinivasa, M. H. G. Taylor, P. P. Singh, T.-C. Yu, M. Soop and A. G. Hill

90 GDFT
This study is dissimilar to previous trials as it explored
80 Fluid restriction the efficacy of GDFT alongside intraoperative fluid
Surgical recovery score

70
restriction. Moreover, this is one of the few trials to have
been conducted in an otherwise optimized perioperative
60
environment. It is notable that the most recent trials
50
exploring GDFT within an ERAS protocol, but without
40
fluid restriction, showed either equivalent or inferior
30 outcomes for patients randomized to GDFT21,22 . Thus,
20 the proposed benefits of GDFT may be offset by the other
17 advances in perioperative care; either that or avoidance of
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Time after surgery (days)
overload using fluid restriction may be an equally valid
intraoperative fluid administration strategy. In the present
Fig. 4Mean(s.d.) surgical recovery scores in goal-directed fluid trial, absolute weight gain was minimal in both groups and
therapy (GDFT) and fluid restriction groups hence unlikely to be associated with adverse outcomes.
Patients randomized to GDFT had a superior corrected
aortic flow time but there were no differences in cardiac
Table 7 Clinical outcomes
index between the groups. It is possible that higher values of
GDFT Fluid restriction P# cardiac indices are a measure of physiological reserve rather
than therapeutic targets23,24 . It is also important to note
No. of patients with complication(s) 26 27 1·000
Grade of complications
that the patients randomized to fluid restriction alone in
I 1 1 1·000 this trial showed acceptable intraoperative cardiac indices.
II 18 17 1·000 It is possible that observed cardiac indices in all patients
III 1 3 0·605
were adequate, and correction of occult hypovolaemia as
IV 6 4 0·744
V 0 2 0·494 seen in previous trials was not required10 , such that fluid
Major complications (grades III–V) 7 9 0·782 optimization had no demonstrable benefit. This may be a
Types of complication result of some of the elements of optimized perioperative
Surgical 14† 15‡ 1·000
Other 12§ 12¶ 1·000
care including free oral fluid until 2 h before surgery
Readmission 9 4 0·221 and preoperative carbohydrate loading25 . Although bowel
Length of hospital stay (days)* 6 (3–41) 5 (2–49) 0·570** preparation was used in some patients with left-sided
pathology in this trial, no statistically significant difference
*Values are median (range). †Intra-abdominal collection (2), anastomotic was observed between the two groups. It is, however,
leak (6), wound infection (5), ileus (1). ‡Anastomotic leak (5), wound
acknowledged that a type II error cannot be excluded.
infection (7), ileus (3). §Cardiorespiratory (5), urinary (3), haemorrhage
(2), other (2). ¶Cardiorespiratory (5), urinary (3), haemorrhage (4). This study was powered for an index of surgical recovery
GDFT, goal-directed fluid therapy. #Two-tailed Fisher’s exact test, in contrast to previous trials that were powered for length
except **Mann–Whitney U test. of hospital stay. Although this is a limitation, it has
been recognized previously that a single intervention is
unlikely to reduce hospital stay in an otherwise optimized
46 respectively; P = 0·853) (Fig. 4). Clinical outcomes are setting4,21,26 . As such, a sensitive measure such as the SRS,
summarized in Table 7. which also correlates with postoperative inflammation as
evidenced by cytokine release18 , is ideal for detection of
Discussion
any clinically meaningful benefits from interventions even
if they do not result in a shorter hospital stay17 . The
In this randomized trial GDFT did not provide congruence between all measured endpoints also suggests
any measurable benefit to patients undergoing elective a lack of benefit from GDFT in this setting. Nonetheless,
colectomy within an ERAS protocol incorporating fluid it is acknowledged that this trial was not powered to detect
restriction. Patients randomized to GDFT received more differences in postoperative complications specifically.
intraoperative colloid and a larger volume of intravenous Although GDFT allows individualized titration of
fluid overall, and had greater aortic flow. However, this did intraoperative intravenous fluids, it is possible that simply
not translate into any differences in surgical recovery, aiming for a neutral perioperative fluid balance is adequate
physiological variables, serum electrolytes, vasoactive for patients with physiological reserve able to correct minor
hormones or clinical outcomes. disturbances of homeostasis4 . The results of this study

 2012 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2013; 100: 66–74
Published by John Wiley & Sons Ltd
Goal-directed fluid therapy within an enhanced recovery protocol for elective colectomy 73

therefore may not be generalizable to patients who are 7 Varadhan KK, Lobo DN. A meta-analysis of randomised
physiologically compromised. However, in a recent trial, controlled trials of intravenous fluid therapy in major elective
patients with greater co-morbidities undergoing colorectal open abdominal surgery: getting the balance right. Proc Nutr
surgery experienced adverse outcomes following GDFT22 . Soc 2010; 69: 488–498.
8 WenKui Y, Ning L, JianFeng G, WeiQin L, ShaoQiu T,
It is also possible that any benefits gained from careful
Zhihui T et al. Restricted peri-operative fluid administration
titration of fluids in the intraoperative setting are offset by
adjusted by serum lactate level improved outcome after major
postoperative fluid administration, which is usually based
elective surgery for gastrointestinal malignancy. Surgery
on clinical assessment alone. 2010; 147: 542–552.
This randomized trial has demonstrated no effect 9 Bundgaard-Nielsen M, Holte K, Secher NH, Kehlet H.
of GDFT on recovery in patients undergoing elective Monitoring of peri-operative fluid administration by
colectomy within an ERAS protocol incorporating fluid individualized goal-directed therapy. Acta Anaesthesiol Scand
restriction. 2007; 51: 331–340.
10 Noblett SE, Snowden CP, Shenton BK, Horgan AF.
Randomized clinical trial assessing the effect of
Acknowledgements
Doppler-optimized fluid management on outcome after
S.S. and P.P.S. are recipients of the Auckland Medical elective colorectal resection. Br J Surg 2006; 93: 1069–1076.
Research Foundation Ruth Spencer Medical Research 11 Wakeling HG, McFall MR, Jenkins CS, Woods WG,
Miles WF, Barclay GR et al. Intraoperative oesophageal
Fellowship. T.-C.Y. is a recipient of a New Zealand
Doppler guided fluid management shortens postoperative
Health Research Council Clinical Training Scholarship.
hospital stay after major bowel surgery. Br J Anaesth 2005;
The ODM was lent by Pharmaco NZ for the duration 95: 634–642.
of the study. All disposable probes were purchased at 12 Gan TJ, Soppitt A, Maroof M, el-Moalem H,
regular cost, and Pharmaco NZ had no input into the Robertson KM, Moretti E et al. Goal-directed intraoperative
study design, data collection, interpretation of results or fluid administration reduces length of hospital stay after
decision to publish. major surgery. Anesthesiology 2002; 97: 820–826.
Disclosure: The authors declare no conflict of interest. 13 Conway DH, Mayall R, Abdul-Latif MS, Gilligan S,
Tackaberry C. Randomised controlled trial investigating the
influence of intravenous fluid titration using oesophageal
References
Doppler monitoring during bowel surgery. Anaesthesia 2002;
1 Brandstrup B, Tønnesen H, Beier-Holgersen R, Hjortsø E, 57: 845–849.
Ørding H, Lindorff-Larsen K et al.; Danish Study Group on 14 Abbas SM, Hill AG. Systematic review of the literature for
Perioperative Fluid Therapy. Effects of intravenous fluid the use of oesophageal Doppler monitor for fluid
restriction on postoperative complications: comparison of replacement in major abdominal surgery. Anaesthesia 2008;
two perioperative fluid regimens: a randomized 63: 44–51.
assessor-blinded multicenter trial. Ann Surg 2003; 238: 15 Jorgensen CC, Bundgaard-Nielsen M, Skovgaard LT,
641–648. Secher NH, Kehlet H. Stroke volume averaging for
2 Nisanevich V, Felsenstein I, Almogy G, Weissman C, individualized goal-directed fluid therapy with oesophageal
Einav S, Matot I. Effect of intraoperative fluid management Doppler. Acta Anaesthesiol Scand 2009; 53: 34–38.
on outcome after intraabdominal surgery. Anesthesiology 2005; 16 Zargar-Shoshtari K, Hill AG. Fast-track open colectomy is
103: 25–32. possible in a New Zealand public hospital. N Z Med J 2008;
3 Lobo DN, Bostock KA, Neal KR, Perkins AC, Rowlands BJ, 121: 33–36.
Allison SP. Effect of salt and water balance on recovery of 17 Zargar-Shoshtari K, Sammour T, Kahokehr A, Connolly AB,
gastrointestinal function after elective colonic resection: a Hill AG. Randomized clinical trial of the effect of
randomised controlled trial. Lancet 2002; 359: 1812–1818. glucocorticoids on peritoneal inflammation and postoperative
4 Abraham-Nordling M, Hjern F, Pollack J, Prytz M, Borg T, recovery after colectomy. Br J Surg 2009; 96: 1253–1261.
Kressner U. Randomized clinical trial of fluid restriction in 18 Paddison JS, Booth RJ, Fuchs D, Hill AG. Peritoneal
colorectal surgery. Br J Surg 2012; 99: 186–191. inflammation and fatigue experiences following colorectal
5 Lassen K, Soop M, Nygren J, Cox PBW, Hendry PO, surgery: a pilot study. Psychoneuroendocrinology 2008; 33:
Spies C et al.; Enhanced Recovery After Surgery (ERAS) 446–454.
Group. Consensus review of optimal perioperative care in 19 Paddison JS, Booth RJ, Hill AG, Cameron LD.
colorectal surgery: Enhanced Recovery After Surgery (ERAS) Comprehensive assessment of peri-operative fatigue:
Group recommendations. Arch Surg 2009; 144: 961–969. development of the Identity-Consequence Fatigue Scale.
6 Zargar-Shoshtari K, Connolly AB, Israel LH, Hill AG. J Psychosom Res 2006; 60: 615–622.
Fast-track surgery may reduce complications following major 20 Dindo D, Demartines N, Clavien P-A. Classification of
colonic surgery. Dis Colon Rectum 2008; 51: 1633–1640. surgical complications: a new proposal with evaluation in a

 2012 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2013; 100: 66–74
Published by John Wiley & Sons Ltd
74 S. Srinivasa, M. H. G. Taylor, P. P. Singh, T.-C. Yu, M. Soop and A. G. Hill

cohort of 6336 patients and results of a survey. Ann Surg therapeutic goals in high-risk surgical patients. Chest 1988;
2004; 240: 205–213. 94: 1176–1186.
21 Senagore AJ, Emery T, Luchtefeld M, Kim D, Dujovny N, 24 Velmahos GC, Demetriades D, Shoemaker WC, Chan LS,
Hoedema R. Fluid management for laparoscopic colectomy: Tatevossian R, Wo CC et al. Endpoints of resuscitation of
a prospective, randomized assessment of goal-directed critically injured patients: normal or supranormal? A
administration of balanced salt solution or hetastarch coupled prospective randomized trial. Ann Surg 2000; 232:
with an enhanced recovery program. Dis Col Rectum 2009; 52: 409–418.
1935–1940. 25 Ljungqvist O. Randomized clinical trial to compare the
22 Challand C, Struthers R, Sneyd JR, Erasmus PD, Mellor N, effects of preoperative oral carbohydrate versus placebo on
Hosie KB et al. Randomized controlled trial of intraoperative insulin resistance after colorectal surgery (Br J Surg 2010; 97:
goal-directed fluid therapy in aerobically fit and unfit patients 317–327). Br J Surg 2010; 97: 327–327.
having major colorectal surgery. Br J Anaesth 2012; 108: 26 Lassen K, Lobo DN. Randomized controlled trial of
53–62. preoperative oral carbohydrate treatment in major abdominal
23 Shoemaker WC, Appel PL, Kram HB, Waxman K, Lee TS. surgery (Br J Surg 2010; 97: 485–494). Br J Surg 2010; 97:
Prospective trial of supranormal values of survivors as 494–495.

 2012 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2013; 100: 66–74
Published by John Wiley & Sons Ltd