Warm water infusion versus air insufflation for unsedated colonoscopy: a randomized, controlled trial.

Abstract

Uncontrolled data suggest that warm water infusion (WWI) instead of air insufflation (AI) during the insertion phase of unsedated colonoscopy improves patient tolerance and satisfaction.
We tested the hypothesis that water could increase the proportion of patients able to complete unsedated colonoscopy and improve patient tolerance compared with the conventional procedure.
Randomized, controlled trial.
Single center, community hospital.
Consecutive outpatients agreeing to start colonoscopy without premedication.
Patients were randomly assigned to either WWI or AI insertion phase of colonoscopy. Sedation and/or analgesia were administered on patient request if significant pain or discomfort occurred.
Percentage of patients requiring sedation/analgesia. Pain and tolerance scores were assessed at discharge by using a 100-mm visual analog scale.
A total of 230 subjects (116 in the WWI group and 114 in the AI group) were enrolled. Intention-to-treat analysis showed that the proportion of patients requesting sedation/analgesia during the procedure (main outcome measurement) was 12.9% in the WWI group and 21.9% in AI group (P = .07). Cecal intubation rates were 94% in the WWI group and 95.6% in the AI group (P = .57). Median (interquartile range) scores for pain were 28 (12-44) and 39 (14-54) in WWI and AI groups, respectively (P = .05); corresponding figures for tolerance were 10 (3-18) and 14 (5-42), respectively (P = .01). The adenoma detection rates were 25% and 40.1% for the WWI and AI groups, respectively (P = .013).
Single-center study, endoscopists not blinded to randomization.
WWI instead of AI is not associated with a statistically significant decrease in the number of patients requiring on-demand sedation, although it significantly improves the overall patient tolerance of colonoscopy. The finding of a lower adenoma detection rate in the WWI group calls for further evaluations. (Clinical trial registration number: NCT00905554).

Full text

Page 1

ORIGINAL ARTICLE: Clinical Endoscopy
Warm water infusion versus air insufflation for unsedated colonoscopy:
a randomized, controlled trial
Franco Radaelli, MD, Silvia Paggi, MD, Arnaldo Amato, MD, Vittorio Terruzzi, MD
Como, Italy
Background: Uncontrolled data suggest that warm water infusion (WWI) instead of air insufflation (AI) during
the insertion phase of unsedated colonoscopy improves patient tolerance and satisfaction.
Objective: We tested the hypothesis that water could increase the proportion of patients able to complete
unsedated colonoscopy and improve patient tolerance compared with the conventional procedure.
Design: Randomized, controlled trial.
Setting: Single center, community hospital.
Patients: Consecutive outpatients agreeing to start colonoscopy without premedication.
Methods: Patients were randomly assigned to either WWI or AI insertion phase of colonoscopy. Sedation and/or
analgesia were administered on patient request if significant pain or discomfort occurred.
Main Outcome Measurements: Percentage of patients requiring sedation/analgesia. Pain and tolerance scores
were assessed at discharge by using a 100-mm visual analog scale.
Results: A total of 230 subjects (116 in the WWI group and 114 in the AI group) were enrolled. Intention-to-treat
analysis showed that the proportion of patients requesting sedation/analgesia during the procedure (main
outcome measurement) was 12.9% in the WWI group and 21.9% in AI group (P ? .07). Cecal intubation rates
were 94% in the WWI group and 95.6% in the AI group (P ? .57). Median (interquartile range) scores for
pain were 28 (12-44) and 39 (14-54) in WWI and AI groups, respectively (P ? .05); corresponding figures for
tolerance were 10 (3-18) and 14 (5-42), respectively (P ? .01). The adenoma detection rates were 25% and 40.1%
for the WWI and AI groups, respectively (P ? .013).
Limitations: Single-center study, endoscopists not blinded to randomization.
Conclusions: WWI instead of AI is not associated with a statistically significant decrease in the number of
patients requiring on-demand sedation, although it significantly improves the overall patient tolerance of
colonoscopy. The finding of a lower adenoma detection rate in the WWI group calls for further evaluations.
(Clinical trial registration number: NCT00905554). (Gastrointest Endosc 2010;72:701-9.)
The recent increase in colonoscopy use in clinical prac-
tice has led to an imbalance between requests and re-
sources in most endoscopy services.1
In this clinical scenario, there is growing interest in
measures and corrective actions aimed at improving the
practice efficiency of endoscopy services by minimizing
patient recovery time and decreasing the procedure-
related costs. In particular, much attention has been gen-
erated by rapid recovery sedation regimens such as
propofol2-5and fospropofol disodium,6which are charac-
Abbreviations: AI, air insufflation; IQR, interquartile range; WWI, warm
water infusion.
DISCLOSURE: All authors disclosed no financial relationships relevant to
this publication.
Copyright © 2010 by the American Society for Gastrointestinal Endoscopy
0016-5107/$36.00
doi:10.1016/j.gie.2010.06.025
Received February 5, 2010. Accepted June 14, 2010.
Current affiliations: Division of Gastroenterology, Valduce Hospital, Como,
Italy.
Presented at the 2009 ACG Annual Scientific Meeting, October 23-28, San
Diego, California.
Reprint requests: Franco Radaelli, MD, Divisione di Gastroenterologia,
Ospedale Valduce, Via Dante 11, 22100 Como, Italy.
If you would like to chat with an author of this article, you may contact Dr.
Radaelli at francoradaelli@virgilio.it.
www.giejournal.org Volume 72, No. 4 : 2010
GASTROINTESTINAL ENDOSCOPY
701

Page 2

terized by immediate resolution of sedation and recovery
of mental and psychomotor capacities. However, new
consideration has been given to unsedated colonoscopy,
which allows further theoretical advantages, such as a
lower incidence of cardiorespiratory adverse events, no
restriction on patient activity after the procedure (patients
are potentially able to work on the examination day), and
overall lower costs, both direct and indirect.7
On the other hand, unsedated colonoscopy is reported
to be painful and poorly tolerated by patients because of
both air insufflation (AI) and instrument insertion, and it is
associated with a worsening of technical performance out-
comes.8These issues are of major importance because
they may negatively affect patient compliance with the
examination.
Recently, some authors reported a possible role for
warm water infusion (WWI) in relieving colonic spasms
and facilitating colonoscopy performance.9-13All of these
studies used WWI as an adjunct to the usual AI during
colonoscope insertion. Leung et al14-16proposed the tech-
nique of using WWI without AI during the insertion phase
of colonoscopy. This technique is based on infusion of
warm water (37°C) to distend areas where the lumen is
collapsed. Their initial and observational studies indicated
that WWI could minimize the procedure-related discom-
fort or pain without compromising the technical perfor-
mance of the colonoscopy itself.14-16In a recent random-
ized trial conducted with veterans undergoing screening
colonoscopywithsedation/analgesia,warmwaterinfusionwas
associated with lower pain scores and decreased sedation
requirements.17
To date, the role of the WWI technique in unsedated
colonoscopy has not been extensively investigated. In this
randomized, controlled trial, we tested the hypothesis that
WWI during the insertion phase of colonoscopy could
increase the proportion of patients able to complete the
colonoscopy without sedation/analgesia and could im-
prove the overall tolerance of the colonoscopy itself com-
pared with the traditional procedure of using AI.
MATERIALS AND METHODS
The study was performed in a community hospital. The
protocol was registered on ClinicalTrials.gov (NCT0090555).
It was reviewed and approved by the Hospital Review
Board of Medical Ethics on March 18, 2009. Written in-
formed consent was obtained from all of the patients at
enrollment.
Study population
All consecutive 18- to 80-year-old outpatients present-
ing for open-access colonoscopy for routine indications
performed by 1 of 3 staff gastroenterologists (F.R., A.A.,
S.P.) between April and July 2009 were considered eligible
to enter the study. Exclusion criteria were (1) patient
request for sedation/analgesia before starting the proce-
dure, (2) previous colorectal surgery, (3) indication for a
bidirectional endoscopy, (4) indication for a partial exam-
ination, (5) patient refusal or inability to provide informed
consent, and (6) advanced heart, lung, liver, or renal
disease (American Society of Anesthesiology risk class
III-IV). Patients meeting inclusion criteria could begin un-
dergoing a colonoscopy without routine sedation/analgesia
and were randomized in case of acceptance.
Study procedure and randomization
Procedures were started without routine sedation and
analgesia by 3 board-certified gastroenterologists. Two of
them (F.R., A.A.) were highly experienced (each having
performed ?5000 colonoscopies), whereas the other one
(S.P.) had much more limited experience and was consid-
ered still in training (having performed ?300 colonosco-
pies). The study was started after a 1-month training
period during which all 3 endoscopists performed approxima-
tely 30 colonoscopies with WWI without AI to become
accustomed with the technique. High-resolution wide-
angle adult video colonoscopes (Olympus HD 180 series;
Olympus Corp, Hamburg, Germany) were used for all
study procedures. A senna-based preparation was given,
consisting in a low-residue diet for 2 days before and
high-dose senna (24 tablets in 2 divided doses) on the day
before the procedure.18
Patients meeting the inclusion criteria and agreeing to
participate in the study were randomly assigned to a WWI
or AI insertion phase of the colonoscopy. Randomization
was done by means of a computer-generated random
number sequence, considering 3 different operators (mixed
block size). Allocation was concealed and kept in a sealed
envelope, opened after the informed consent was ob-
tained. The patient, but not the endoscopist, was blinded
to the randomization modality.
Before starting the procedure, a brief questionnaire
recording demographic data, previous abdominal surgery,
main indication for the colonoscopy, and level of patient
anxiety (none, mild, moderate, or severe) was adminis-
tered by the endoscopist. Colonoscopy began with the
patient in the left lateral position.
Take-home Message
● The use of warm water infusion during endoscope
insertion is associated with a trend in a decrease in
patients requiring on-demand sedation. Overall tolerance
and pain scores of unsedated procedures are significantly
higher with this technique, whereas the cecal intubation
rate is not affected.
● This technique might play a role in improving acceptance
and tolerability of colonoscopy for patients willing or
needing to undergo unsedated procedures. The lower
adenoma detection rate associated with this novel
technique, never reported before, is a matter of major
concern and needs further evaluation.
Warm water infusion for unsedated colonoscopy Radaelli et al
702
GASTROINTESTINAL ENDOSCOPY
Volume 72, No. 4 : 2010www.giejournal.org

Page 3

For patients randomized to the WWI group, a liquid
reservoir was filled with 37°C water before starting the
colonoscopy; the water temperature was kept constant by
coating the reservoir with a heat-saver envelope. AI was
used just at instrument introduction until the rectosigmoid
junction was reached, and then turned off during the
procedure until the cecum was reached. During colono-
scope insertion, warm water was intermittently infused by
means of a peristaltic flushing pump (Olympus OFP;
Olympus Corp) through an auxiliary water channel of the
colonoscope to obtain lumen distention. A press button on
the colonoscope controlled water feed and the flow rate
was set at 210 mL/min. There was no restriction of the
overall volume of water that could be infused to get
adequate lumen distention. However, if the visualization
of colonic lumen could not be achieved with WWI and the
endoscopist judged it unsafe to advance the colonoscope,
air was briefly (lasting no more than 10 seconds) turned
on. Any turning on of air was specifically recorded in the
patient data sheet. If 3 or more episodes of AI were
needed, the water pump was turned off and the procedure
was completed with air. In this case, the cecal intubation
outcome was recorded as a failure. During the withdrawal
phase, water and stool residuals were suctioned in each
colonic tract before mucosal inspection; moreover, AI was
used to obtain an adequate distention of the colonic
lumen.
For patients randomized to the AI group, colonoscopy
was performed in the standard fashion, with routine AI.
Aliquots of 30 to 50 mL of water at room temperature were
used for washing residual stool, as needed.
For both groups, polyps were always removed in the
withdrawal phase; if they were detected during the as-
cending phase of colonoscopy, they were relocated and
removed during the subsequent withdrawal phase of the
procedure.
At regular intervals during the colonoscopy, patients
were asked by both the endoscopist and GI assistant about
abdominal discomfort and pain and whether they wanted
sedation and/or analgesia. Sedation and analgesia were
administered at the patient’s request at fixed intravenous
boluses of 2.5 mg midazolam and 50 mg meperidine.
Additional medication could be given at the endoscopist’s
discretion at any time during the procedure depending on
the perceived depth and quality of sedation. Patient posi-
tion was changed to facilitate the passage of the colono-
scope, if needed.
The success of cecal intubation was verified by the
endoscopist and the GI assistant by the identification of
the ileocecal valve and the appendix orifice. Insertion time
to the cecum and total procedure time (insertion to re-
moval of the colonoscope) were measured.
All patients were monitored by continuous pulse oxim-
etry and visual clinical assessment of ventilatory status.
Significant oxygen desaturation was diagnosed when val-
ues less than 90% were maintained for more than 15
seconds. Supplemental oxygen (2 L/min by means of a
nasal cannula) was administered for all episodes in which
oxygen saturation was less than 85% or less than 90% on
3 separate occasions; these data were recorded as signifi-
cant outcomes. A vagal reaction was defined as an episode
of bradycardia (heart rate ?60 beats per minute) associ-
ated with excessive sweating and nausea and/or vomiting.
After the completion of the procedure, any abnormal
findings, operative procedure (ie, polypectomy), change
of patient position during colonoscopy, and procedure-
related adverse event were recorded. The level of techni-
cal difficulty of the examination (subjectively rated by the
endoscopist as easy/fairly easy to difficult/failure), the
procedure and recovery times, total hospital stay (time
from the start of the procedure to discharge, calculated
adding recovery time to total procedure time), and the
quality of bowel preparation (excellent, good, sufficient,
poor, according to the Aronchik scale)19were also re-
corded (excellent: minimal amount of liquid stools, mini-
mal aspiration needed; good: mainly liquid stools, fre-
quent aspiration needed, no limitation of the colonoscopy;
sufficient: liquid and semisolid stools, frequent aspiration
and washings needed, small lesions may be missed; and
inadequate: ?10% of mucosa not visualized, presence of
solid or semisolid stools, aspiration not possible, repeat
colonoscopy required).
If polyps were detected, pathology reports were re-
corded to assess adenoma prevalence.
Patient assessment
Patient pain and tolerance were assessed at discharge
(Time 0) by an independent interviewer, blind to the
randomization, by means of a validated visual analog
scale.20,21Patients were asked to place a vertical mark
anywhere along a 100-mm scale to represent their toler-
ance (0 ? excellent, 100 ? unbearable) and pain (0 ? no
pain, 100 ? worst pain ever). Patients were also asked
whether they would be willing to have the procedure
repeated in the future under the same circumstances (yes,
probably yes, probably no, certainly no). Finally, the over-
all tolerance of the examination (excellent, tolerable,
poorly tolerable, or unacceptable) and the willingness to
repeat the colonoscopy under the same circumstances
(yes, probably yes, probably no, certainly no) were as-
sessed after 24 hours (time 24) by phone by a blinded GI
assistant.
Study endpoints
The primary endpoint of the study was to evaluate
whether the use of WWI during the insertion phase of
colonoscopy could reduce the proportion of patients re-
quiring on-demand sedation/analgesia. Secondary out-
come measures included (1) pain and tolerance scores, (2)
willingness to repeat the colonoscopy (time0and time24),
(3) procedure times and total hospital stay, and (4) cecal
intubation rate.
Radaelli et al Warm water infusion for unsedated colonoscopy
www.giejournal.org Volume 72, No. 4 : 2010
GASTROINTESTINAL ENDOSCOPY
703

Page 4

Sample size and statistical analysis
Statistical power was calculated before starting the
study for the primary outcome variable (proportion of
patients requiring on-demand sedation/analgesia). The
sample size calculation was based on the estimation that a
decrease in the percentage of patients requiring on-
demand sedation from 25% to 10% would have been
clinically relevant. In our previous experience, the 25% of
patients agreeing to undergo unsedated colonoscopy
asked for sedation to complete the examination; these
data are also confirmed by the literature.22Based on a
decrease from 25% to 10% and taking into account the
presence of 3 different examiners, we estimated that 226
patients (113 patients in each arm) would have been
required for an ? value of .05 and a power of 80%.
The 3 examiners were divided into 2 groups (2 experts
[F.R., A.A.] and a trainee [S.P.]), and a subgroup analysis
was preliminarily planned according to the endoscopy
expertise for principal outcome measures (proportion of
patients requiring on-demand sedation/analgesia, pain
and tolerability scores, insertion to cecum procedure time,
and cecal intubation rate).
For statistical analysis, data were entered into a statistical
software program (Intercooled Stata; Stata Corp, College Sta-
tion, Tex). All analyses were performed by using an
intention-to-treat approach. Descriptive statistics were com-
puted for all variables. Normally and non-normally distrib-
uted data were represented by using standard deviation and
interquartile range (IQR), respectively. Chi-square and t tests
were used to compare proportions and means for normally
distributed data, as appropriate; median values for non-
normallydistributeddatawerecomparedbyusingtheMann-
Whitney rank sum test. For all tests, a P value ?.05 was
considered statistically significant.
RESULTS
Baseline characteristics
Among 454 outpatients evaluated by the 3 investigators
during the study period, 397 met the inclusion criteria and
were offered enrollment in the study. Of these, 167 sub-
jects (42% men, mean age 60.4 ? 12.1 [standard deviation]
years) refused enrollment in the study; of them, 68 (40.7%)
had undergone a previous colonoscopy (29 and 39 sub-
Figure 1. Study flow diagram.
Warm water infusion for unsedated colonoscopy Radaelli et al
704
GASTROINTESTINAL ENDOSCOPY
Volume 72, No. 4 : 2010www.giejournal.org

Page 5

jects with and without premedication, respectively). The
remaining 230 (58%) patients agreed to participate and
were randomly allocated to the investigational arm (WWI
group, 116 patients) or the control arm (AI group, 114
patients) (Fig. 1). The 2 groups were well balanced with
regard to age, sex, body mass index, previous major ab-
dominal or pelvic surgery, previous colonoscopy, and
stated anxiety (Table 1). The procedural features (indica-
tion for the colonoscopy, quality of bowel preparation)
were also equally distributed between the 2 groups, as
listed in Table 1.
Of 230 colonoscopies, 219 (95.2%) were completed to
the cecum. In the remaining 11 cases (6 in the WWI group
and 5 in the AI group), cecal intubation was not achieved
because of poor bowel preparation (8 cases), tight malig-
nant strictures (2 cases), and technical difficulties (1 case).
Patient tolerance and pain
Among 230 patients enrolled in the trial, 190 (82.6%)
completed the procedure without sedation/analgesia. The
number of patients who requested and were given
sedation/analgesia during the insertion phase of the ex-
amination (primary outcome measurement) was 15 in the
WWI group and 25 in the AI group (12.9 vs 21.9%, P ?
.07). Patient assessments of tolerance and pain (median,
IQR) by using the visual analog scale are displayed in
Figure 2 and summarized in Table 2. Median (IQR) scores
for pain were 28 (12-44) and 39 (14-54) in the WWI group
and AI group, respectively (P ? .05); corresponding fig-
ures for tolerance were 10 (3-18) and 14 (5-42), respec-
tively (P ? .01). The proportion of patients who, at dis-
charge, reported they would be willing to have the
procedure repeated in the future under the same circum-
stances was 90.5% in the WWI group and 81.6% in the AI
group (P ? .05). At phone interview the day after the
Figure 2. A, Patient pain scores. Scores (median, IQR) obtained by using
100-mm visual analog scale (0 ? no pain, 100 ? worst pain ever). B,
Patient tolerance scores. Scores (median, IQR) obtained by using 100-mm
visual analog scale (0 ? excellent, 100 ? unbearable).
TABLE 1. Demographic and clinical characteristics of
subjects
WWI group
(n ? 116)
AI group
(n ? 114)
P
value
Male*68 (58.6)66 (57.8) .91
Age (y)†
58.4 (11.5)58.8 (13.3).88
Body mass index
(kg/m2)†
24.8 (4.6)25.1 (3.6) .89
Previous major
abdominal or pelvic
surgery*
37 (31.9)34 (29.8) .73
Previous colonoscopy*37 (31.9)42 (36.8) .92
Patient anxiety.97
None/mild*71 (61.2)70 (61.4)
Moderate/severe45 (38.8) 44 (38.6)
Indication for
colonoscopy
Screening/positive
fecal occult blood
test results
46 (39.7) 42 (36.8).66
Surveillance* 20 (17.2)24 (21.1) .58
Rectal bleeding*19 (16.4) 19 (16.7).97
Pain as main
symptom*
17 (14.7) 16 (14.0).89
Other indications*14 (12.1)13 (11.4) .91
Excellent-good
preparation*
109 (94) 108 (94.7) .80
Endoscopist expertise.85
Experts* 77 (66.4)77 (67.5)
Trainee*39 (33.6)37 (32.5)
AI, Air insufflation; WWI, warm water infusion.
*Number (%) of patients, ?2test.
†Mean (standard deviation), Student t test.
Radaelli et al Warm water infusion for unsedated colonoscopy
www.giejournal.org Volume 72, No. 4 : 2010
GASTROINTESTINAL ENDOSCOPY
705

Page 6

procedure, the proportion of patients who reported the
overall tolerance of the procedure as excellent or good
was 92.2% in the WWI group and 81.6% in the AI group
(P ? .02), whereas no significant difference between the 2
groups was found with regard to willingness to repeat the
procedure (90.5% vs 83.3%, P ? .10).
Procedural data
Procedural data are summarized in Table 3. Cecal intu-
bation was achieved in 110 (94.8%) procedures in the
WWI group and in 109 (95.6%) cases in the AI group (P ?
.57). In 12 patients of the WWI group, it was necessary to
briefly turn on air once or twice to get an adequate visu-
alization of the colonic lumen. Moreover, in another pa-
tient in the same group, repeated AIs were needed to
proceed with the examination, which was subsequently
completed with air. According to the intention-to-treat
analysis, this patient was included in the investigational
group; however, in this case, cecal intubation was consid-
ered “failed” for the purpose of the study, even if the
cecum was reached. As a consequence, the cecal intuba-
tion in the WWI group decreased from 110 to 109 of 116
patients (94%). The median (IQR) time to reach the cecum
was 7 (5-11) and 5 (4-8) minutes in the WWI and AI
groups, respectively (P ? .001). No significant differences
were observed between the 2 groups regarding either the
total procedure or recovery times or the total hospital stay.
There was no significant difference between the 2
groups with regard to the proportion of patients requiring
positional changes during the procedure (40.5% vs 42.3%,
P ? .70) and the overall difficulty of the examination, as
rated by the endoscopists.
A significantly lower proportion of patients with at
least 1 polyp (polyp detection rate) was found in the
WWI group than in the AI group (31.9% vs 50%, P ?
.005). The adenoma detection rate was 25% and 40.1%,
in the WWI and AI groups, respectively (P ? .013). In
the WWI group, a total of 65 polyps were found, of
which 49 were adenomas (12 hyperplastic, the other 4
were normal mucosa). In the AI group, corresponding
figures for polyps and adenomas were 100 and 80,
respectively (17 hyperplastic and 3 normal mucosa). In
the WWI and AI groups, no difference was observed
with regard to either the mean number of polyps per
patient (1.75; 95% CI, 1.34-2.18 vs 1.76; 95% CI, 1.31-2.0;
P ? .82) or the mean number of adenomas per patient
(1.70; 95% CI, 1.28-2.22 vs 1.74; 95% CI, 1.30-2.11; P ?
.46). In the WWI and AI groups, no difference was
observed with regard to either the advanced adenoma
detection rate (7% vs 10.5%, P ? .32) or the mean
number of advanced adenomas per patient (1.83; 95%
CI, 0.99-2.21 vs 1.84; 95% CI, 0.96-2.35; P ? .45).
Significantly more operative procedures were per-
formed (32.7% vs 49.1%, P ? .01) in the AI compared with
the WWI group.
Adverse events
Four patients (3.4%) in the WWI group and 9 patients
(7.9%) in the AI group experienced sustained oxygen
desaturation (P ? .14). All of them were given sedation/
analgesia during the procedure. The proportions of pa-
tients requiring supplemental oxygen, according to the
protocol, were similar in the 2 groups (3% vs 5.3%, P ?
.30). Two patients in the AI group experienced vagal
reactions. There were 2 cases of postpolypectomy bleed-
ing (1 in each group) requiring endoscopic intervention;
their subsequent clinical course was uneventful.
Endoscopists’ expertise
According to randomization blocks, 154 procedures
were performed by the 2 expert endoscopists and the
remaining 76 by the trainee. To evaluate whether there
were differences between the WWI and AI groups accord-
ing to the level of endoscopist expertise, the same analysis
performed on the whole cohort was repeated by splitting
the experts and the trainee. In this subgroup analysis, no
difference was found with regard to the primary endpoint
(request of on-demand medications) and other secondary
outcomes (time to reach the cecum, cecal intubation,
polyp and adenoma detection rates) for both the experts
and the trainee. Conversely, when WWI was used instead
of AI, tolerance and pain scores were significantly lower
for the experts only (Table 4).
TABLE 2. Patient tolerability and pain
WWI group
(n ? 116)
AI group
(n ? 114)
P value
On-demand
sedation/
analgesia*
15 (12.9) 25 (21.9) .07
Pain score†‡ 28 (12-44) 39 (14-54).05
Tolerability score†‡ 10 (3-18)14 (5-42).01
Tolerability
excellent-good
time24*
107 (92.2)93 (81.6) .02
Willingness to
repeat the
procedure
Yes/probably yes
time0*
105 (90.5)93 (81.6).05
Yes/probably yes
time24*
105 (90.5) 95 (83.3).10
AI, Air insufflation; WWI, warm water infusion.
time 0 ? at discharge.
time 24 ? the day after the procedure.
*Number (%) of procedures, ?2test.
†Median (interquartile range), Mann-Whitney rank sum test.
Warm water infusion for unsedated colonoscopyRadaelli et al
706
GASTROINTESTINAL ENDOSCOPY
Volume 72, No. 4 : 2010www.giejournal.org

Page 7

DISCUSSION
This study shows that in patients undergoing colonos-
copy without routine premedication, the use of WWI in-
stead of AI during the insertion phase improves the overall
tolerability of the examination, although it does not sig-
nificantly decrease the proportion of patients requiring
on-demand sedation during the procedure.
The proportion of patients asking for on-demand seda-
tion was chosen as the primary outcome measurement
because the study was mainly aimed at evaluating a strat-
egy that could potentially increase the proportion of pa-
tients able to complete colonoscopy unsedated, without
compromising the success rate of cecal intubation, and
therefore could favorably affect procedure-related costs
and performance efficiency of endoscopy services. This
issue was not supported by our data, although a trend in
favor of WWI was observed. The overall tolerance of the
procedure is a major determinant of patient compliance
with the examination. For this purpose, the perceived
levels of pain and tolerance as well as willingness to
undergo a repeat unsedated procedure were chosen as
secondary outcome measures, and all were significantly
better in with WWI than AI.
These results partially confirm those reported in a re-
cent randomized, controlled study in minimally sedated
patients in which a possible role for warm water in mini-
mizing both pain and drug dose for sedation was demon-
strated.17It is hypothesized that these positive effects of
WWI depend on both the decrease in colonic spasms and
local distention of the colonic wall. Furthermore, the
weight of the water in the left side of the colon also
straightens the sigmoid colon.9,11Our technique was
slightly different from that applied by Leung et al15be-
cause brief periods of AI were allowed in cases of extreme
technical difficulty. However, approximately 90% of the
procedures were completed with WWI, without use of AI,
which makes our findings generally comparable to those
reported in the literature.15Moreover, considering that
occasional periods of AI were allowed by the protocol, the
proportion of procedures performed in accordance with
the study design increased to more than 99%. These data,
along with the high rate of acceptance of on-demand
sedation colonoscopy in our study population, indicate a
high applicability of the technique in current clinical prac-
tice. It could be argued that the occasional use of AI could
have affected the tolerance and pain of the examination;
TABLE 3. Procedural data
WWI group
(n ? 116)
AI group
(n ? 114)
P value
Cecal intubation*109 (94) 109 (95.6) .57
Time to cecum (min)†
7 (5-11) 5 (4-8)
?.001
Total procedure time (min)†
15 (12-20)14 (11-18).07
Recovery time (min)†
5 (0-15)5 (0-20) .42
Total hospital stay (min)†
21 (15-34) 21 (12-37) .89
Polyp detection rate* 37 (31.9)57 (50).01
No. of polyps per patient‡
1.75 (1.34-2.18)1.76 (1.31-2.0).82
Adenoma detection rate* 29 (25)46 (40.1) .01
No. of adenomas per patient‡
1.70 (1.28-2.22) 1.74 (1.30-2.11).46
Advanced adenoma detection rate* 8 (7) 12 (10.5).32
No. of advanced adenomas per patient‡
1.83 (0.99-2.21) 1.84 (0.96-2.34).45
Operative procedures* 38 (33)56 (49) .01
Technical difficulty (endoscopist assessment).95
Easy/fairly easy* 94 (81) 92 (80.7)
Difficult/failure*22 (19) 22 (19.3)
Patient positional changes during the procedure* 47 (40.5)48 (42.1) .70
AI, Air insufflation; WWI, warm water infusion.
*Number (%) of procedures, ?2test.
†Median (interquartile range), Mann-Whitney rank sum test.
‡Mean (95% CI), Student t test.
Radaelli et alWarm water infusion for unsedated colonoscopy
www.giejournal.org Volume 72, No. 4 : 2010
GASTROINTESTINAL ENDOSCOPY
707

Page 8

for this purpose, we performed a sensitivity analysis that
did not show any difference in these items.
From a technical point of view, the use of WWI does
not seem to increase the difficulty of the procedures. No
difference was observed between the 2 groups with regard
to the evaluation of the level of difficulty of the examina-
tions as rated by the endoscopists and the proportions of
patients requiring positional changes during the proce-
dure. Our data are partially in line with those recently
reported by Ramirez et al23and Ransibrahmanakul et al,24
who reported that the technique is characterized by a short
learning curve and is not influenced by endoscopist ex-
pertise. In fact, the level of endoscopist expertise seems to
affect the benefits of the use of WWI. Tolerance and pain
scores were lower in the WWI group, although the sub-
group analysis showed that this finding seems to be influ-
enced by the performance of the experts only (no differ-
ence between the WWI and AI groups for the trainee).
This observation suggests that the most “profitable” appli-
cation of WWI may require a kind of technical endoscopic
background for it to be really advantageous.
According to our data, the use of WWI is characterized
by a significantly longer time to reach the cecum, which is
likely related to the quicker efficiency of air in obtaining
lumen distention. A longer withdrawal time and as a con-
sequence a longer total procedure time were expected for
the WWI group, based on the hypothesis that more liquid
suction would be needed in this arm. This issue could not
be confirmed because we did not measure the true mu-
cosal inspection time by using a stopwatch. Moreover, a
significantly higher number of operative procedures were
performed in the AI group, thus increasing the overall
withdrawal time in this group and potentially hiding dif-
ferences in the total procedure time between the 2 groups.
Thelowerpolypandadenomadetectionratesinthestudy
arm, although the 2 groups were well balanced with regard
to age, indication for the procedure, and quality of bowel
preparation, are one of the most impressive findings of the
study and deserve comment. It is worth emphasizing that all
3 endoscopists consistently obtained a lower adenoma and
polyp detection rate in the WWI versus AI group. However,
the clinical relevance of this finding should not be overem-
phasized because no difference was observed concerning
the advanced adenoma detection rate. Our data are in con-
trast with those presented by Leung et al,15who observed a
trend toward improved adenoma detection in a water instil-
lation group. It is well-known that most colonic lesions are
detected in the withdrawal phase of the procedure, which in
this study was also performed with the standard technique
(AI). We may speculate that greater lumen distention, posi-
tively influencing the adenoma detection, might have been
achieved in the AI group because of AI in the ascending
phase of colonoscopy. Indeed, a recent study by Morini et
al25assessed a potential role for the insertion phase of
colonoscopy in detecting colonic lesions, thus determining a
potentialbiasfortheWWIarm,althoughthisobservationhas
been debated and needs for further confirmation.26
Compared with previous reports on the use of WWI,
the current study has some strengths. First, the WWI tech-
nique for unsedated colonoscopy had never been inves-
tigated in a randomized, controlled trial. The randomized
design allows the comparison of 2 homogeneous sub-
groups and the presence of female subjects, not only
males, as in the studies with U.S. veterans,15,17increases
the generalizability of data. Second, the inclusion of a
larger patient population than that analyzed in previous
studies15,17makes our trial adequately powered to evalu-
ate the predefined outcome measures.
Several limitations of the study deserve comment. The
main drawback was the unblinding of the endoscopist
with regard to the insertion phase of the colonoscopy
TABLE 4. Outcome measures according to
endoscopists’ expertise
WWI group
(n ? 116)
AI group
(n ? 114)
P value
On-demand
sedation/analgesia*
Experts 9/77 (11.7)17/77 .08
Trainee 6/39 (15.4)8/37 (21.6).48
Tolerance score†
Experts10 (3-19)14 (5-43).03
Trainee13 (4-24) 14 (5-42).35
Pain score†
Experts 26 (12-43)39 (11-62).05
Trainee 31 (14-45) 37 (15-45).62
Time to reach cecum
(min)†
Experts7 (5-10) 6 (4-10).02
Trainee 6 (5-12) 5 (4-6).01
Cecal intubation*
Experts72/77 (93.5) 73/77 (94.8) .73
Trainee 38/39 (97.4) 36/37 (97.3).97
Polyp detection rate
Experts25/77 (32.4) 37/77 (48.1) .05
Trainee 12/39 (30.8) 20/37 (54.1).04
Adenoma detection
rate
Experts 20/77 (26)31/77 (40.3) .06
Trainee 9/39 (23.1) 15/37 (40.5).10
AI, Air insufflation; WWI, warm water infusion.
*Number (%) of procedures, ?2test.
†Median (interquartile range), Mann-Whitney rank sum test.
Warm water infusion for unsedated colonoscopyRadaelli et al
708
GASTROINTESTINAL ENDOSCOPY
Volume 72, No. 4 : 2010www.giejournal.org

Page 9

because he was directly involved in the decision to ad-
minister on-demand sedation/analgesia on patient re-
quest. It might be argued that the operator could have
unintentionally chosen to delay sedation/analgesia in the
interventional arm. This hypothesis does not seem to be
supported by our data because no difference was ob-
served either in the proportion of patients requiring drugs
or in the tolerability levels of the procedure, the latter
being higher for the WWI group. Another potential limi-
tation may be attributable to the 2:1 ratio between experts
and trainee in our study because the positive or negative
impact of a novel technique may affect more the perfor-
mances of less-experienced operators. However, this ratio
was determined by local human resources, and this pitfall
should have been minimized by the calculation of the
sample size of the study targeted by 3 different endosco-
pists, then arbitrarily grouped as experts and trainee. Fur-
thermore, the percentage of patients requesting sedation/
analgesia observed in the study was quite low. This fact
might be partially explained by a positive selection bias
because patients asking for sedated procedure were not a
priori included in the study. Last but not least, this was a
single-center study and its results need to be externally
validated.
In conclusion, this randomized, controlled trial shows
that the use of WWI instead of AI during colonoscopy is
associated with a decreased number of patients requiring
on-demand sedation and significantly increases the overall
tolerance of the procedure, without compromising the
success rate of cecal intubation. We do not believe that
unsedated colonoscopy may replace sedated procedures.
However, this technique might play a role in improving
acceptance and tolerability of colonoscopy for patients
willing or needing to undergo unsedated procedures. The
lower adenoma detection rate associated with this novel
technique, never reported before, is a matter of major
concern, and further evaluation is needed.
REFERENCES
1. Rex DK, Johnson DA, Anderson JC, et al. American College of Gastroen-
terology.AmericanCollegeofGastroenterologyguidelinesforcolorec-
tal cancer screening 2009. Am J Gastroenterol 2009;104:739-50.
2. Standards of Practice Committee of the American Society for Gastro-
intestinal Endoscopy, Lichtenstein DR, Jagannath S, Baron TH, et al.
Sedation and anestesia in GI endoscopy. Gastrointest Endosc 2008;
68:815-26.
3. Ulmer BJ, Hansen JJ, Overley CA, et al. Propofol versus midazolam/
fentanyl for outpatient colonoscopy: administration by nurses super-
vised by endoscopists. Clin Gastroenterol Hepatol 2003;1:425-32.
4. Rex DK, Deenadayalu VP, Eid E, et al. Endoscopist-directed administra-
tionofpropofol:aworldwidesafetyexperience.Gastroenterology2009;
137:1229-37; quiz 1518-9.
5. Vargo JJ, Bramley T, Meyer K, et al. Practice efficiency and economics:
thecaseforrapidrecoverysedationagentsforcolonoscopyinascreen-
ing population. J Clin Gastroenterol 2007;41:591-8.
6. FechnerJ,IhmsenH,HatterscheidD,etal.Comparativepharmacokinet-
icsandpharmacodynamicsofthenewpropofolprodrugGPI15715and
propofol emulsion. Anesthesiology 2004;10:626-39.
7. RexDK,KhalfanHK.Sedationandthetechnicalperformanceofcolonos-
copy. Gastrointest Endosc Clin N Am 2005;15:661-72.
8. RadaelliF,MeucciG,SgroiG,etal.ItalianAssociationofHospitalGastro-
enterologists (AIGO). Technical performance of colonoscopy: the key
role of sedation/analgesia and other quality indicators. Am J Gastroen-
terol 2008;103:1122-30.
9. ChurchJM.Warmwaterinfusionfordealingwithspasmduringcolonos-
copy: simple, inexpensive, and effective. Gastrointest Endosc 2002;56:
672-4.
10. Brocchi E, Pezzilli R, Tomassetti P, et al. Warm water or oil-assisted
colonoscopy: toward simpler examinations? Am J Gastroenterol 2008;
103:581-7.
11. Baumann UA. Water intubation of the sigmoid colon: water instillation
speeds up left-sided colonoscopy. Endoscopy 1999;31:314-7.
12. Hamamoto N, Nakanishi Y, Morimoto N, et al. A new water instillation
methodforcolonoscopywithoutsedationasperformedbyendoscopists-
in-training.GastrointestEndosc2002;56:825-8.
13. Leung FW. Water-related techniques for performance of colonoscopy.
Dig Dis Sci 2008;53:2847-50.
14. LeungJW,MannS,LeungFW.Optionsforscreeningcolonoscopywith-
outsedation:apilotstudyinUnitedStatesveterans.AlimentPharmacol
Ther 2007;26:627-31.
15. Leung FW, Aharonian HS, Leung JW, et al. Impact of a novel water
methodonscheduledunsedatedcolonoscopyinU.S.veterans.Gastro-
intest Endosc 2009;69:546-50.
16. LeungJW,SaleraR,ToomsenL,etal.Pilotfeasibilitystudyofthemethod
of water infusion without air insufflation in sedated colonoscopy. Dig
Dis Sci 2009;54:1997-2001.
17. LeungJW,MannSK,Siao-SaleraR,etal.Arandomized,controlledcom-
parisonofwarmwaterinfusioninlieuofairinsufflationversusairinsuf-
flationforaidingcolonoscopyinsertioninsedatedpatientsundergoing
colorectalcancerscreeningandsurveillance.GastrointestEndosc2009;
70:505-10.
18. Radaelli F, Meucci G, Imperiali G, et al. High-dose senna compared with
conventional PEG-ES lavage as bowel preparation for elective colonos-
copy: a prospective, randomized, investigator-blinded trial. Am J Gas-
troenterol 2005;100:2674-80.
19. AronchickCA,LipshutzWH,WrightSH,etal.Anoveltabletedpurgative
forcolonoscopicpreparation:efficacyandsafetycomparisonswithCo-
lyte and Fleet phospho-soda. Gastrointest Endosc 2000;52:346-52.
20. Revill SI, Robinson JO, Rosen M, et al. The reliability of a linear analogue
for evaluating pain. Anesthesia 1976;31:1191-8.
21. Morrow JB, Zuccaro G Jr, Conwell DL, et al. Sedation for colonoscopy
using a single bolus is safe, effective, and efficient: a prospective, ran-
domized, double-blind trial. Am J Gastroenterol 2000;95:2242-7.
22. Petrini JL, Egan JV, Hahn WV. Unsedated colonoscopy: patient charac-
teristics and satisfaction in a community-based endoscopy unit. Gas-
trointest Endosc 2009;69:567-72.
23. Ramirez F, Leung F, Hayden CT, et al. Water method for colonoscopy in
sedated patients is easy to learn for an experienced colonoscopist and
may minimize the need for technical assistance. Am J Gastroenterol
2009;104(Suppl 3):S525.
24. Ransibrahmanakul K, Leung J, Mann S, et al. Water infusion in lieu of air
insufflations (water method) diminished discomfort in patients under-
going minimal sedation colonoscopy in the hands of supervised train-
ees. Am J Gastroenterol 2009;104(Suppl 3):S509.
25. Morini S, Hassan C, Zullo A, et al. Detection of colonic polyps according
to insertion/withdrawal phases of colonoscopy. Int J Colorectal Dis
2009;24:527-30.
26. RexDK.Endoscopy:insertionversuswithdrawalphasesforpolypdetec-
tion. Nat Rev Gastroenterol Hepatol 2009;6:443-4.
Radaelli et alWarm water infusion for unsedated colonoscopy
www.giejournal.org Volume 72, No. 4 : 2010
GASTROINTESTINAL ENDOSCOPY
709