Keywords
Precancerous conditions & cancerous lesions (displasia and cancer) stomach - Barrett's
and adenocarcinoma - Preparation, quality and logistical aspects - Quality management
- Diagnosis and imaging (inc chromoendoscopy, NBI, iSCAN, FICE, CLE)
Introduction
Clear visualization of the gastrointestinal mucosal surface is essential for adequate
upper endoscopy. Previous research has shown that in patients with upper gastrointestinal
cancers, misdiagnosis during previous endoscopy is not uncommon, with missing rates
ranging between 6.0% to 9.4% [1]
[2]
[3]
[4]
[5]
[6]. Misdiagnosis can be partly attributed to lesions overlooked by endoscopists [6]
[7]. Optimal mucosal visibility, therefore, may not only shorten total procedure time,
as it reduces the need for washing and suctioning, but might also improve early detection
of small neoplastic lesions.
Standard preparation for upper endoscopy includes a fasting period [8]
[9], with or without the use of premedication. Premedication for upper endoscopy may
include antifoaming agents and mucolytic agents. While in most East Asian countries
administration of premedication is strongly recommended and has become standard of
care [10]
[11], it is not uniformly implemented in the United States and Europe.
Simethicone (which is a mixture of polydimethylsiloxane and silicon dioxide) is one
of the substances frequently used as premedication for upper endoscopy. In the majority
of previous studies, it has shown to improve mucosal visibility compared to no premedication
or water only [12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]. Simethicone has an antifoaming effect because it lowers the surface tension of
air bubbles, thereby causing the coalescence of small bubbles into larger ones. Several
dosages of simethicone have been studied over the years, and dosages ranging from
40 to 1000 mg all have shown to improve mucosal visibility [12]
[13]
[17]
[20]
[24]. Fewer data are available on the timing of simethicone as premedication for upper
endoscopy.
Therefore, the aim of the current study was to evaluate the effect of timing of simethicone
as premedication for upper endoscopy on mucosal visibility.
Patients and methods
Study design
This prospective, endoscopist-blinded, parallel-group randomized clinical trial with
a
superiority study design was performed in two different community hospitals in the
Netherlands between February 2020 and February 2022. All participants signed informed
consent prior to enrollment in the study. The medical ethics review boards of the
two
participating hospitals approved the trial, after which it was registered in the Dutch
Trial
Register (www.trialregister.nl; NL8383). The study was conducted in accordance with
the
principles of the Declaration of Helsinki and reported according to the Consolidated
Standards of Reporting Trials (CONSORT) guidelines for parallel-group randomized studies
[27].
Participants
Patients were considered eligible if they: (1) had reached an age of 18 years or older;
and (2) were scheduled for an elective upper endoscopy. Under the original protocol,
only patients undergoing endoscopy with sedation were eligible. During the trial,
the study team decided patients could be included regardless of the use and type of
sedation, as the use of sedation would be unlikely to influence our outcome measurements
(the trial protocol was amended accordingly and reviewed by the medical ethics review
board).
Exclusion criteria were as follows: (1) previous upper gastrointestinal surgery; (2)
achalasia; (3) known stenosis of the upper gastrointestinal tract;(4) known gastroparesis;
(5) allergies to simethicone; and (6) pregnancy.
Endoscopic procedures
Participants were randomly assigned to one of three parallel study groups after arrival
at the Endoscopy Department: the early group, late group, and split-dose group. The
early
group received 40 mg simethicone in 5 mL water on arrival at the endoscopy department
and 5
mL water in the endoscopy suite shortly before endoscopy; the late group received
5 mL water
on arrival at the endoscopy department and 40 mg simethicone in 5 mL water in the
endoscopy
suite shortly before endoscopy; and the split-dose group received 20 mg simethicone
in 5 mL
water on arrival at the endoscopy department and 20 mg simethicone in 5 mL water in
the
endoscopy suite shortly before endoscopy. Time of intake was reported for both drinks.
Of
note, the mucolytic agent N-acetylcysteine was not part of the premedication as this
is not
standard practice in our hospitals.
All endoscopies were performed by four experienced endoscopists (R.V., J.B., L.A.H.,
B.W.). Endoscopic images were captured at nine predefined locations before flushing:
the proximal esophagus, distal esophagus, corpus (antegrade position), antrum, angulus
(in partial inversion) corpus including greater curvature (retroflex position), cardia
(in inversion), duodenal bulb, and the descending part of the duodenum. Suctioning
of fluid in the stomach was allowed to avoid pulmonary aspiration. During the procedure,
the endoscopist scored his or her satisfaction for cleanness of the mucosa on a scale
from 1 to 10, with 1 being least satisfaction and 10 complete satisfaction.
Thereafter, the endoscopist was asked to start flushing with water until adequate
mucosal views were obtained in the duodenum, stomach, and esophagus. An endoscopic
flushing pump was used for this purpose. The number of flushes and the total flushing
time required to achieve adequate mucosal views were recorded. Total procedure time
and endoscopic findings were also recorded.
Mucosal visibility score
Two independent, experienced endoscopists (A.B. and L.A.H.), blinded to the timing
of
administration of simethicone, scored all still images using a 4-point scale previously
described by Basford et al ([Fig. 1]) [20]:
Fig. 1 Mucosal visibility score examples for one predefined location (corpus antegrade position).
a Score 1: no adherent mucus and clear views. b Score 2: a thin coating of mucus that does not obscure views. c Score 3: mucus/bubbles partially obscuring views. d Score 4: heavy mucus/bubbles obscuring views.
-
No adherent mucus and clear views of the mucosa.
-
A thin coating of mucus that did not obscure views of the mucosa.
-
Some mucus/bubbles partially obscuring views of the mucosa (i.e. a small mucosal lesion
might be missed without flushing).
-
Heavy mucus/bubbles obscuring views of the mucosa (i.e. extensive flushing is needed
to avoid missing small mucosal lesions).
The endoscopists did not receive formal training prior to their assessments. In case
of discrepancies between the two reviewers, the images were also scored by a third
independent endoscopist (A.A-T.), and the median of the three readings was used as
the final score.
Per individual patient, mucosal visibility was evaluated on three different levels:
1)
total mucosal visibility (TMV), defined as adequate if each score of the nine predefined
locations was ≤ 2 and considered inadequate if the score was ≥ 3 for one or more locations;
2) per organ (i.e. esophagus, stomach, duodenum), with mucosal visibility defined
as
adequate if all scores for the predefined locations in that specific organ were ≤
2 and
considered inadequate if the score was ≥ 3 for one or more locations in the concerning
organ; and 3) for each of the nine predefined locations, with mucosal visibility defined
as
adequate if the score for the still images of that particular location was ≤ 2. A
score ≥ 3
was considered inadequate.
Outcomes
The predefined primary outcome for this study was the percentage of patients with
adequate TMV. Predefined secondary outcomes were: 1) the percentage of patients with
adequate mucosal visibility for each organ separately; 2) the percentage of patients
with adequate mucosal visibility for each of the nine predefined locations; 3) the
satisfaction of the performing endoscopist concerning the cleanness of the mucosa,
as assessed during the procedure; 4) the total flushing time; 5) the number of fluid
flushes; and 6) the percentage of patients with newly detected dysplastic or neoplastic
lesions during endoscopy.
In post-hoc analyses, we also evaluated the percentage of patients with a mucosal
visibility score of 4 for at least one of the nine locations and the percentage of
patients with adequate mucosal visibility (i.e. TMV and mucosal visibility per organ
separately) in relation to the recorded time of simethicone administration before
endoscopy.
Sample size
Our sample size calculation was based on a pilot in our own patient population consisting
of 59 patients, assigned to the three study groups based on the day of the endoscopy
program: the early group, late group, and split-dose group. One blinded endoscopist
assessed nine endoscopic images in the esophagus, stomach, and duodenum on the mucosal
visibility scale described by Basford et al [20]. Adequate TMV in this pilot was 14% for the early group, and 26% and 67% for the
late and split-dose groups, respectively. Because in most Dutch hospitals the administration
of simethicone shortly before upper endoscopy is the current standard of care, we
considered the late group as the reference group. To demonstrate an increase of 20%
in adequate TMV in the other study groups, the calculated sample size for each group
was 119 patients (power of 80%, two-sided 1.7% significance level (5%/3) using Bonferroni
correction rather than Hommel’s method for multiplicity adjustment [28], assuming 10% dropout rate), resulting in 393 patients in total.
Randomization
After the nurse practitioner or research fellow obtained consent, patients were randomized
into one of three study groups on arrival at the endoscopy department on the day of
the procedure. We performed 1:1:1 block-randomization with block-sizes of six to create
three parallel groups, stratified by performing endoscopist. The randomization was
performed by a nurse practitioner or research fellow. The random allocation sequence
was generated using the randomization website https://www.sealedenvelope.com. We used the REDCap randomization module to guide the randomization process [29].
Blinding
The endoscopists performing the endoscopies and the independent endoscopists scoring
the images for TMV were blinded to the randomization process.
Statistical methods
Means with standard deviations were used for normally distributed variables, and medians
with 25th and 75th percentiles (p25-p75) for variables with a skewed distribution. Categorical variables
are presented as frequencies and percentages of total.
The percentages of adequate TMV in the study groups were compared using Chi-square
tests. Subanalyses per organ and for each predefined location were performed in the
same manner. Mean time of fluid flushing, and mean number of additional fluid flushes
required to achieve adequate mucosal views were compared using One-Way Anova or Kruskal
Wallis test, depending on the distribution. The percentage of newly diagnosed lesions
was compared using a Chi-square test, as well as the percentage of patients with a
mucosal visibility score of 4 in at least one of the nine locations. The percentage
of adequate mucosal visibility scores in relation to the recorded time of simethicone
administration was analyzed using a Chi-square test for trend in proportions. Statistical
significance was set at P < 0.05. In case of multiple comparisons, adjusted P values derived using Hommel’s method were derived and reported [30]. Therefore, P < 0.05 represents statistically significant findings.
As this was a clinical trial, missing data were limited to a very small number of
patients (3%) and observations (< 0.1%). Therefore, no imputation methods were used
to handle missing data. Observations with missing values are listed as frequencies
and percentages of total. R version 3.5.1 for Windows was used for all statistical
analyses.
Results
Patient characteristics
Of the 393 randomized patients, 132 were randomized in the early group, 130 in the
late group, and 131 in the split-dose group. Seven patients were excluded; two patients
were excluded because of not meeting the eligibility criteria and five patients due
to procedural failures ([Fig. 2]). Therefore, 386 patients were included in the final analysis. Patient baseline
characteristics are presented in [Table 1]. Most included patients were male (70%) and the mean age was 66 years. The main
indications for endoscopy were Barrett’s esophagus (BE) or follow-up after treatment
of BE neoplasia, which accounted for 68% of all endoscopies. Eighteen percent of patients
used medication that could affect gastric emptying, without significant differences
between study groups. While more than half of endoscopies were performed by the same
endoscopist, the number of endoscopies per study group was similar for each performing
endoscopist separately due to the stratified randomization ([Table 1]).
Fig. 2 Patient flow diagram.
Table 1 Baseline characteristics of the included patients and their endoscopies.
|
|
Total
(n = 386)
|
Early group
(n = 132)
|
Late group
(n = 128)
|
Split-dose group
(n = 126)
|
|
*Metoclopramide, ondansetron, erythromycin, domperidone, opioids, levodopa, tricyclic
antidepressants, beta-agonists, anticholinergics.
GI, gastrointestinal; NA, not applicable; p25-p75, 25th to 75th percentile; SD, standard deviation.
|
|
Baseline characteristics patients
|
|
Age, mean (SD)
|
66 (12)
|
66 (12)
|
67 (13)
|
66 (12)
|
|
Male, n (%)
|
269 (70)
|
92 (70)
|
91 (71)
|
86 (68)
|
|
ASA classification, n (%)
|
|
I
|
34 (9)
|
10 (8)
|
12 (9)
|
12 (10)
|
|
II
|
320 (83)
|
115 (87)
|
104 (81)
|
101 (80)
|
|
III
|
32 (8)
|
7 (5)
|
12 (9)
|
13 (10)
|
|
Indication endoscopy, n (%)
|
|
Barrett’s esophagus
|
127 (33)
|
46 (35)
|
40 (31)
|
41 (33)
|
|
Follow-up after treatment Barrett’s esophagus
|
137 (35)
|
42 (32)
|
49 (38)
|
46 (37)
|
|
Dyspepsia/epigastric pain
|
41 (11)
|
13 (10)
|
14 (11)
|
14 (11)
|
|
Gastroesophageal reflux
|
21 (5)
|
6 (5)
|
7 (5)
|
8 (6)
|
|
Anemia
|
12 (3)
|
5 (4)
|
5 (4)
|
2 (2)
|
|
Screening malignancy
|
12 (3)
|
5 (4)
|
3 (2)
|
4 (3)
|
|
Dysphagia
|
6 (2)
|
1 (0.8)
|
2 (2)
|
3 (2)
|
|
Suspected upper GI bleeding
|
1 (0.3)
|
0 (0)
|
0 (0)
|
1 (1)
|
|
Suspected celiac disease
|
1 (0.3)
|
0 (0)
|
0 (0)
|
1 (1)
|
|
Other
|
28 (7)
|
14 (11)
|
8 (6)
|
6 (5)
|
|
Diabetes mellitus, n (%)
|
53 (14)
|
22 (17)
|
19 (15)
|
12 (10)
|
|
Medication affecting gastric emptying*, n (%)
|
68 (18)
|
17 (13)
|
28 (22)
|
23 (18)
|
|
Baseline characteristics endoscopy
|
|
Performing endoscopist
|
|
Endoscopist 1
|
211 (55)
|
70 (53)
|
70 (55)
|
71 (56)
|
|
Endoscopist 2
|
84 (22)
|
28 (21)
|
28 (22)
|
28 (22)
|
|
Endoscopist 3
|
55 (14)
|
19 (14)
|
18 (14)
|
18 (14)
|
|
Endoscopist 4
|
43 (11)
|
15 (11)
|
14 (11)
|
14 (11)
|
|
Sedation, n (%)
|
|
Midazolam/fentanyl
|
239 (62)
|
80 (61)
|
82 (64)
|
77 (61)
|
|
Propofol
|
79 (20)
|
27 (20)
|
27 (21)
|
25 (20)
|
|
No sedation
|
68 (18)
|
25 (19)
|
19 (15)
|
24 (19)
|
|
Timing simethicone, minutes, median (p25-p75)
|
|
First simethicone drink
|
|
25 (19–33)
|
NA
|
25 (19–33)
|
|
Second simethicone drink
|
|
NA
|
8 (6–10)
|
8 (6–9)
|
|
Duration endoscopy, minutes, median (p25-p75)
|
8 (5–13)
|
8 (5–12)
|
9 (6–14)
|
9 (5–14)
|
Mucosal visibility scores
In 183 of the 386 included patients (47%), the TMV was considered adequate ([Table 2]). TMV was assessed as adequate in 72 (55%), 54 (42%), and 77 patients (61%) in the
early, late, and split-dose groups, respectively (P < 0.01). The percentage of adequate TMV in the late group was significantly lower
compared to the split-dose group (42% vs 61%, P < 0.01), but not for the late group compared to the early group (42% vs 55%, P = 0.09) ([Fig. 3]).
Fig. 3 Percentages adequate total mucosal visibility (TMV) and adequate mucosal visibility
per organ. Adequate TMV was defined as a score ≤ 2 for all nine of the predefined
locations. Adequate mucosal visibility per organ was defined as a score ≤ 2 for each
of the respective predefined locations in the esophagus, stomach or duodenum separately.
Early group = 40 mg simethicone 20 to 30 minutes prior to gastroscopy. Late group
= 40 mg simethicone 0 to 10 minutes prior to gastroscopy. Split-dose group = 20 mg
simethicone 20 to 30 minutes, and 20 mg simethicone 0 to 10 minutes prior to gastroscopy.
*** denotes statistical significance.
Table 2 Primary and secondary outcomes.
|
|
Total
(n = 386)
|
Early group
(n = 132)
|
Late group
(n = 128)
|
Split-dose group
(n = 126)
|
P value
|
|
*Newly detected lesions in esophagus stomach, duodenum after flushing.
†Adjusted P values using Hommel’s correction method.
TMV, total mucosal visibility; p25-p75, 25th to 75th percentile; SD, standard deviation.
|
|
Primary outcome
|
|
Adequate TMV, n (%)
|
183 (47)
|
72 (55)
|
54 (42)
|
77 (61)
|
< 0.01
|
Early vs. late = 0.09†
Early vs. split-dose = 0.29†
Late vs. split-dose < 0.01†
|
|
Secondary outcomes
|
|
Adequate mucosal visibility esophagus, n (%)
|
314 (81)
|
101 (77)
|
110 (86)
|
103 (82)
|
0.15
|
|
Adequate mucosal visibility stomach, n (%)
|
245 (63)
|
90 (68)
|
68 (53)
|
87 (69)
|
0.01
|
Early vs. Late = 0.03†
Early vs. split-dose = 0.88†
Late vs. split-dose = 0.02†
|
|
Adequate mucosal visibility duodenum, n (%)
|
362 (94)
|
130 (98)
|
112 (88)
|
120 (95)
|
< 0.01
|
Early vs. late < 0.01†
Early vs. split-dose = 0.13†
Late vs. split-dose = 0.06†
|
|
Satisfaction performing endoscopist, median (SD)
|
7.0 (6.0–8.0)
2 (0.5%) missing
|
7.0 (6.5–8.0)
1 (0.8%) missing
|
7.0 (6.0–8.0)
1 (0.8%) missing
|
8.0 (6.0–8.0)
|
0.62
|
|
Flushing time, seconds, median (p25-p75)
|
39 (20–60)
3 (0.8%) missing
|
38 (20–58)
|
40 (22–62)
3 (2%) missing
|
39 (17–60)
|
0.73
|
|
Number of additional flushes, median (p25-p75)
|
3 (2–5)
3 (0.8%) missing
|
3 (2–5)
|
4 (3–5)
3 (2%) missing
|
3 (2–5)
|
0.36
|
|
Detected dysplastic lesions*, n (%)
|
41 (11)
|
16 (12)
|
12 (9)
|
13 (10)
|
0.77
|
Differences in adequate mucosal visibility per organ were most pronounced in the stomach
([Table 2], [Fig. 3]). Both the early and split-dose groups had higher percentages of adequate mucosal
visibility in the stomach compared to the late group (68% vs 53%, P = 0.03 and 69% vs 53%, P = 0.02). As for the duodenum, only the early group scored higher in adequate mucosal
visibility than the late group (98% vs 88%, P < 0.01), whereas this was not the case for scores in the split-dose group compared
to the late group (95% vs 88%, P = 0.06). There were no significant differences among groups regarding adequate mucosal
visibility of the esophagus. The distributions of mucosal visibility scores for each
of the nine predefined locations separately are shown in Supplementary Table S1.
A mucosal visibility score of 4 (i.e. heavy mucus/bubbles obscuring views of the mucosa
for which extensive flushing is needed) was given for at least one location in the
esophagus, stomach and duodenum in 50 individual patients (13%). In the late group,
24 patients (19%) received a score of 4 for at least one of the predefined locations,
compared to 15 (11%) in the early group and 11 (9%) in the split-dose group (P = 0.047).
Endoscopic parameters
Median satisfaction scores for the performing endoscopists concerning the cleanness
of the mucosa during the endoscopy were 8.0 (p25-p75 6.0–8.0) for the split-dose group,
as opposed to a median score 7.0 for both the early (p25-p75 6.5–8.0) and late groups
(p25-p75 6.0–8.0) (P = 0.62).
In the total cohort, a median of 39 seconds of flushing was necessary in a median
of three flushes to achieve clear mucosal views in the esophagus, stomach, and duodenum,
without significant differences between study groups. In addition, there were no significant
differences in the percentage of newly detected (pre)cancerous lesions in the esophagus,
stomach, or duodenum (P = 0.77).
Simethicone administration time
[Fig. 4] demonstrates the percentages of adequate mucosal visibility for patients randomized
to the early and late groups only (n = 260) in relation to the recorded time of simethicone
administration before endoscopy. The percentage of adequate TMV was highest when simethicone
was administered between 20 to 30 minutes prior to upper endoscopy. For the esophagus,
the highest percentage of adequate mucosal visibility was seen when a short time interval
was used between simethicone intake and upper endoscopy (0–10 minutes), with decreasing
percentages when the time interval increased (P = 0.02 for trend). For both the stomach and duodenum, percentages of adequate mucosal
visibility increased with an increasing time interval between simethicone administration
and the start of the procedure (P < 0.01 for trend).
Fig. 4 Adequate mucosal visibility (%) in all patients (n = 260) randomized to either the
early group (n = 132) or the late group (n = 128) in relation to the measured time
(minutes) of simethicone administration prior to endoscopy. Early group = 40 mg simethicone
scheduled at 20 to 30 minutes prior to gastroscopy. Late group = 40 mg simethicone
scheduled at 0 to 10 minutes prior to gastroscopy. Of note: patients included in the
split-dose group are not represented in this figure. Adequate total mucosal visibility
was defined as a score ≤ 2 for all nine of the predefined locations. Adequate mucosal
visibility per organ was defined as a score ≤ 2 for each of the respective predefined
locations in the esophagus, stomach or duodenum separately.
Discussion
The presence of mucus and foam in the esophagus, stomach, and duodenum can hamper
thorough assessment of the mucosal surface. An adequate pre-procedure preparation
may help to improve mucosal visibility. In this multicenter, single-blind, randomized
trial, we aimed to clarify the effect of timing of simethicone administration as premedication
for upper endoscopy in relation to mucosal visibility. We compared three different
simethicone intake regimens: early intake (20–30 minutes prior to endoscopy), late
intake (0–10 minutes prior to endoscopy), and split-dose intake (20–30 minutes and
0–10 minutes prior to endoscopy). Our study demonstrated that overall, clearest mucosal
views were obtained when simethicone was administered early (i.e. 20–30 minutes before
endoscopy), either in a split-dose regimen or as a single dose. Although the highest
percentage of adequate TMV was found in the split-dose group, the difference between
the split-dose and early groups failed to reach statistical significance.
Current guidelines provide recommendations on fasting for solids and fluids, although
the routine use of premedication prior to upper endoscopy is, unlike bowel preparation
for colonoscopy [31]
[32], not uniformly advised. Without specific guideline recommendations, the use of premedication
varies considerably in daily practice. A relatively low dosage of simethicone was
administered in a 5-mL solution in our study, according to the standard of care in
our practices. Many studies have been published on the effects of simethicone as premedication
for upper endoscopy. Irrespective of the dosage used, the vast majority of studies
demonstrated clear beneficial effects of simethicone on improving mucosal visibility
compared to placebo, including a reduction in total flushing and total procedure times
[14]
[15]
[16]
[17]
[19]
[21]
[23].
Fewer data are available on the optimal administration time for simethicone. Two studies
have been published on the timing of premedication; however, assessing gastric mucosal
visibility only. Sun et al. studied different time intervals for simethicone intake
and concluded that a time interval of 31 to 60 minutes resulted in the best gastric
visibility [33]. Woo et al. investigated a premedication mixture of pronase, dimethylpolysiloxane,
and sodium bicarbonate, and found that the optimal time interval was between 10 to
30 minutes before upper endoscopy [34]. To our knowledge, our trial is the first primarily designed to evaluate the effect
of timing of simethicone intake on the complete upper gastrointestinal tract.
Subsequently, we found that the effect of timing of simethicone intake is different
for the esophagus, stomach, and duodenum. In the esophagus, decreasing percentages
of adequate mucosal visibility with increasing time between simethicone intake and
the start of the endoscopy were observed. On the contrary, for the stomach and duodenum,
the percentage of adequate mucosal visibility scores increased with increasing time
of intake. The highest percentage of adequate TMV in the split-dose group, therefore,
is a logical consequence deriving from these findings, demonstrating the results of
a balanced approach to maximize the effect in the esophagus, stomach and duodenum
altogether.
How should our findings influence clinical practice? We demonstrated that impaired
mucosal visualization occurred frequently in the stomach. Extensive cleansing of the
gastric mucosa is particularly challenging and time-consuming as compared to the esophagus
or duodenum, and longer procedure time can cause more discomfort in patients. At the
same time, lesions in the stomach can be subtle and easy to overlook if they are covered
by bubbles [3]
[6]. For practical reasons, simethicone administration strategies, therefore, may mainly
focus on improving gastric mucosal visibility, with standard simethicone administration
20 to 30 minutes prior to upper endoscopy. In case of known or suspected esophageal
pathology (e.g. if the endoscopy is performed in the context of BE surveillance),
a second dose of simethicone administered 0 to 10 minutes prior to the procedure may
be considered.
The main strengths of our study are its randomized, endoscopist-blinded design and
the high-quality data collection. To reduce the level of subjectivity, mucosal visibility
scores were derived from still images, assessed by three independent, blinded endoscopists.
Lastly, the complete esophagus, stomach, and duodenum were explored and assessed regarding
mucosal visibility, using nine standard photo documentation locations mentioned by
the European Society of Gastrointestinal Endoscopy guideline on performance measures
for upper gastrointestinal endoscopy [9].
Nonetheless, several limitations should be taken into account while interpreting the
results of this study. First, when our study was conducted, no validated scoring tool
was available to evaluate mucosal visibility during upper endoscopy. Many different
scores have been used in previous studies. We chose to use the scoring system previously
described by Basford et al [20], based on the possibility of missing small mucosal lesions rather than required
flushing volume. In our opinion, this is a more clinically relevant scale. However,
in a recent study, a standardized scoring tool was found to have strong evidence of
validity for the assessment of gastric and duodenal mucosal visibility [35]. Second, because two of the participating endoscopists were BE expert endoscopists
in a tertiary referral hospital, the majority of study participants were BE patients.
This will account for the high percentage of newly diagnosed dysplastic lesions in
this study. Apart from this, we believe the results of our study can be generalized
to all patients undergoing elective upper endoscopy. Third, we did not include a study
arm without premedication, nor did we study the effect of timing of simethicone administration
at different dosages. Considering that the beneficial effects of simethicone compared
to placebo have been demonstrated in previous studies, it was decided to focus solely
on simethicone administration time. As for dosage, low-dose simethicone was used in
this study because it is routinely used in our practices. However, none of the study
arms showed adequate TMV above 70%. The TMV might have been improved if we had used
higher simethicone dosages. Fourth, this study was not designed nor powered to detect
differences in dysplastic lesions during upper endoscopy. Dysplastic lesions were
recorded after all study procedures were completed, i.e. when the performing endoscopists
had obtained adequate mucosal views through water flushing. Evaluating the effect
of different simethicone administration times on detection of dysplastic lesions in
the upper gastrointestinal tract would require an extremely large sample size. Therefore,
we used mucosal visibility as a surrogate endpoint. Finally, we did not evaluate the
role of mucolytic agent N-acetylcysteine in addition to simethicone, because in most
Western countries this is not typically used.
Conclusions
In conclusion, this randomized trial demonstrated that timing of small-volume simethicone
as premedication in upper endoscopy is of particular importance for mucosal visibility.
To optimize mucosal visibility in daily practice, simethicone can best be administered
20 to 30 minutes prior to upper endoscopy, either as a single administration or in
a split-dose regimen.
