Dig Dis Sci (2012) 57:2527–2534

DOI 10.1007/s10620-012-2188-2

ORIGINAL ARTICLE

What Factors Are Associated with the Difficult-to-Sedate

Endoscopy Patient?
Bikram S. Bal • Michael D. Crowell •
Divyanshoo R. Kohli • Jiana Menendez •
Farzin Rashti • Anjali S. Kumar • Kevin W. Olden

Received: 9 November 2011 / Accepted: 14 April 2012 / Published online: 8 May 2012
Ó Springer Science+Business Media, LLC 2012

Abstract Conscious sedation was administered for the endoscopic

Background Difficult sedation during endoscopy results
in inadequate examinations and aborted procedures. We
hypothesized that gender, alcohol abuse, physical/sexual
abuse, and anxiety are predictors of difficult-to-sedate
endoscopy patients.
Methods This is a prospective cohort study. At the time
of enrollment, subjects completed the following three
validated questionnaires: state-trait anxiety inventory,
self-report version of alcohol use disorder inventory,
and Drossman questionnaire for physical/sexual abuse.
Kevin W. Olden serves as the guarantor and assumes full
responsibility for this study.
B. S. Bal (&)
D. R. Kohli
F. Rashti
Division of Gastroenterology, Department of Internal Medicine,
Washington Hospital Center, 110 Irving St NW, Suite 3A-3,
Washington, DC 20010, USA
e-mail: bal.bikram@gmail.com
M. D. Crowell
Division of Gastroenterology and Hepatology, Mayo Clinic,
Scottsdale, AZ, USA
J. Menendez
Mount Sinai Medical School, New York, NY, USA
procedures at the discretion of the endoscopist and was
graded in accordance with the Richmond agitation sedation
scale (RASS). Subjects’ perceptions of sedation were
documented on a four-point Likert scale 24 h after their
procedure.
Results One-hundred and forty-three (79 %) of the 180
subjects enrolled completed the study. On the basis of the
RASS score, 56 (39 %) subjects were found to be difficult
to sedate of which only five were dissatisfied with their
sedation experience. State (n = 39; p = 0.003) and trait
(n = 41; p = 0.008) anxiety and chronic psychotropic use
(p = 0.040) were associated with difficult sedation. No
association was found between difficult sedation and
gender (p = 0.77), alcohol abuse (p = 0.11), sexual abuse
(p = 0.15), physical abuse (p = 0.72), chronic opioid use
(p = 0.16), or benzodiazepines (BDZ) use (p = 0.10).
Conclusion Pre-procedural state or trait anxiety is asso-
ciated with difficult sedation during endoscopy. In this
study neither alcohol abuse nor chronic opiate/BDZ use
was associated with difficult sedation.
Keywords Endoscopy
Difficult to sedate
Abuse

Anxiety

A. S. Kumar
Department of Surgery, Washington Hospital Center,
110 Irving St NW, P.O. Box 2100, Washington, DC 20010, USA
K. W. Olden
Department of Medicine, St Joseph’s Hospital and Medical
Center, Phoenix, AZ 85013, USA
K. W. Olden
Department of Medicine and Psychiatry, Creighton University
School of Medicine, California Plaza, 2500, Omaha,
NE 68178, USA
Introduction
Adequate sedation is an integral part of a good-quality
endoscopic examination. Sedation for endoscopy reduces
patient anxiety and pain. It increases the acceptability of
procedures to the patients and results in greater willingness
to undergo repeat procedures [1, 2]. The difficult-to-sedate
endoscopy patient has been a long-standing challenge in
the practice of gastrointestinal endoscopy. Inadequate

123
2528 Dig Dis Sci (2012) 57:2527–2534

sedation results in patient discomfort and incomplete abuse, physical/sexual abuse, and difficult sedation during
examinations. The inconvenience of re-prepping because endoscopic examination.
of a failed endoscopic examination and the time lost from
work or school can be a significant burden to patients.
Additionally, failed endoscopic procedures also result in Methods
financial loss and waste of scarce medical resources.
The phenomenon of inability to adequately sedate a Design and Setting
patient for an endoscopic examination is poorly understood
and few studies have addressed this pertinent issue. Vari- This was a prospective cohort study carried out solely at
ous suggested predictors of difficult sedation perform the Washington Hospital Center, a 920 bed tertiary-care
inconsistently and usually have non-validated tools to hospital in Washington, DC, USA. Subjects were enrolled
support their cogency. Excessive alcohol use ([40 g/day), from the gastroenterology and colorectal surgery outpatient
chronic benzodiazepine use, and ‘‘longer procedures’’ are clinics from February 15, 2010 to November 15, 2010. The
thought to portend difficult sedation, on the basis of pre- number of subjects recruited was based on the sample-size
vious studies [3–5]. Despite the implication of alcohol calculation described below. All endoscopic procedures
abuse in interfering with successful sedation, it is com- were completed at the hospital endoscopy center. The study
monly observed that patients with active alcohol use and/or was approved by the Institution Review Board and all
use of other drugs cross-tolerant of those used for con- participants signed informed consent. Subjects were
scious sedation are not routinely sedation challenges. deemed enrolled in the study from the time of recruitment
Anecdotally, physical and sexual abuse is thought to be until the follow-up call 24 h after their endoscopic
associated with difficult sedation but hitherto no studies procedure.
have investigated this association.
The best study to date to investigate this issue was pub-
Subject Recruitment
lished by Pena et al. [4]. This study reported pre-procedure
‘‘nervousness’’ scored on a ten-point Likert scale as a pre-
Subjects were recruited for the study at the time of
dictor of difficult sedation. Investigators utilized the Spiel-
scheduling their medically indicated endoscopic procedure;
berger state-trait anxiety index (STAI) to objectively
an esophagogastroduodenoscopy (EGD) or colonoscopy.
document anxiety, but these data were not reported in their
Informed consent was obtained by the physician scheduling
article. In that same study, alcohol abuse was also found to be
them for their procedure. To eliminate any perceived dur-
associated with difficult sedation, however no validated
ess no subjects were enrolled in the study on the day of
instrument was used to document alcohol abuse. The primary
their procedure. Subjects undergoing endoscopy via our
outcome measured was patient satisfaction scored on a seven-
‘‘open access’’ system were scheduled by the office staff
point Likert scale, which was subject to multiple biases.
and were not enrolled in the study because they did not
Mahajan and colleagues also attempted to measure the
have the opportunity to discuss the purposes and intent of
anxiety related to GI procedures but utilized non stan-
the study in advance of their procedure. Subjects under-
dardized anxiety measures [6]. In their study, anxiety was
going ERCP, EUS, and luminal stenting were also exclu-
measured on a four-point Likert scale and visual linear
ded from the study, because these subjects require
analog scale in which the patient was asked to place a
aggressive sedation owing to the complex nature of these
mark, representing anxiety, on a 10-cm long line, that was
procedures. Exclusion criteria for our study also included
scored from 0 to 10, with 0 representing no anxiety and 10
alcohol withdrawal within 30 days before enrollment and
being maximum anxiety. Again both of these were subject
the inability to give informed consent because of dementia
to reporting bias and could not be interpreted in a stan-
or developmental disabilities.
dardized fashion. The hospital anxiety and depression scale
At the time of enrollment; age, gender, and current
(HADS) was used to objectively diagnose anxiety. How-
medication use were recorded. Each subject was assigned a
ever, the HADS is more accurately used in hospitalized
unique identifying number. Subject identifiers such as
patients and their study was limited to outpatients only [7].
name and medical record number were kept on a password
Identifying factors affecting difficult sedation during
protected computer and not displayed on any subsequent
endoscopy have been an elusive pursuit for gastroenterol-
forms. At enrollment, subjects completed the following
ogists. Multiple factors have been anecdotally associated
three questionnaires:
with difficult sedation; however limited research and effort
have been used to describe these factors. 1. Alcohol use disorder inventory (AUDIT)—self report
The purpose of our study was to use validated instru- version The AUDIT was created by the World Health
ments to study the association between anxiety, alcohol Organization as a simple method of screening for

123
Dig Dis Sci (2012) 57:2527–2534
alcohol abuse [8]. The AUDIT has been developed and
validated over a period of two decades, and has been
found to enable accurate assessment of alcohol abuse
across gender, age, and cultures [9]. The self report
version of the AUDIT consists of 10 questions; each
question with a set of responses scored 0–4, by the
subject, on the basis of the subject’s alcohol use. It
usually takes 2–4 min to complete the AUDIT self-
report questionnaire. A score of 10 or greater is
regarded as highly specific for diagnosis of alcohol
abuse [10]. This score was used to diagnose alcohol
abuse in our study.
2. STAI The STAI has been used extensively in research
and clinical practice for diagnosis of state and trait
anxiety. It has been translated into 30 languages, and
since its development in 1970 more than 2,000 studies
using the STAI have appeared in the research literature
[11]. For the purpose of our study the revised version
of the test; ‘‘Form-Y’’ was used. The STAI-Y is a self
administered test and takes 6–15 min to complete,
depending on the subject’s level of education. The
S-anxiety scale (STAI Form Y-1) consists of 20
statements that evaluate how the respondent feels
‘‘right now, at this moment’’. The T-anxiety scale
(STAI Form Y-2) consists of 20 statements that
evaluate how individuals ‘‘generally feel’’. A score
of over 40 on the STAI Form Y-1 and STAI Form Y-2
was taken to be diagnostic of state of anxiety and trait
anxiety respectively. Complete details regarding scor-
ing and norms for STAI have been published else-
where [11].
3. Screening questionnaire for sexual and physical
abuse history This is a self-report questionnaire
developed by Drossman et al. Since its design, the
questionnaire has been validated by their group
against a detailed psychological interview [12–15].
The questionnaire has two parts: the first part has five
questions to identify sexual abuse as a child or adult;
the second part has two questions to identify physical
abuse as a child or adult. A copy of the questionnaire
can be obtained from the review article published by
Drossman et al. [28]. Any positive answer in either
section is regarded as indicative of sexual or physical
abuse, respectively. For the purpose of our study, no
distinction was made between abuse as an adult or
child.
Because all three questionnaires inquired about infor-
mation which may have been potentially distressing to the
subjects, one of the authors (KO), who is a board-certified,
psychiatrist was available to the subjects in case emotional
and mental distress was caused or detected by the protocol-
related questions.
2529
Endoscopy
Subjects were checked in and prepared according to the
routine protocols of our endoscopy center on the day of
their procedure. Procedures were performed in the standard
fashion by our endoscopy team. Each endoscopy team
consisted of an attending gastroenterologist, gastroenter-
ology fellow, registered nurse, and endoscopy technician.
Four different gastroenterologists, unaware of the results of
the questionnaires and with experience ranging from 20 to
35 years, performed the EGDs and colonoscopies. Stan-
dard medications, documentation of findings, and other
medical care were administered or performed according to
our endoscopy unit rules, policies, and protocols. Con-
scious sedation in our unit is delivered by the endoscopy
nurse under the direct supervision of the endoscopist. We
use a combination of intravenous midazolam and fentanyl
or meperidine as the standard sedative. Rarely, diphenhy-
dramine was given for additional sedation, on the basis of
the judgment of the supervising physician.
Sedation was administered at the discretion of the gas-
troenterologist performing the procedure. The four differ-
ent endoscopists used escalating doses of benzodiazepines
(midazolam) in combination with opiates (fentanyl or
meperidine) in accordance with their preference and
experience to keep subjects sedated for an adequate
endoscopic examination. The dosage of intravenous seda-
tion medication was recorded but not used to stratify sub-
jects as being difficult or easy to sedate. We did not record
the time required for the procedure to be completed. After
endoscopy, the difficulty of sedation was recorded objec-
tively on the Richmond agitation sedation scale (RASS).
The RASS is a ten-level (?4 ‘‘combative’’ to -5
‘‘unarousable’’) scale for assessment of sedation and agi-
tation of subjects undergoing conscious sedation (Table 1)
[16]. It is designed to have precise, unambiguous defini-
tions for levels of sedation that rely on an assessment of
arousal, cognition, and sustainability using common
responses (eye opening, eye contact, physical movement)
to common stimuli (spoken voice, physical stimulation)
presented in a logical progression. Since its inception, the
RASS has been tested extensively, including by user
groups representing nurses, physicians, and pharmacists, in
research and clinical settings and for a variety of subject
populations [17–21]. It is a well validated tool used for
assessing conscious sedation of ventilated and non-venti-
lated patients. For the purpose of our study, a score of -1
(sustained awakening to voice—eye opening/eye contact
[10 s or greater) or greater was classified as difficult
sedation by the attending gastroenterologist [16].
In addition, the nurse, technician, gastroenterology
fellow, and attending gastroenterologist, unaware of each
123
2530 Dig Dis Sci (2012) 57:2527–2534

Table 1 Richmond agitation

RASS score RASS description
sedation scale (RASS)
?4 Combative, violent, danger to staff
?3 Pulls or removes tube(s) or catheters; aggressive
?2 Frequent non-purposeful movement, fights ventilator
?1 Anxious, apprehensive, but not aggressive
0 Alert and calm
-1 Awakens to voice (eye opening/contact) [10 s
-2 Light sedation, briefly awakens to voice (eye opening/contact) \10 s
-3 Moderate sedation, movement or eye opening. No eye contact
-4 Deep sedation, no response to voice, but movement or eye opening to physical stimulation
-5 Unarousable, no response to voice or physical stimulation

Table 2 Likert scale to assess difficulty of sedation Statistical Analysis

1. Easy to sedate
2. Somewhat easy to sedate
3. Somewhat difficult to sedate
4. Very difficult to sedate
5. Unable to sedate. Procedure aborted
Table 3 Likert scale to assess satisfaction with sedation experience
1. Very satisfied and does not remember anything
2. Satisfied but remembers some discomfort
3. Dissatisfied and was uncomfortable during the procedure
4. Very dissatisfied and experienced significant pain and
discomfort during the procedure
others’ responses, individually rated the difficulty of
sedation on a five-point Likert scale (Table 2). This
information was used to compare the perception of difficult
sedation between different members of the endoscopy
team, and was correlated with the RASS score. Scores of 1
and 2 were classified as easy sedation, and scores of 3, 4,
and 5 were classified as difficult sedation.
Follow-Up
Twenty-four hours post-procedure, a follow-up call was
made to the subject to record self-reported sedation expe-
rience. The sedation experience was rated on a four-point
Likert scale (Table 3). Selections 1 and 2 were classified as
satisfied with sedation experience. Selections 3 and 4 were
classified as dissatisfied with sedation experience. If the
subject could not be reached, a message to call back was
left and subjects were called repeatedly until they could be
contacted. If the subject could not be contacted after a
period of 1 month no further attempts were made to doc-
ument their sedation experience.
Data were entered manually and statistically assessed by
use of the Statistical Package for the Social Sciences
(SPSS version 18.0; SPSS, Chicago, IL, USA). Normality
of data sets was determined by use of SPSS explore and
descriptive functions. Stem-and-leaf plots and histograms
were used to evaluate variable distributions and assess
outliers. Normality plots were used to display normal
probability and detrended normal probability plots. Where
appropriate, the Kolmogorov–Smirnov statistic was used
for testing normality. Frequency distributions were eval-
uated for all categorical variables (e.g., gender). Student’s
t tests were performed to evaluate differences in demo-
graphic variables with continuous, normal distributions,
for example weight, height, or BMI, between groups.
Unadjusted tests for proportionality between groups were
made using v2 tests with the Mantel–Haenszel odds ratios
used to test for independence between factors of interest.
Adjusted multivariate logistic regression was used to
evaluate the strength of association between each pre-
dictor and the criterion variables after adjusting for gen-
der. The inter-rater reproducibility for classifying the
perception of difficult sedation among members of the
endoscopy team was analyzed by use of Cohen’s kappa
statistic. These statistical coefficients are commonly used
to analyze data sets for strength of agreement between
raters. The kappa statistic (j) is a measure of agreement
over and above chance that is used as a measure of
reproducibility between repeated assessments of the same
variable by different raters. Fleiss’ guidelines were fol-
lowed to characterize j over 0.75 as excellent, 0.40–0.75
as fair to good, and below 0.40 as poor agreement
between raters [22]. Summary statistics included point
estimates and standard deviations or 95 % confidence
intervals for all variables. All significance levels were set
to p \ 0.05.

123
Dig Dis Sci (2012) 57:2527–2534 2531

Sample-Size Calculations Table 5 Factors associated with difficult sedation

Characteristics No difficulty in Difficult p value
A primary objective of the proposed study was to test the sedation n (%) sedation n (%)
hypothesis that difficult sedation was associated with state/
trait anxiety, sexual/physical abuse, or alcohol abuse. A Female gender 57 (66 %) 38 (68 %) 0.77
secondary objective was to evaluate the relationship State anxiety 16 (22 %) 23 (48 %) 0.003
between sedation and concomitant medication use. The Trait anxiety 18 (25 %) 23 (48 %) 0.008
null hypothesis was that the proportion of difficult-to- Alcohol abuse 10 (12 %) 12 (21 %) 0.11
sedate subjects would not differ significantly from subjects Sexual abuse 30 (35 %) 26 (46 %) 0.15
with adequate sedation with regard to gender, state and trait Physical abuse 43 (49 %) 26 (46 %) 0.72
anxiety, alcohol abuse, or physical/sexual abuse. Power Medications
was computed on the basis of primary analysis to reject the Opiates 4 (5 %) 6 (11 %) 0.16
null hypothesis if state or trait anxiety resulted in a mini- Benzodiazepines 4 (5 %) 0 (0 %) –
mally clinically relevant 25 % difference in the incidence Psychotropics 4 (5 %) 8 (14 %) 0.04
of difficult sedation between the groups. The criterion for Bold values indicate statistically significant
significance (a) was set at 0.05, two-tailed. Given these
values, a study sample of 58 subjects distributed equally
would provide at least 80 % power to reject the null
hypothesis. Table 5 shows the association between different vari-
ables and difficult sedation during endoscopy. Increased
state and trait anxiety were significantly associated with
Results increased difficulty of sedation. Elevated state anxiety
scores were identified in 39/121 (32 %) subjects, 23 of
A total of 180 subjects were enrolled in the study. One- whom were difficult to sedate (p = 0.003). Complete STAI
hundred and forty-three subjects—48 males and 95 scores could not be obtained for 22 subjects. After con-
females-completed their endoscopic procedure and were trolling for gender, difficult-to-sedate subjects were three
deemed to have completed the study. Thirty-seven subjects times as likely to have increased state anxiety (OR 3.29,
did not attend for their scheduled procedure and did not 95 % CI 1.49, 7.26; p = 0.003). Trait anxiety was found in
make any attempt to re-schedule their procedures. These 41/121 (34 %) of subjects, 23 (48 %) of whom were dif-
cases were classified as lost to follow-up and were not ficult to sedate (p = 0.008). Of those subjects with ele-
included in the study for the purpose of data analysis. vated state anxiety, 72 % also had elevated trait anxiety
Twenty percent cancellation or ‘‘no show’’ is average for scores. After controlling for gender, difficult-to-sedate
our practice and not thought to affect or be affected by the subjects were also nearly three times as likely to have
study. increased trait anxiety compared with subjects with normal
On the basis of the RASS score (-1 or greater), 56 sedation (OR 2.92, 95 % CI 1.32, 6.46; p = 0.008). These
subjects (39 %) were found to be difficult to sedate (18 findings suggest that both state and trait anxiety are sig-
males and 38 females). Subject demographics and sedation nificant contributors to difficult sedation during endoscopy.
difficulty are summarized in Table 4. No statistically sig- No significant relationship was found between sexual
nificant association was found between gender and difficult abuse (p = 0.15) and physical abuse (p = 0.72), as judged
sedation (p = 0.77). by the Drossman abuse questionnaire, and difficulty with
sedation. Interestingly, sexual abuse was not significantly
associated with state-anxiety (p = 0.12) or trait-anxiety
Table 4 Patient characteristics
(p = 0.22). Similar outcomes were found between physical
Characteristic n abuse state-anxiety (p = 0.11) or trait-anxiety (p = 0.45).
Total number of patients 180
Even after controlling for gender, difficult-to-sedate sub-
jects were no different from subjects with normal sedation
Completed study 143 (79 %)
for sexual abuse (OR 1.64, 95 % CI 0.82, 3.30; p = 0.16)
Males 48 (33.5 %)
or physical abuse (OR 0.89, 95 % CI 0.46, 1.76; p = 0.74).
Females 95 (66.5 %)
Twenty-two subjects (15 %) met criteria for alcohol
Mean age (years) 51.5
abuse, on the basis of their AUDIT score. The proportion
Difficult to sedate 56 (39 %)
of subjects that was difficult to sedate was not statistically
Males 18 (32 %)
different from those with normal sedation (p = 0.11).
Females 38 (68 %)
There was no association between alcohol use and

123
2532
Table 6 Patients’ sedation experience
Characteristics n
Total patients 180
Completed study 143 (79 %)
Satisfied with sedation experience 125 (87 %)
Dissatisfied with sedation experience 5 (3 %)
Unable to contact 13 (10 %)
difficulty in sedation. Concomitant chronic opioid use was
not significantly associated with difficulty in sedation
(p = 0.16). Although the sample size for benzodiazepine
use (n = 4) was very small, there was no suggestion of an
association with difficulty in sedation. Psychotropic drug
use was found to be associated with difficult sedation
during endoscopy (p = 0.04). Not surprisingly, psycho-
tropic drug use also was associated with increased state
anxiety (p = 0.02) and trait anxiety (p = 0.03). However,
interpretation is limited by the small number of subjects
who were taking psychotropic medications.
Although 56 of the 143 subjects reported difficult
sedation, only five subjects reported dissatisfaction with
their sedation experience during the follow-up telephone
call. All five of these subjects had been rated as difficult to
sedate on the basis of their RASS score. Twelve subjects
could not be contacted for follow-up. One hundred and
twenty-five subjects (87 %) were satisfied with their
sedation experience (Table 6).
The five-point Likert-scale rating of difficulty of seda-
tion by the members of the endoscopy team were analyzed
with the Cohen’s j coefficient. For inter-rater agreement
between attending physician and nurse the j coefficient
was 0.512. The j coefficient for the attending physician
and fellow was 0.684. On the basis of Fleiss’s guidelines
mentioned in the ‘‘Methods’’ section, this signified good
agreement regarding perception of the difficulty of sedation
by members of the endoscopy team. However, physicians
were in closer agreement with each other than with the
nursing staff.
Doses of medications administered for sedation during
endoscopy were also recorded. The mean doses of different
medications were: midazolam 4.5 mg, fentanyl 30 mcg,
meperidine 47 mcg. Notably, the dose of medication was
not taken into consideration for classifying a subject as
difficult to sedate. It was based purely on the objective
criteria of arousal during the procedure and its sustain-
ability to stimuli.
Discussion
Our study demonstrated that higher than normal levels of
both state anxiety and trait anxiety were associated with
123
Dig Dis Sci (2012) 57:2527–2534
difficult sedation. Subjects who perceived elevated anxiety
overall in their life (trait anxiety) experienced difficulty
with sedation. In addition, elevated anxiety because of an
imminent procedure (state anxiety) was also associated
with difficult sedation. These findings were consistent with
results from previous studies [4, 6, 23] and with experience
reported by the gastroenterology community. The signifi-
cant association identified between state anxiety and diffi-
cult sedation prompts discussion regarding use of different
tools to enhance patient relaxation before endoscopic
procedures.
Interestingly, use of psychotropic medications (anti-
depressants and anti-psychotics) has been associated with
increased difficulty with sedation during endoscopy [3, 4,
24]. We did not find any association of chronic opioid or
benzodiazepine use and difficulty in sedation. As men-
tioned previously, our results should be interpreted with
caution given the small number of patients on these drugs
in this study. We noted strong covariation between psy-
chotropic medication use and both state and trait anxiety.
Therefore, it is likely that the observed association between
difficult sedation and psychotropics was confounded by the
use of these medications to treat anxiety disorders. How-
ever, the intriguing association between psychotropic
medications and difficult sedation during endoscopy could
also indicate these subjects’ underlying psychiatric dis-
eases, rather than anxiety, are causing difficult sedation.
Further studies should be conducted to investigate a variety
of psychiatric disorders and their association with difficult
sedation during endoscopy.
Our study did not find evidence to support the common
belief that alcohol abuse results in difficult sedation during
endoscopy [4, 24]. However, our results must be inter-
preted with some caution, because the number of subjects
meeting criteria for alcohol abuse was relatively small.
Along similar lines, chronic opiate or benzodiazepine use
was not associated with difficult sedation during endos-
copy. This makes it clear that subjects with alcohol abuse
and opiate/benzodiazepine use are not difficult to sedate.
Such subjects may require higher doses of sedation medi-
cation but will almost universally have an appropriate
response to the sedating medication.
Anecdotally, sexual and/or physical abuse may make
subjects feel vulnerable when undergoing invasive endo-
scopic procedures [25–27], resulting in agitation and dif-
ficulty with sedation. We evaluated this hypothesis by
documenting sexual/physical abuse with a validated ques-
tionnaire, and applying standardized measures of difficult
sedation. We found no significant relationship between
sexual or physical abuse and difficult sedation during
endoscopy. We also found no significant association
between histories of abuse and state or trait anxiety in this
study. Despite these findings, invasive procedures such as
Dig Dis Sci (2012) 57:2527–2534
endoscopy and colonoscopy may cause irrational fears and
provoke stress reactions in some subjects. Hence, attempts
should be made to identify and manage these subjects with
a compassionate multi-disciplinary approach [27].
In this study, we attempted to identify factors associated
with difficult sedation during endoscopy. The greatest
strength of our study is that standardized and validated
instruments were used to study factors often quoted to be
associated with difficult sedation. Additionally, the Rich-
mond Agitation and Sedation scale was used to objectively
describe difficult sedation by use of precise, unambiguous
definitions for different levels of sedation and agitation.
Previously quoted studies [6] have used Likert scales to
document difficult sedation from the physician’s perspec-
tive and can be biased by physician perception. Our study
recorded perception of the quality of sedation by members
of the endoscopy team but this was not used for diagnosing
difficult sedation.
Another novelty of our study was that doses of sedating
drugs were not used to define a difficult to sedate endos-
copy subject. Our definition of difficult sedation was purely
based on assessment of arousal, cognition, and sustain-
ability using common responses (eye opening, eye contact,
physical movement) to stimuli, as defined by the RASS. In
our opinion, escalating doses of sedation medications are
not indicative of difficult sedation if the subject has an
appropriate response to the administered drugs i.e. decrease
in consciousness. This is in contrast with the common
belief that subjects requiring higher doses of sedating
medications are difficult to sedate [24]. This ‘‘upper limit’’
on dosing is heavily biased based on the comfort level of
the endoscopist and varies widely among different
physicians.
The study also reinforces the observation that conscious
sedation using benzodiazepines (midazolam) and opiates
(fentanyl/meperidine) has a strong amnesiac effect [27]
thereby causing the subjects to forget their endoscopic
experience. Over 90 % of the subjects who were difficult to
sedate did not remember being uncomfortable during the
procedure and reported being satisfied with their sedation
experience.
The major limitation of our study is that our subjects
were primarily an inner city population with poor access to
medical care and consequently multiple medical problems.
This may have resulted in our selecting subjects who had
undergone multiple invasive medical procedures in the
past. They may have received conscious sedation previ-
ously and were familiar with its effectiveness in keeping
them comfortable during the procedure.
In conclusion, our study addresses an important issue
of identifying a difficult-to-sedate endoscopy patient.
Designing an algorithm to identify these subjects before
their procedure maybe helpful in preventing inadequate
2533
and aborted endoscopic procedures. Our results can also
be extrapolated to other medical specialties, for exam-
ple interventional radiology, trans-esophageal echocardi-
ography, obstetrics/gynecology, etc. for which conscious
sedation is routinely used. A major benefit of identifying a
difficult-to-sedate subject is reduction of health-care costs
associated with failed procedures and wastage of scarce
medical resources.
Conflict of interest No conflict of interest.
References
1. Abraham NS, Fallone CA, Mayrand S, Huang J, Wieczorek P,
Barkun AN. Sedation versus no sedation in the performance of
diagnostic upper gastrointestinal endoscopy: a Canadian ran-
domized controlled cost-outcome study. Am J Gastroenterol.
2004;99:1692–1699.
2. Zubarik R, Ganguly E, Benway D, Ferrentino N, Moses P,
Vecchio J. Procedure-related abdominal discomfort in patients
undergoing colorectal cancer screening: a comparison of colon-
oscopy and flexible sigmoidoscopy. Am J Gastroenterol.
2002;97:3056–3061.
3. Cook PJ, Flanagan R, James IM. Diazepam tolerance: effect of
age, regular sedation, and alcohol. Br Med J (Clin Res Ed).
1984;289:351–353.
4. Pena LR, Mardini HE, Nickl NJ. Development of an instrument
to assess and predict satisfaction and poor tolerance among
patients undergoing endoscopic procedures. Dig Dis Sci.
2005;50:1860–1871.
5. Schutz SM, Lee JG, Schmitt CM, Almon M, Baillie J. Clues to
patient dissatisfaction with conscious sedation for colonoscopy.
Am J Gastroenterol. 1994;89:1476–1479.
6. Mahajan RJ, Johnson JC, Marshall JB. Predictors of patient
cooperation during gastrointestinal endoscopy. J Clin Gastroen-
terol. 1997;24:220–223.
7. Zigmond AS, Snaith RP. The hospital anxiety and depression
scale. Acta Psychiatr Scand. 1983;67:361–370.
8. Saunders JB, Aasland OG, Babor TF, de la Fuente JR, Grant M.
Development of the alcohol use disorders identification test
(AUDIT): WHO collaborative project on early detection of per-
sons with harmful alcohol consumption II. Addiction.
1993;88:791–804.
9. Allen JP, Litten RZ, Fertig JB, Babor T. A review of research on
the alcohol use disorders identification test (AUDIT). Alcohol
Clin Exp Res. 199;21:613–619.
10. Babor TF, Higgins-Biddle JC, Saunders JB, Monteiro MG. The
Alcohol Use Disorders Identification Test Guidelines for Use in
Primary Care. 2nd ed. Geneva: World Health Organization;
2001.
11. Spielberger CD. The State Trait anxiety inventory: A Compre-
hensive Bibliography. CA: Consulting Psychologists Press; 1983.
12. Briere J, Runtz M. Multivariate correlates of childhood psycho-
logical and physical maltreatment among university women.
Child Abuse Negl. 1988;12:331–341.
13. Resnick HS, Kilpatrick DG, Dansky BS, Saunders BE, Best CL.
Prevalence of civilian trauma and posttraumatic stress disorder in
a representative national sample of women. J Consult Clin Psy-
chol. 1993;61:984–991.
14. Badgley R, Allard H, McCormick N, et al. Occurrence in the
population. In: Kortet R, Vainikka A, eds. Anonymous Sexual
123
2534
Offences against Children, vol. 1. Ottawa: Canadian Government
Publishing Centre; 1984:175–193.
15. Leserman J, Drossman DA, Li Z, Toomey TC, Nachman G. The
reliability and validity of a sexual and physical abuse history
questionnaire in female patients with gastrointestinal disorders.
Behav Med. 1995;21:141–150.
16. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond agita-
tion-sedation scale: validity and reliability in adult intensive care
unit patients. Am J Respir Crit Care Med. 2002;166:1338–1344.
17. Ely EW, Truman B, Shintani A, et al. Monitoring sedation status
over time in ICU patients: reliability and validity of the Richmond
agitation-sedation scale (RASS). JAMA. 2003;289:2983–2991.
18. Rhoney DH, Murry KR. National survey of the use of sedating
drugs, neuromuscular blocking agents, and reversal agents in the
intensive care unit. J Intensive Care Med. 2003;18:139–145.
19. Samuelson KA, Larsson S, Lundberg D, Fridlund B. Intensive
care sedation of mechanically ventilated patients: a national
Swedish survey. Intensive Crit Care Nurs. 2003;19:350–362.
20. Nisbet AT, Mooney-Cotter F. Comparison of selected sedation
scales for reporting opioid-induced sedation assessment. Pain
Manag Nurs. 2009;10:154–164.
123
Dig Dis Sci (2012) 57:2527–2534
21. Almgren M, Lundmark M, Samuelson K. The Richmond agita-
tion-sedation scale: translation and reliability testing in a Swedish
intensive care unit. Acta Anaesthesiol Scand. 2010;54:729–735.
22. Fleiss JL. Statistical Methods for Rates and Proportions. 2nd ed.
New York: Wiley; 1981:38–46.
23. Froehlich F, Thorens J, Schwizer W, et al. Sedation and analgesia
for colonoscopy: patient tolerance, pain, and cardiorespiratory
parameters. Gastrointest Endosc. 1997;45:1–9.
24. Standards of Practice Committee of the American Society for
Gastrointestinal Endoscopy, Lichtenstein DR, Jagannath S, Baron
TH, et al. Sedation and anesthesia in GI endoscopy. Gastrointest
Endosc. 2008;68:815–826.
25. Davy E. The endoscopy patient with a history of sexual abuse:
strategies for compassionate care. Gastroenterol Nurs. 2006;
29:221–225.
26. Borum ML. Childhood sexual trauma as a potential factor for
noncompliance with endoscopic procedures. Gen Hosp Psychia-
try. 1998;20:381–382.
27. Cohen LB, Delegge MH, Aisenberg J, et al. AGA Institute review
of endoscopic sedation. Gastroenterology. 2007;133:675–701.
28. Drossman D, et al. Ann Intern Med. 1995; 123:782–794.