Y-ECCO Literature Review: Benjamin Zare

Benjamin Zare

Patient satisfaction of propofol versus midazolam and fentanyl sedation during colonoscopy in inflammatory bowel disease

Steenholdt C, Jensen JT, Brynskov J, et al.

Clin Gastroenterol Hepatol. 2022;20:559–68.e5

Benjamin Zare
© Benjamin Zare


Many clinicians have anecdotally observed patients opting out of colonoscopies due to unpleasantness related to the procedure, and within Inflammatory Bowel Disease (IBD) populations a number of factors have been implicated in non-adherence, including logistics, health perceptions, stress and procedure problems (including discomfort) [1].

Prior studies have demonstrated patient preferences for propofol sedation over midazolam and fentanyl sedation for outpatient colonoscopy in general [2]. Furthermore, propofol has been shown to be safe, without severe adverse events or accidents [3], and nurse-administered propofol has specifically proven to be an efficient means of sedation for endoscopy in low-risk patients [4]. Nevertheless, this area has yet to be explored in the specific cohort of IBD patients.

Modern management of IBD requires the employment of ileocolonoscopy for diagnosis, as well as for the surveillance and guidance of future management. The investigators here looked to fill the aforementioned knowledge gap through design of a trial investigating the effectiveness of deep nurse-administered propofol sedation (NAPS), versus moderate midazolam and fentanyl sedation, as a means of improving patient satisfaction and future attitude towards colonoscopies.


This study is a prospective, randomised unblinded clinical trial comparing use of NAPS to moderate sedation. Patients included in the study were ≥18 years old and either had an existing diagnosis of IBD, with surveillance colonoscopy due, or were suspected of having IBD; thus, in the latter group colonoscopy represented part of the diagnostic work-up.

After generation by a third party of a random sequence list assigning each participant to either form of sedation, participant numbers were packed in envelopes (concealed from investigation team members and patients) with allocation revealed in the preparation area prior to entering the endoscopy suite. A total of 320 patients were initially screened for eligibility, with most excluded due to insistence on deep sedation. Based on superiority of deep sedation over moderate sedation, with a minimum difference of 7 in total satisfaction score, equal randomisation and power of 90% with an alpha of 5%, a targeted 130 participants were included. Of these, 63 were originally randomised to deep sedation and 67 to moderate sedation. After one drop out from the propofol group due to ASA class change leading to last-minute exclusion, and two unavoidable drop outs from the midazolam and fentanyl group, data were collected for 62 and 64 participants in each group, respectively.

Deep sedation with propofol was provided by dedicated nurses trained in NAPS: doses comprised an initial bolus dose in milligrams of 100 – age, followed by 10–20 mg every 1–2 minutes. In the moderate sedation group, midazolam was administered as a starting bolus of 1–5 mg with 0.5–2 mg supplementary doses, while fentanyl was given as an initial 25 μg in those ≤60 years old or 50 μg in those ≥61 years old, followed by top-up doses at the discretion of the endoscopists.

All participants were asked to fill out baseline questionnaires prior to their procedure and post-procedure Satisfaction Questionnaires (comprising 13 questions with each rated on a five-point Likert scale, giving a possible 13–65 points) at the time of discharge from the unit. A week after their procedure, an otherwise identical but electronic Satisfaction Questionnaire was forwarded to patients for completion on a voluntary basis, with a view to assessing for any change in participants’ original answers. Higher satisfaction scores were deemed to represent more positive outcomes.

The primary end point was defined as total satisfaction at time of discharge from the endoscopy unit. Secondary end points included: satisfaction scores at 1 week post-procedure, individual factors impacting on total satisfaction and whether participants’ opinion regarding future colonoscopy had altered. Adverse events were also considered amongst secondary outcomes.

Key Findings

On review of results, regarding the primary outcome, patients who were administered deep sedation with propofol had significantly higher satisfaction scores (mean 60.1, SD 3.4) than those administered moderate sedation with midazolam and fentanyl (mean 51.2, SD 8.4; p<0.001). Reassessment questionnaires at 1 week after discharge (completed by 60% of the deep and 48% of the moderate sedation group, respectively) revealed no major differences in post-procedure satisfaction, demonstrating these differences in degree of satisfaction to be enduring and to remain significant.

Multivariate linear regression analysis demonstrated that, aside from experience of last endoscopy as identified by the pre-procedure questionnaire, there were three main items that contributed significantly to the difference in overall satisfaction score: increased sedation/drowsiness (difference of 27% in deep versus moderate sedation groups), decreased algesia (45% difference between the groups) and improved general experience as compared to previous procedures (18% difference between the groups). There was also a 158% difference in degree of amnesia in those administered propofol as compared to those administered midazolam and fentanyl. Perhaps most crucially, on the question of whether patients would accept more frequent colonoscopies on the basis that they would be offered the same sedation, scores were 22% higher in the NAPS group.

Finally, whilst this study was not designed to compare safety-related outcomes between the two groups, 14 total episodes of safety concern were noted: six desaturations <92%, all within the moderate sedation group (p=0.03), one severe desaturation <85% (requiring assisted ventilation with face mask) and seven hypotensive episodes (mean arterial pressure drop >30%), including three in the moderate sedation and four in the deep sedation group (p=0.71).


The research team conducting this study should be commended on an excellently designed and well-powered study that answers a clinical question that is even more relevant as the treatment target of histological remission comes to the fore in IBD management.

Amongst the study patients, most of whom had prior experience of endoscopic procedures, deep sedation with propofol, as compared to moderate sedation with midazolam and fentanyl, led to increased satisfaction, provided a superior experience to previous procedures and increased willingness to undergo further surveillance ileocolonoscopy if the same sedation were to be offered. This was despite an increase in amnesia, with an otherwise comparable safety profile.

Whilst questions remain regarding a likely increase in costs in association with provision of NAPS (formal cost-benefit analyses in comparison with moderate sedation have yet to be conducted), increased availability of propofol sedation could catalyse improved adherence to endoscopy-based monitoring programmes underway globally across the IBD field.


    1. Friedman S, Cheifetz AS, Farraye FA, et al. Factors that affect adherence to surveillance colonoscopy in patients with inflammatory bowel disease. Inflamm Bowel Dis. 2013;19:534–9.
    2. Padmanabhan A, Frangopoulos C, Shaffer LET. Patient satisfaction with propofol for outpatient colonoscopy: A prospective, randomized, double-blind study. Dis Colon Rectum. 2017;60:1102–8.
    3. Kanno Y, Ohira T, Harada Y, et al. Safety and recipient satisfaction of propofol sedation in outpatient endoscopy: A 24-hour prospective investigation using a questionnaire survey. Clin Endosc. 2021;54:340–7.
    4. Lin OS, La Selva D, Kozarek RA, et al. Nurse-administered propofol continuous infusion sedation: A new paradigm for gastrointestinal procedural sedation. Am J Gastroenterol. 2021;116:710–6.

Benjamin Zare - Short Biography

Benjamin Zare is a gastroenterology trainee in London, UK. His research interests include luminal gastroenterology, specifically Inflammatory Bowel Disease and use of newer biologic and small molecule therapies. He is Communication Lead for GLINT (Gastro London Investigative Network for Trainees) and is currently working as local Associate PI for the RECEDE (REducing Colonoscopies in patients without significant bowEl DiseasE) study. He hopes to continue his clinical research in an upcoming post at Guy's and St Thomas' NHS Foundation Trust. 

Posted in ECCO News, Y-ECCO Literature Reviews, Committee News, Y-ECCO, Volume 17, Issue 3