Secondary Logo

Journal Logo

Patient comfort

Deep sedation for endoscopic cholangiopancreatography with or without pre or intraprocedural opioids

A double-blind randomised controlled trial

Fassoulaki, Argyro; Iatrelli, Ioanna; Vezakis, Antonios; Polydorou, Andreas

Author Information
European Journal of Anaesthesiology: September 2015 - Volume 32 - Issue 9 - p 602-608
doi: 10.1097/EJA.0000000000000187



Deep sedation or general anaesthesia for endoscopic retrograde cholangiopancreatography (ERCP) is a prerequisite because the procedure is more complex and lasts longer than routine endoscopies such as gastroscopy or colonoscopy. If a technique of deep sedation is chosen, patients breathe spontaneously after administration of sedatives and/or opioids. Deep sedation is needed to prevent uncontrolled movements and coughing and to increase the patient's comfort. The risks of deep sedation are hypoventilation, hypoxaemia and arterial hypotension.

Propofol used for deep sedation during endoscopy has the advantage of a rapid onset of action, a short duration and easy titration; it has, therefore, improved the conditions for ERCP and the quality of recovery.1–3 Furthermore, it has antiemetic properties, which make the drug ideal for ambulant patients.4,5 Sedation with propofol compared with traditional sedatives is associated with faster recovery but with similar adverse events when used for ERCP.6,7

Opioids are often administered with propofol for sedation during ERCP with the intention of decreasing propofol requirements. The effect of preprocedural intranasal fentanyl on sedation during ERCP has never been tested. We hypothesised that pre or intraprocedural opioids would decrease propofol requirements during ERCP. The primary aim of our study was to investigate the effect of pre or intraprocedural opioids on propofol requirement during deep sedation for ERCP procedure. Secondarily, we aimed to assess the effect of pre or intraprocedural opioids on recovery, early postoperative pain and cognitive function.

Materials and methods

Ethical approval for this study (No. M-18/21–12–2010) was provided by the Institutional Review Board of the Aretaieio Hospital, University of Athens, Greece (Chairperson Prof D Botsis) on 21 December 2010. Written informed consent was obtained from all patients who participated in the study. The study was conducted in the Endoscopy Unit, Aretaieio University Hospital. The first patient entered the study on 1 February 2011 and the last patient on 5 July 2013.

This was a single-centre, placebo-controlled, double-blinded randomised trial with three parallel groups. Adult patients of American Society of Anesthesiologists’ (ASA) physical status 1 to 3 aged between 45 and 75 years of either sex scheduled for elective ERCP were eligible to enter the study. Exclusion criteria were chronic pain, sedative, hypnotic, opioid or other analgesic consumption during the last month, allergy to the drugs used in the study protocol and refusal to give written informed consent.

Patients were assigned randomly to the remifentanil (R), intranasal fentanyl (F) or placebo (P) group. All patients received propofol for sedation. In group R, patients received intranasal placebo before ERCP and a remifentanil infusion (10 μg ml−1) was administrated at a rate of 0.1 ml kg−1 h−1 during ERCP; the infusion of remifentanil was started 5 min before administration of propofol. In group F, patients received intranasal fentanyl 200 μg before ERCP and a placebo infusion during ERCP; an infusion of 0.9% NaCl was started 5 min before administration of propofol. In group P, patients received intranasal placebo before ERCP and an infusion of 0.9% NaCl was started 5 min before administration of propofol. The intranasal fentanyl or intranasal placebo was administered with the patient in the sitting position.

Randomisation was performed using a research randomiser programme, using 60 sets of numbers and three numbers per set for a total of 180 patients, keeping each number in a set to remain unique and without sorting the numbers that are generated ( The random allocation sequence was implemented by an independent endoscopist-surgeon who also prepared the remifentanil or the 0.9% NaCl infusion in 60-ml syringes and administered the nasal spray according to the order number indicating the group of assignment.

The fentanyl and placebo nasal sprays were in dark bottles similar in appearance, the placebo bottle being filled with 0.9% NaCl by the hospital pharmacy. The label from the intranasal fentanyl spray was removed and both bottles were covered with stickers and coded by the hospital pharmacy. The anaesthesiologists who also assessed and recorded the primary and secondary outcomes, the surgeon-endoscopists, the data analyst and the patient were blinded to the group assignment.

The MiniMental state test was undertaken in all participants before sedation and the ERCP procedure. When patients arrived at the Endoscopy Unit, two cannulae (20-gauge) were inserted in peripheral veins of the right hand and connected via extension tubes to two separate infusion pumps (Graseby 3400; Graseby Medical Ltd., Watford, Herts, UK) for propofol and remifentanil or placebo infusions. The patient was asked to breathe 100% oxygen for 3 to 5 min via a close-fitting face mask and the nasal treatment was administered according to group assignment. All patients were then placed in the semiprone position required for the ERCP procedure. Monitoring included peripheral oxygen saturation (SpO2), heart rate (HR), respiratory rate with impedance pneumography, ECG and noninvasive arterial blood pressure (Nichon Kohden, Bedside Monitor, Model BSM-2301K; Tokyo, Japan). SBP and DBP were measured and recorded before starting sedation and at the end of the procedure because frequent measurements of blood pressure elicit pain due to cuff inflation and interfere with the bispectral index (BIS) measurements. Additional monitoring included BIS (BIS View Monitoring System; Aspect Medical Systems Inc., Norwood, Massachusetts, USA).

All patients were sedated with a bolus dose of propofol 1 mg kg−1 followed by a propofol infusion at a rate of 5 to 9 mg kg−1 h−1 titrated to obtain BIS values around 40 to 70. During the procedure, oxygen at a flow rate 5 l min−1 was administered via a nasal catheter. If the patient became apnoeic or the SpO2 decreased below 90%, the anaesthesiologist intervened by manipulations such as jaw thrust and/or increasing the inspired oxygen concentration. If the patient moved or coughed, the rate of the propofol infusion was increased.

Baseline values (before propofol and/or opioid administration) of BIS, SpO2, HR and respiratory rate were recorded, and further recordings were made every 3 min throughout the procedure. During the ERCP, the number of patients who reacted with a cough or movement and required an increase in the rate of the propofol infusion, the number who required intervention due to decreases in SpO2 below 90% and the number of interventions in each group were recorded.

When the ERCP was finished and the endoscope was removed, the infusions were discontinued and the total dose of propofol was recorded. Thirty minutes after the end of the procedure, patients were assessed using the Ramsay Sedation Scale8 and Observer Assessment of Alertness Sedation (OAAS) score.9 Postoperative pain was assessed using a visual analogue scale (VAS; 0 mm no pain, 100 mm worst pain imaginable) at 30 min. The requirement for analgesics in the first 30 min was recorded. Cognitive function assessment using the modified MiniMental state10 test was repeated 30 min after the ERCP. For patients who had not achieved full recovery after 30 min, the sedation scores and assessment of cognitive function were repeated at 60 and 120 min. Details of the Ramsay Sedation Scale, OAAS score and MiniMental test are shown in a supplemental data file,

Patient and endoscopist satisfaction were quantified using a 0 to 100 mm VAS scale. Patients scored globally their satisfaction on the basis of the lack of awareness during the procedure, the postprocedural presence of pain and/or postoperative nausea and vomiting. The endoscopist scored satisfaction on the basis of patients’ involuntary movements and/or coughing during the procedure as well as on events requiring interruption of the procedure in order to ventilate and oxygenate the patient.

The primary outcome of the study was the total dose of propofol administered to accomplish the procedure. Secondary outcomes were quality of recovery, postoperative pain and cognitive function.


To detect a significant difference of 15% in total propofol requirements among the three groups with a power of 0.8 (or 80%) at 5% level of significance, 57 patients were needed in each group. This was based on the means and standard deviations of the first 14 patients in each group. The mean ± SD propofol requirements in these patients were 14.9 ± 4.4, 13.5 ± 4.2 and 11.9 ± 3.0 mg kg−1 in groups R, F and P, respectively.

Differences in patients’ characteristics (age, BMI), SBP, DBP, propofol requirements, MiniMental test scores, postoperative pain scores, patient satisfaction and endoscopist satisfaction were analysed using the Kruskal–Wallis test. The number of interventions for SpO2 decreases was compared using the Kruskal–Wallis test. The differences in Ramsay and OAAS scores were analysed using Fisher's exact test.

The number of patients with cancer or jaundice, the number of patients who reacted by coughing and/or movement intraoperatively or required one or more interventions due to decreases in oxygen saturation below 90% and the number of patients who required analgesics in each group were analysed. Differences between incidences were compared using the χ2 test.

For all measurements made during ERCP (BIS, HR, respiratory rate, SpO2), the mean overall observations after baseline were computed for each group and an analysis of covariance (ANCOVA) was performed to assess the overall means among the three groups after adjustment for the baseline values.

All analyses were carried out using SPSS v.11.0 (SPSS Inc., Chicago, Illinois, USA).


The number of patients excluded after group allocation and the reasons for exclusion are shown in the flow diagram of the study (Fig. 1). The data obtained from the remaining 173 patients were analysed as per protocol. The three groups did not differ regarding age, BMI, ASA physical status, duration of ERCP, incidence of cancer or incidence of jaundice (Table 1).

Fig. 1:
Flow diagram of the study.
Table 1:
Patient characteristics, duration of endoscopic retrograde cholangiopancreatography, men/women, American Society of Anesthesiologists physical status, number of patients with cancer and number of patients with jaundice in the remifentanil (R), the intranasal fentanyl (F) and the placebo (P) groups

The procedures performed were endoscopic sphincterotomy as well as common bile duct stone removal (82 patients), endoscopic sphincterotomy as well as stent insertion (70 patients), stent removal or exchange (14 patients) and diagnostic (seven patients).

In group F, patients reported significantly less pain (P = 0.007), and fewer patients required analgesics (P = 0.049), than patients in groups R and P (Table 2). Two, four and two patients in groups R, F and P, respectively, complained of nausea. Two patients in group R and one patient in group F also vomited. Patient satisfaction based on awareness during the procedure and on postprocedure pain and nausea/vomiting did not differ among the three groups. The endoscopists’ satisfaction scores were also similar among the groups (Table 2).

Table 2:
MiniMental state test before, 30 and 60 min after termination of the infusions, postoperative pain (visual analogue score mm), number of patients who required analgesia 30 min after the end of endoscopic retrograde cholangiopancreatography, and patient and endoscopist satisfaction in the remifentanil (R), fentanlyl (F) and placebo (P) groups

The total dose of propofol and the dose of propofol given by infusion were similar in the three groups (Table 3). The groups differed significantly regarding the incidence of patients who reacted, with fewer patients reacting during the ERCP in group F (Table 3). The number of patients who required intervention to restore and/or maintain SpO2 did not differ among the groups. However, the number of interventions to restore and/or to maintain SpO2 above 90% was higher in group F.

Table 3:
Total propofol dose, propofol dose administered by infusion only, number of patients who reacted, number of patients who required intervention for SpO2 decreases less than 90% and number of interventions in the remifentanil (R), fentanyl (F) and placebo (P) groups

During the procedure, there were no significant differences among groups in HR, respiratory rate, SpO2 or BIS values (Table 3). Arterial blood pressure did not differ before or after the procedure among the groups (Table 4).

Table 4:
SBP and DBP, heart rate, respiratory rate, SpO2 and bispectral index values in the remifentanil (R), intranasal fentanyl (F) and placebo (P) groups during the procedure

The maximum scores of the Ramsay and OAAS scales were obtained 30 min after completion of the procedure in all patients except two and did not differ among the groups. In one patient in group R, the maximum value of the Ramsay scale was obtained 60 min after the procedure, and in one patient in group F, the Ramsay and OAAS scales reached the maximum values 120 min after the procedure. Scores obtained from the MiniMental state test did not differ among the three groups at any time (Table 2).


This investigation found that the administration of pre or intraprocedural opioids had no effect on propofol requirements for deep sedation in patients undergoing ERCP. Recovery rate and cognitive function were good after 30 min. The group that received intranasal fentanyl had less minor pain compared with the other two groups.

In patient-controlled sedation with propofol and remifentanil for ERCP, less propofol was required when compared with propofol administered as a continuous infusion and fentanyl titrated by the anaesthesiologist.11 Our patients were deeply sedated because the endoscopist required an immobile patient. Although patients were asleep, the sedation required to prevent unintentional movements or coughing was too deep to allow patient-controlled sedation.

Berzin et al.12 showed that ASA physical status is a confounding factor for sedation-related cardiovascular adverse events, while BMI is a confounding factor for respiratory adverse events. In the same study, severe sedation-related adverse events were SpO2 less than 85%, unplanned tracheal intubation, arrhythmias and termination of the procedure.

We did not observe any intraprocedure arrhythmias that required intervention. However, no patients with ASA physical status more than 3 were included in our study. In addition, the gas insufflation was air rather than CO2, which predisposes to arrhythmias. Hypoxic episodes were successfully managed by interventions treating the airway obstruction and increasing the inspired oxygen concentration. None of the procedures was terminated prematurely or postponed due to hypoxaemia or cardiovascular complications associated with the sedation administered. Our patients were preoxygenated with 100% oxygen for about 3 to 5 min before being placed in the semiprone position, and fewer decreases in SpO2 during the procedure were expected. For these reasons, our results regarding adverse effects of sedation are not comparable with those of the previous study.

End-tidal CO2 monitoring has been shown to decrease the incidence of hypoxaemia during colonoscopy.13 During the ERCP procedures, we did not obtain reliable end-tidal CO2 values because the tube collecting the expired air was frequently misplaced by the endoscopist. Visual monitoring of breathing and respiratory rate obtained by impedance pneumography proved adequate for intraoperative respiratory monitoring.

Gas insufflation during ERCP was with air, and bowel distension and postoperative pain might be increased. However, all groups were exposed to the same type of gas insufflation.

The doses of remifentanil were subanalgesic. Preprotocol tests with higher doses of remifentanil were associated with apnoeic and hypoxaemic episodes requiring frequent interruptions of the intervention. The low dose of remifentanil may explain the absence of an effect on propofol requirement in our investigation. In another study, higher doses of remifentanil added to propofol sedation for colonoscopy resulted in higher incidences of respiratory and cardiovascular depression and slower recovery from sedation.14

In the present study, intranasal fentanyl improved the quality of sedation in terms of fewer intraprocedural unintentional movements and less postoperative pain, although the number of patients who required intervention for low SpO2 was no higher than in the other groups. Although all interventions were administered after the insertion of the cannulae in peripheral veins, intranasal fentanyl can be given before insertion of the cannulae, alleviating patients’ anxiety and pain during venepuncture. This may be advantageous, particularly in ambulatory patients in whom scheduled premedication is not feasible.

The majority of patients achieved the preprocedure MiniMental state score within the first 30 min after ERCP. This time period is consistent with the time required for patient recovery in the endoscopic unit as assessed by the Aldrete score in a previous study.15 The same investigators reported that the total dose of propofol administered did not affect the time to discharge.

The main strength of this study was the design; it was a double-blinded randomised trial with three groups. However, the study has a number of limitations. First, two different opioids were tested, administered at different times. The opioid administration during ERCP was subanalgesic. Therefore, the two groups with opioids did not have equipotent analgesia. Second, there is a controversy regarding whether BIS can estimate the degree of sedation accurately. Targeting deep sedation with BIS alone may be an oversimplification. Third, the MiniMental state test is a screening test for cognitive dysfunction; therefore, the possibility that more accurate instruments could identify differences among the groups cannot be excluded. Furthermore, a learning effect between a pre and postprocedural assessment cannot be excluded. Fourth, blood pressure measurement was not undertaken during ERCP; therefore, adverse reporting may be incomplete. Finally, this was a single-centre study and the results of the study cannot be generalised.

In conclusion, the addition of opioids before or during ERCP had no clinically relevant benefits in this investigation. In all groups, the recovery was fast and without cognitive dysfunction. Intranasal fentanyl decreased postoperative pain intensity.

Acknowledgements relating to this article

Assistance with the study: we thank Ms Katerina Dimitriou BSc for help in the statistical analysis of the data, and Mr Nikos Karakostas for nursing assistance during the ERCP procedures.

Financial support and sponsorship: none.

Conflicts of interest: none.

Presentation: none.


1. Fanti L, Testoni PA. Sedation and analgesia in gastrointestinal endoscopy: what's new? World J Gastroenterol 2010; 16:2451–2457.
2. Lee CK, Lee S-H, Chung IK, et al. Balanced propofol sedation for therapeutic GI endoscopic procedures: a prospective randomized study. Gastrointest Endosc 2011; 73:206–214.
3. Angsuwatcharakon P, Rerknimitr R, Ridtitid W. Cocktail sedation containing propofol versus conventional sedation for ERCP: a prospective, randomized controlled study. BMC Anesthesiol 2012; 12:20.
4. Borgeat A, Wilder-Smith OHG, Saiah M, Rifat K. Subhypnotic doses of propofol possess direct antiemetic properties. Anesth Analg 1992; 74:539–541.
5. Ewalenko P, Janny S, Dejonckheere M, et al. Antiemetic effect of sybhypnotic doses of propofol after thyroidectomy. Br J Anaesth 1996; 77:463–467.
6. Bo L-L, Bai Y, Bian J-J, et al. Propofol vs traditional sedative agents for endoscopic retrograde cholangiopancreatography: a meta-analysis. World J Gastroenterol 2011; 17:3538–3543.
7. Garewal D, Powell S, Milan SJ, et al. Sedative techniques for endoscopic retrograde cholangiopancreatography. Cochrane Database Syst Rev 2012; CD007274.
8. Ramsay MA, Savege TM, Simpson BR, et al. Controlled sedation with alphaxalone-alphadolone. Br Med J 1974; 2:656–659.
9. Chernik DA, Gillings D, Laine H, et al. Validity and reliability of the Observer's Assessment of Alertness/Sedation Scale: study with intravenous midazolam. J Clin Psychopharmacol 1990; 10:244–251.
10. Crum RM, Anthony JC, Bassett SS, et al. Population-based norms for the Mini-Mental State Examination by age and educational level. JAMA 1993; 269:2386–2391.
11. Mazanikov M, Udd M, Kylänpää L, et al. Patient-controlled sedation with propofol and remifentanil for ERCP: a randomized, controlled study. Gastrointest Endosc 2011; 73:260–266.
12. Berzin TM, Sanaka S, Barnett SR, et al. A prospective assessment of sedation-related adverse events and patient and endoscopist satisfaction in ERCP with anesthesiologist-administered sedation. Gastrointesr Endosc 2011; 73:710–717.
13. Friedrich-Rust M, Welte L, Welte C, et al. Capnographic monitoring of propofol-based sedation during colonoscopy. Endoscopy 2014; 46:236–244.
14. Moerman AT, Struys M, Vereecke HE, et al. Remifentanil used to supplement propofol does not improve quality of sedation during spontaneous respiration. J Clin Anesth 2004; 16:237–243.
15. Fanti L, Agostini M, Casati A, et al. Target-controlled propofol infusion during monitored anesthesia in patients undergoing ERCP. Gastrointest Endosc 2004; 60:361–366.

Supplemental Digital Content

© 2015 European Society of Anaesthesiology