Colonoscopy is one of the most commonly performed outpatient procedures for the diagnosis, serial surveillance, and treatment of gastrointestinal disorders. The pain and anxiety associated with this procedure necessitates the use of sedation-analgesia to achieve patient comfort and cooperation (1). Older patients undergoing colonoscopy are at the highest risk of hemodynamic instability, respiratory depression, and delays in the time to home readiness associated with the use of narcotic analgesia (2).
Remifentanil is an ultra short-acting μ-opiate receptor agonist that is capable of permitting a rapid transition from intense analgesia to minimal residual effect (3,4). This study was designed to test the hypothesis that a remifentanil infusion would be more effective than meperidine boluses for analgesia in older patients undergoing ambulatory colonoscopy.
After IRB approval and written informed consent, 100 patients scheduled for ambulatory colonoscopy were randomized to either the Remifentanil or Meperidine group. Eligible patients were between the ages of 18 and 90 yr of age and had an ASA physical status of I-IV. Patients were excluded from the study if they had a history of chronic narcotic and/or benzodiazepine use, drug abuse, previous colon resection, or allergy to meperidine or fentanyl.
A modified Ramsay scale (6-point sedation scale) was used to assess the baseline sedation level and to titrate midazolam throughout the procedure (5). Midazolam was titrated using 0.5 mg increments until a sedation score of 2 (cooperative, oriented, tranquil) was achieved. If the patient’s baseline sedation score was 2, midazolam was titrated to a sedation score of 3 (responds to verbal commands). Subsequent doses of midazolam could be administered after the colonoscopy started if: 1) sedation score decreased to less than 2 or 3; 2) patient’s verbal rating of their anxiety level was in the moderate-to-severe range (5–10); or 3) at the request of the blinded operator.
After initial sedation but before the start of the procedure, patients were administered an initial dose of the study drug. The Meperidine group received a 1 mg/kg (patients ≤65 yrs) or 0.5 mg/kg (patients >65 yrs) bolus of meperidine with a placebo infusion. The Remifentanil group received a 5-min infusion of either 0.05 μg kg−1 min−1 (≤ 65 yrs) or 0.025 μg kg−1 min−1 (> 65 yrs) before beginning the procedure with a placebo bolus. On completion of the 5-min initial dose, the procedure started and the maintenance infusion continued at 50% of the initial infusion. No bolus doses of remifentanil were administered.
If patients spontaneously complained of pain or if their verbal analog score increased, the infusion rate was increased and bolus aliquots were simultaneously administered. The infusion of remifentanil was increased by either 0.05 μg kg−1 min−1 for patients ≤ 65 yr of age or 0.025 μg kg−1 min−1 for patients > 65 yr old. Additional doses of meperidine equaled 0.5 mg/kg for patients ≤ 65 yrs and 0.25 mg/kg >65 yr. Increases in infusion (placebo or remifentanil) rate were always accompanied by an appropriate additional simultaneous injection of either meperidine or placebo respectively. The infusion could be decreased by 50% if patients experienced any of the following: respiratory depression (respiratory rate < 8 breaths per minute or Sao2 < 94% or end-expiratory carbon dioxide > 50 mm Hg), hypotension (systolic blood pressure 20% less than baseline or < 80 mm Hg for >1 min), or bradycardia (decreases in heart rate [HR] by > 30% or HR < 50 bpm for > 1 min).
Standard anesthesia monitors were used to monitor HR, blood pressure, Sao2, electrocardiogram, and end-expiratory CO2 (at a nasal cannula without oxygen flow). The patient was continuously monitored with hemodynamic and respiratory measurements recorded every 5 min throughout the procedure. Oxygen at 5 L/M by simple face mask was administered throughout the procedure. A baseline minute volume was obtained by using an OHMEDA TVX Flow Transducer/Volume monitor sensor unit™ attached to a tight-fitting anesthesia face mask connected to the Ohmeda RGM 5250™.
Patient tolerance was assessed by recording the number of episodes of hemodynamic and respiratory disturbances throughout the study. Hemodynamic disturbances included the incidence of increased blood pressure (systolic blood pressure > 20% of baseline), tachycardia (HR > 110 bpm or > 20% of baseline), hypotension (systolic blood pressure ≤ 80 mm Hg or > 20% decrease from baseline), bradycardia (HR < 50 bpm or > 30% decrease from baseline). Respiratory disturbances included respiratory rate (< 8 breaths per minute) and reductions in arterial saturation (Sao2 < 92%). The incidence of nausea, vomiting, and pruritus were also recorded.
Patients were reassessed for pain and anxiety at the following anatomical points: 1) colonoscope insertion; 2) sigmoid; 3) cecum; 4) retroflexed view of the rectum; and 5) withdrawal of scope. A blinded investigator asked each patient to assess his/her level of pain and anxiety using an 11-point (0–10) verbal pain and anxiety rating scale (6,7). The verbal pain and anxiety scales were anchored by word pairs (e.g., no pain to severe pain, no anxiety to most severe anxiety ever).
The infusion was discontinued at the conclusion of the colonoscopy and no further injections were administered. Minute ventilation was reassessed at the conclusion of the procedure and repeated again 5 and 10 min later. A modified Aldrete score (8,9) of 8 defined completion of Phase I recovery and a Postanesthetic Discharge Scoring (10) of 9 defined time to home readiness (8–10).
Patient and operator satisfaction were surveyed using a four-point scale (1 = poor, 2 = satisfactory, 3 = good, 4 = excellent). Patients were surveyed 30 min after the end of the procedure and on postoperative day 1 (telephone interview) to assess their satisfaction with the management of their sedation-analgesia. Patient satisfaction was assessed by asking about overall satisfaction, residual drug effects, number of hours it took before they returned to normalcy, and whether or not they would prefer the same sedation-analgesia medication should they need to undergo a similar procedure in the future. Operator satisfaction was assessed by rating patient tolerance to the procedure, ease of examination, level of analgesia, and overall impression.
Descriptive statistics were expressed as proportions (frequency count) for categorical-type variables and means ± sd for interval (ratio)-type variables. All data were analyzed by using SAS software ( version 6.12™; SAS, Cary, NC). Repeated-measures analysis of variance was performed on outcome measurements across time (physiologic variables). The results were generally consistent between groups over time and therefore were not reported. Data were analyzed by using either t-test (age, weight, procedural and recovery characteristics), Fisher’s exact test (patient demographics), or Mann-Whitney U-test (episodes of side effects). The frequency data were analyzed by Fisher’s exact test. A P value of ≤ 0.05 was considered significant unless otherwise indicated.
One hundred patients (97 male, 3 female) were randomized for this study. Patient demographics are illustrated in Table 1. The most common indication for colonoscopy was follow-up for colonic polyps (65%). The overall procedural characteristics for patients receiving monitored anesthesia care in both treatment groups are demonstrated in Table 2. Doses of meperidine used are consistent with those reported in other studies in patients undergoing colonoscopy (11–14).
The cardiopulmonary response to each analgesic regimen is shown in Figure 1. Although there was no significant difference in group variability, significant changes over the course of procedure were found for mean arterial pressure, respiratory rate, and end-tidal carbon dioxide by using two-way analysis of variance (P = 0.02). The number of episodes of adverse hemodynamic events are depicted in Table 3. The incidence of respiratory rates <8 breaths/min was similar between groups. The incidence of nausea (3 vs 8;P = 0.03), vomiting (0 vs 5;P = 0.06), and pruritus (1 vs 3;P = 0.75) also tended to be decreased in the remifentanil group compared with those receiving meperidine.
Figure 2 illustrates the verbal pain and anxiety scores for each group before, during, and after the procedure. Compared with the Meperidine group, the baseline anxiety (P = 0.02) and pain scores (P = 0.06) were increased in the Remifentanil group. Even after normalizing the values in each group, the adjusted verbal pain scores for the Remifentanil group were higher than those for the Meperidine group at the sigmoid flexure (P = 0.03) and cecum (P = 0.05). The adjusted verbal anxiety scores were also increased in the Remifentanil group after the initial dose (P = 0.03), sigmoid flexure (P = 0.05), and cecum (P = 0.03). There was no significant difference between the percentages of patients or operators that rated the conditions as excellent or good versus satisfactory or poor.
This study demonstrates that remifentanil infusions and meperidine boluses are equally well tolerated in older patients undergoing ambulatory colonoscopy when administered by an anesthesia provider. Although the patient and operator surveys revealed overall satisfaction in both groups, patients in the Meperidine group had more side effects (i.e., hypotension, tachycardia, nausea) and those receiving remifentanil had higher adjusted verbal pain and anxiety scores during parts of the procedure. The hemodynamic and analgesic differences noted between the two treatment groups were most likely related to the doses of remifentanil and meperidine used and the distinctive pharmacokinetic properties of each drug.
There were no significant differences in the variability of several physiologic variables between the two treatment groups. The differences in HR between the two treatment groups are most likely related to the known chronotropic effects of meperidine and remifentanil on the autonomic nervous system. The increased incidence of hypotension and tachycardia in the Meperidine group suggests that bolus dosing of this drug may not be the most hemodynamically stable method of administering sedation-analgesia in patients with significant cardiovascular disease. Finally, the increased incidence of nausea and vomiting in the Meperidine group is the most likely explanation for the 8-minute delay in time to home readiness compared with the Remifentanil group.
The significant increase in the baseline verbal anxiety scores in the Remifentanil group was surprising. The trend toward a higher level of education and cardiovascular disease in the Remifentanil group could have contributed to this unexpected finding. The increased level of anxiety may have had an influence on the increased verbal pain scores noted in the Remifentanil group during portions of the procedure (15–17). The wide standard deviation in doses of remifentanil used and higher frequency of dosage increases in the Remifentanil group suggests that the higher verbal pain scores may also have been because of insufficient analgesia during portions of the procedure. Overall, the gastroenterologist could not distinguish between treatment groups and noted that in general the quality of sedation-analgesia was much better in patients participating in the study.
The ability to achieve equipotent doses of remifentanil and meperidine in this study was complicated by marked pharmacokinetic differences between remifentanil and meperidine. Also, although the dosing of meperidine in patients undergoing colonoscopy is well described, the ability to convert the doses of meperidine into equivalent doses of remifentanil is almost impossible given its ultra-short context-sensitive half-life (3,4,18). The initial remifentanil infusion rates were chosen largely on the basis of previous studies that used remifentanil in conjunction with midazolam for monitored anesthesia care (19–22). Our dosing protocol allowed for rapid adjustment of the remifentanil infusions based on patient response. Although our average infusion rate was 0.14 μg kg−1 min−1, the verbal pain and anxiety scores suggested that the remifentanil and/or midazolam doses used may have been insufficient in some patients.
Sedation-analgesia for colonoscopy is most commonly administered under the supervision of a gastroenterologist using bolus doses of meperidine (23). Our decision to use of a double-blinded design to compare the tolerance of using meperidine and remifentanil for colonoscopy in a nonoperating room environment necessitated the use of anesthesia providers. Indeed, the current package insert for remifentanil recommends that the administration should be limited to anesthesia providers. Unfortunately, the economic implications of using an anesthesia provider to administer remifentanil for routine colonoscopies severely limit its broadest application. The use of remifentanil in older patients may be made more economically viable in patients requiring monitored anesthesia care for significant coexisting medical disease. Alternatively, the establishment of a safe and effective delivery system for fast-acting drugs such as remifentanil could allow patients undergoing colonoscopy to adjust their dose on an “as needed” basis (24).
In summary, remifentanil infusions and meperidine boluses were equally well tolerated in older patients undergoing ambulatory colonoscopy when administered by an anesthesia provider. Additional studies will be required to identify optimal dosing as well as a safe and effective drug delivery system if remifentanil is going to achieve wider use in patients undergoing colonoscopy.
The authors would like to thank members of anesthesia care teams and gastroenterology and day surgery units who were instrumental in conducting this multidisciplinary study. In addition, we would like to thank Wanda Jones for her administrative support in the preparation of this manuscript.
1. Schutz S, Lee J, Schmitt C,et al. Clues to patient dissatisfaction with conscious sedation for colonoscopy. Am J Gastroenterol 1994; 89: 1476–9.
2. Wayne J, Kahn O, Aurebach M. Complications of colonoscopy and flexible sigmoidoscopy. Gastrointest Endosc Clin N Am 1996; 6: 343–77.
3. Patel S, Spencer C. Remifentanil. Drugs 1996; 52: 417–21.
4. Kapila A, Glass P, Jacobs J,et al. Measured context-sensitive half-times of remifentanil and alfentanil. Anesthesiology 1995; 83: 968–75.
5. Ramsey M, Savege T, Simpson B, Goodwin R. Controlled sedation with alphaxalone-alphadolone. BMJ 1974; 2: 656–9.
6. Huskisson E. Measurement of pain. Lancet 1974; 2: 1127–31.
7. Joyce C, Zutshi D, Hrubes V, Mason R. Comparison of fixed interval and visual analogue scales for rating chronic pain. Eur J Clin Pharm 1975; 8: 415–21.
8. Aldrete A. The post-anesthesia recovery score revisited [letter]. J Clin Anesth 1995; 7: 89–91.
9. Chung F, Chan V, Ong D. A post-anesthetic discharge scoring system for home readiness after ambulatory surgery. J Clin Monit 1995; 7: 500–6.
10. Chung F. Recovery pattern and home-readiness after ambulatory surgery. Anesth Analg 1995; 80: 896–902.
11. Chokhavatia S, Nguyen L, Williams R,et al. Sedation and analgesia for gastrointestinal endoscopy. Am J Gastroenterol 1993; 88: 393–6.
12. DiPalma J, Herrera J, Weiss R,et al. Alfentanil for conscious sedation during colonoscopy. South Med J 1995; 88: 630–4.
13. Rodney W, Dabov G, Orientale E. Sedation associated with a more complete colonoscopy. J Fam Pract 1993; 89: 394–400.
14. Seow-Choen F, Leong A, Tsang C. Selective sedation for colonoscopy. Gastrointest Endosc 1994; 40: 661–4.
15. Maranets I, Kain Z. Perioperative anxiety and intraoperative anesthetic requirements. Anesth Analg 1999; 89: 1347.
16. Wallace LM. Trait anxiety as a predictor of adjustment to and recovery from surgery. Br J Clin Psychol 1987; 26: 73–4.
17. Johnston M. Pre-operative emotional states and post-operative recovery. Adv Psychosom Med 1986; 15: 1–22.
18. Glass PS, Gan TJ, Howell S. A review of the pharmacokinetics and pharmacodynamics of remifentanil. Anesth Analg 1999; 89: S7–14.
19. Servin F, Desmont JM, Watkin WD. Remifentanil as an analgesic adjunct in local/regional anesthesia and in monitored anesthesia care. Anesth Analg 1999; 89(4S): 528–32.
20. Avramov MN, Smith I, White PF. Interactions between midazolam and remifentanil during monitored anesthesia care. Anesthesiology 1996; 85: 1283–9.
21. Gold MI, Watkins WD, Sung YF,et al. Remifentanil versus remifentanil/midazolam for ambulatory surgery during monitored anesthesia care. Anesthesiology 1997; 87: 51–7.
22. Bowdle TA, Camporesi EM, Maysick L,et al. A multicenter evaluation of remifentanil for early postoperative analgesia. Anesth Analg 1996; 83: 1292–7.
23. Keeffe E, O’Oconnor K. A/S/G/E survey of endoscopic sedation and monitoring practices. Gastrointest Endosc 1990;1989:36:S13–18.
© 2001 International Anesthesia Research Society
24. Roseveare C, Seavell C, Patel P,et al. Patient-controlled sedation with propofol and alfentanil during colonoscopy: A pilot study. Endoscopy 1998; 30: 482–3.